Standards and CSS in IE

I’m very happy that we’ve shipped IE 7 beta 1. I wanted to make it clear
that we know Beta 1 makes little progress for web developers in improving
our standards support, particularly in our CSS implementation. I feel badly
about this, but we have been focused on how to get the most done overall for
IE7, so due to our lead time for locking down beta releases and ramping up
our team, we could not get a whole lot done in the platform in beta 1.
However, I know this will be better in Beta 2 – and I want to share how we
are placing our priorities in IE.

In the web platform team that I lead, our top priority is (and will
likely always be) security – not just mechanical “fix buffer overruns” type
stuff, but innovative stuff like the anti-phishing work and low-rights IE.
For IE7 in particular, our next major priority is removing the biggest
causes of difficulty for web developers. To that end, we’ve dug through a
lot of sites detailing IE bugs that cause pain for web developers, like
PositionIsEverything and Quirksmode, and categorized and investigated those
issues; we’ve taken feedback from you directly (yes, we do read the
responses to our blog posts) on what bugs affect you the most and what
features you’d most like to see, and we’ve planned out what we can and can’t
do in IE7.

In IE7, we will fix as many of the worst bugs that web developers hit as
we can, and we will add the critical most-requested features from the
standards as well. Though you won’t see (most of) these until Beta 2, we
have already fixed the following bugs from
PositionIsEverything and

  • Peekaboo bug
  • Guillotine bug
  • Duplicate Character bug
  • Border Chaos
  • No Scroll bug
  • 3 Pixel Text Jog
  • Magic Creeping Text bug
  • Bottom Margin bug on Hover
  • Losing the ability to highlight text under the top border
  • IE/Win Line-height bug
  • Double Float Margin Bug
  • Quirky Percentages in IE
  • Duplicate indent
  • Moving viewport scrollbar outside HTML borders
  • 1 px border style
  • Disappearing List-background
  • Fix width:auto

In addition we’ve added support for the following

  • HTML 4.01 ABBR tag
  • Improved (though not yet perfect) <object> fallback
  • CSS 2.1 Selector support (child, adjacent, attribute, first-child
  • CSS 2.1 Fixed positioning
  • Alpha channel in PNG images
  • Fix :hover on all elements
  • Background-attachment: fixed on all elements not just body

I want to be clear that our intent is to build a platform that fully
complies with the appropriate web standards, in particular CSS 2 ( 2.1, once
it’s been Recommended). I think we will make a lot of progress against that
in IE7 through our goal of removing the worst painful bugs that make our
platform difficult to use for web developers.

In that vein, I’ve seen a lot of comments asking if we will pass the
Acid2 browser test
published by the Web Standards
when IE7 ships. I’ll go ahead and relieve the suspense by
saying we will not pass this test when IE7 ships. The
original Acid Test tested only the CSS 1 box model, and actually became
part of the W3C CSS1 Test Suite since it was a fairly narrow test – but the
Acid 2 Test covers a wide set of functionality and standards, not just from
CSS2.1 and HTML 4.01, selected by the authors as a “wish list” of features
they’d like to have. It’s pointedly not a compliance test (from the Test
Guide: “Acid2 does not guarantee conformance with any specification”). As a
wish list, it is really important and useful to my team, but it isn’t even
intended, in my understanding, as our priority list for IE7.

We fully recognize that IE is behind the game today in CSS support.
We’ve dug through the Acid 2 Test and analyzed IE’s problems with the test
in some great detail, and we’ve made sure the bugs and features are on our
list - however, there are some fairly large and difficult features to
implement, and they will not all sort to the top of the stack in IE7. I
believe we are doing a much better service to web developers out there in
IE7 by fixing our known bang-your-head-on-the-desk bugs and usability
problems first, and prioritizing the most commonly-requested features based
on all the feedback we've had.

I do want to be clear that I believe the Web Standards Project and my
team has a common goal of making the lives of web developers better by
improving standards support, and I’m excited that we’re
working together to that end.

- Chris Wilson

Comments (705)
  1. Anonymous says:

    I hope that you’ll retain font embedding even if you’re shooting for CSS 2.1 compliance.

  2. Anonymous says:

    Sounds great!

  3. Keith Farmer says:

    Out of curiosity (since I’m not familiar with the bug names you list), is something going to be done about the layering of DHTML popups over combo controls and the like. I’m sure you’ve seen it — the controls in question live permanently in a layer above the pop-up, making for poor co-existance between DHTML menus and forms.

  4. Anonymous says:

    That’s *fantastic* Chris! This is exactly the sort of post I’ve been hoping for ever since the IE Blog first started. Thank you!

    The selection of further support you have chosen seems reasonable enough, you’ve certainly covered the extremely annoying and incomprehensible bugs that tend to completely screw up a website.

    The major style thing that is currently lacking that is extremely useful and widely implemented is CSS tables. Would it be possible to implement those too? The major scripting thing would be the DOM event model. One extremely simple thing to add (i.e. a five minute job) would be the new media types for ECMAScript.

    > Background-attachment: fixed on all elements not just body

    This completes your support for CSS 1.0, doesn’t it?

    > We’ve dug through the Acid 2 Test and analyzed IE’s problems with the test in some great detail, and we’ve made sure the bugs and features are on our list

    Would it be possible to make this list public?

    Please keep posting like this. It doesn’t matter if it’s merely to say "we definitely won’t be implementing [x]", so long as we know one way or the other.

  5. Anonymous says:

    Just to add to my previous comment, if you implement the new ECMAScript media types, that means we can hide scripts from Internet Explorer 6 and below without any browser checking or non-standard hacks.

  6. Anonymous says:


    Yes, background-attachement: fixed was the last thing we were aware of. If you guys know of anything else missing form CSS1 please post it here.


    — Markus

  7. Anonymous says:


    Just remembered a few things:

    Specificity is screwed up somewhere when dealing with lists. I never bothered to debug it fully, because I can usually work around it by cramming a load of specificity into the selectors I use. Are you aware of this issue?

    The forward-compatible parsing of CSS 1 requires you to ignore certain things in error conditions that you do not (or was this fixed in 6.0?).

    Stylesheets served as something other than text/css still get processed – although this is an HTTP bug, not a CSS bug per se. If you fix this, it would let us serve stylesheets to Internet Explorer 6.0 and below, but not Internet Explorer 7+.

  8. Anonymous says:

    Very encouraging.

  9. Anonymous says:

    Actually, Jim, I don’t think I’ve ever seen a specificity problem with lists. If you can find an example, we can take a look.

  10. Anonymous says:

    I like those lists. No, I *really* like those lists. I hope they get bigger before beta 2.

    "In that vein, I’ve seen a lot of comments asking if we will pass the Acid2 browser test published by the Web Standards Project when IE7 ships. [….] As a wish list, it is really important and useful to my team, but it isn’t even intended, in my understanding, as our priority list for IE7."

    I think most, if not all, web developers won’t have a problem with this. Recent Firefox nightlies don’t render it right, so I don’t honestly expect IE 7 to render it completely correct.

  11. Anonymous says:

    > Improved (though not yet perfect) <object> fallback

    One thing that the <object> element type doesn’t really address is how to signal to the user that a fallback has been used.

    In some cases, an author would want to have a "silent" fallback, for example, where a video is available in Ogg Theora and MPEG formats.

    In other cases, an author would want some kind of "you aren’t getting the best experience" notification, for example where a Flash presentation is available with a PNG screenshot fallback.

    Would it be possible to include some mechanism to allow authors to say which is more suitable? For example:

    <meta name="object-fallback" content="silent">

    In the cases where notification is suitable, you can put the notice in the yellow bar, and in the other cases, you can just silently use the fallback.

    Please do this in a way that doesn’t conflict with valid code though, traditionally, you tend to do it in such a way as to make it impossible to use it in valid pages, e.g. invalid attributes.

    Exposing to scripts whether a fallback was used would also be useful.

    This is a wishlist item though, further HTML, CSS, DOM and HTTP improvements are far more welcome.

  12. Anonymous says:

    I would like to note, that I would like to see this in IE. The autoscroll image looks terrible, do you really think that looks like a final product? Also maybe some form of a download manager, or more than 2 downloads at once implemented? Some users live in a broadband world where you can handle more than 2 downloads!

  13. Anonymous says:

    Next time I run across the list specificity problem, I’ll whip up a test case.

  14. Anonymous says:

    What about HTML q element? It should be very easy to support that.

  15. Anonymous says:

    Thank you. This is what I wanted to hear. Many of what I consider necessities are not mentioned on your list, but I’m much happier now that you’re actually making progress.

    I do want to remind you guys, though, that IE won’t be forgiven so easily. Many web developers think of Microsoft as a company who will do the bare minimum just to make their customers content enough to stay with them. So even if you do make considerable progress and bring IE near where other browsers are at, many of us will still question your long-term commitment. I would ask that you promise us that you won’t let IE’s standards support development ever fall stagnant like IE6 again, but I understand that, as a business, you aren’t in much of a position to make such long-term promises. Instead, I just want you to keep it in mind: web developers don’t trust Microsoft’s commitment anymore, and unless you work your butt off to please them, they maybe never will. Doing a minimal job will not work this time.

    That said, I really hope to get a pleasant surprize come IE7. I’m personally sticking with Firefox, but as long as the majority population is using something that reasonably complies with standards, I’ll be happy.

  16. Anonymous says:

    I’d like to know if IE7 will support the * html { … } selector still. This selector isn’t supported in other browsers, and it’s what lets us differentiate between browsers in a lot of cases. If this is changed in IE7 and not supported (or, you might say fixed, since technically I guess supporting this would be considered a bug), many existing sites that rely on this bug will cease to display properly, unless IE7 fixes all of the bugs with web page formatting (such as the box model).

    One other thing, will border-style: dotted be supported in IE7?

  17. Anonymous says:

    This post just made me smile, Chris. It really did. 🙂

    A big thank you to you and your team for the work you folks are putting in. Keep the momentum going!

  18. Anonymous says:

    Great news, and I like that you intend to keep up with your efforts with CSS 2.1!

    This article would be hilarious to read the comments to if it was posted on Slashdot.

  19. Anonymous says:

    Jim: Maybe I’m not following what you mean, but Couldn’t you do something like this:

    <!– Silent –>

    <object type="video/x-something" data="foo.v1">

    <object type="video/x-another" data="foo.v2">



    <!– Not best experience –>

    <object type="video/x-something" data="foo.v1">

    <img src="foo.png" alt="Foobar">

    <p>Your user agent cannot display this video correctly. Above is a screenshot.</p>


    Also, I sure hope they don’t add more proprietary features. That’s the last thing we want. If you want to propose something new, do it through the W3C.

  20. Anonymous says:

    "I do want to remind you guys, though, that IE won’t be forgiven so easily."

    Well, lets be honest. Most of the people making the most noise will >never< forgive MS simply because it >is< MS. Their whole personal self image is defined by how rebellious they are and how they can read /. and "stick it to the corporations!" by pirating movies and complaining about MS.

    MS could ship perfect CSS2 support and they would complain it isn’t Firefox "bug compatible", that it doesn’t retro-install into Windows 3.1. For an encore they would complain that because a fully compliant browser is in Windows that they need to fire up the whole monopoly machinery again because they can’t compete anymore on features.

    Jealous hatred never ends.

  21. Anonymous says:

    > Yes, background-attachement: fixed was the last

    > thing we were aware of. If you guys know of

    > anything else missing form CSS1 please post it

    > here.


    > Thanks

    > — Markus

    Definately. font-weight is not correct when it is set to 600.

  22. Anonymous says:

    > One thing that the <object> element type

    > doesn’t really address is how to signal to the

    > user that a fallback has been used.

    No, you can try something like this:

    <object type="image/svg+xml" data="example.svg">

    <p>You are not viewing SVG</p>

    <object type="image/png" data="example.png">

    <p>You can’t view image</p>



  23. Anonymous says:

    And also, "blink" in text-decoration is not currently supported (I guess it should be in CSS1, right).

  24. Anonymous says:

    i am a end user and if this lack in anyway of innovation then you will lose. it is a fact.

  25. Anonymous says:

    *Loud applause*

    Chris! Thank you so much for that down-to-earth and informative blog post.

  26. Anonymous says:

    This is great news; thanks for posting and for working on those CSS bugs.

    Is there any hope of IE7 supporting the application/xml+xhtml and application/xml+xslt media types?

  27. Anonymous says:

    Nanobot, minghong: Doh! I can’t believe I’ve never thought of (or seen) that before! Thanks.

    > "blink" in text-decoration is not currently supported

    Is that really a bad thing? 😉 It’s an optional part of the specs anyway.

  28. Anonymous says:

    > Is there any hope of IE7 supporting the application/xml+xhtml and application/xml+xslt media types?

    I’d guess XHTML support is very unlikely – it won’t affect as many people as HTML/CSS/DOM/HTTP updates, and it’s a bigger job than many people realise. It’s not simply a case of adding a media type to a list and failing on malformed documents. There are changes to the DOM, changes to CSS, etc that also have to be added.

    On the off-chance you guys are considering it – please don’t. Anne van Kesteren’s suggestion for an Internet Explorer that can follow the specs 100% and still render old websites with quirks is a great idea, and I hope you guys will consider it for Internet Explorer 8 – implementing a buggy application/xhtml+xml will ruin any chance of it working though.

  29. Anonymous says:

    Today, Chris Wilson spilled the beans I’ve been wanting to spill for awhile –&amp;nbsp; The majority of…

  30. Rosyna says:

    The Acid2 test is also an ERROR checking test. To see how well a browser deals with malformed HTML/CSS in strict mode. Not handling errors correctly can also cause huge problems on web pages. The Acid2 test is just some of the errors that could occur. As it is now, I think only Safari (webkit) tip of the tree passes Acid2 now. (Well, and KHTML).

  31. Anonymous says:

    >MS could ship perfect CSS2 support and they would complain it isn’t Firefox "bug compatible", that it doesn’t retro-install into Windows 3.1.

    Soulhuntre, I have no doubt that some people would hate Microsoft even if it went open source, implemented every webstandard perfectly, fixed every security flaw and stbility problem in windows, and bought them a car. Still, give them a little more credit than that.

    They wouldn’t want IE to conform to Firefox bugs. If anything, they’d just find something else to legitimately complain about. Or perhaps they’d try to fix more bugs in Firefox and then complain that IE doesn’t do that properly (nevermind that Firefox didn’t either, a week ago).

    As for Windows 3.1, they wouldn’t care, because anyone who hates Microsoft that much is running Linux or OS X.

  32. Anonymous says:

    From css1, a little thing come to mind: font-size inheritance in tables.

    Also, support for the keyword ‘inherit’.

  33. Anonymous says:

    > The Acid2 test is also an ERROR checking test. To see how well a browser deals with malformed HTML/CSS in strict mode.

    It contains CSS errors to determine if a browser follows the error handling defined by the CSS specifications. It doesn’t contain HTML errors, because there is no defined error handling for HTML, so there’s nothing to check.

    > As it is now, I think only Safari (webkit) tip of the tree passes Acid2 now. (Well, and KHTML).

    The latest iCab beta does as well.

    > Soulhuntre, I have no doubt that some people would hate Microsoft even if it went open source, implemented every webstandard perfectly, fixed every security flaw and stbility problem in windows, and bought them a car.

    Bear in mind that once upon a time, IBM were just as bad as Microsoft ever was, but IBM have pretty much redeemed themselves and are now the "good guys". They’ve done a lot less to earn their credibility than you are saying Microsoft would have to do.

  34. Anonymous says:

    This is certainly good to see. Seems you’ve addressed the most important issues.

    Main things left over from my wish list:

    – support for min-height/min-width (and correspondingly: height/width, which currently seem to act as min-height/min-width should)

    – support for :before and :after selectors, especially in combination with the "content" property (which may already be covered in the item "CSS 2.1 Selector support")

    The progress you have already made goes a long way to resolving my IE headaches. If the above points make it in, too, I’ll be a very happy camper indeed.

    Thanks for finally giving some details in addressing the concerns we’ve expressed here over the past few months.

  35. Anonymous says:

    Jim, it has an HTML comment error.

    <div class="parser-container"><div class="parser"><!– — —>ERROR<!- —— ></div></div> <!– two dashes is what delimits a comment, so the text "->ERROR<!-" earlier on this line is actually part of a comment –>

  36. Anonymous says:

    Wow, this really is good news! But why the secrecy, why didn’tyou let us know this until know?

    Support for the correct DOM event handling and the application/xhtml+xml MIME type are the only two things I cant think of right now that’s missing from that list.

    Any possibility that you will have any release in the future that will work in Windows XP prior to SP 2 or Windows 2000 (not likely, I know, but one’s gotta ask)?

    An updated review can be found at

  37. Anonymous says:

    Thank you Chris.

    Really, thank you.

    > And also, "blink" in text-decoration is not currently supported (I guess it should be in CSS1, right).

    Well, I myself wouldn’t rank that as a "bad thing", less blinkage = more surfage

    > It contains CSS errors to determine if a browser

    > follows the error handling defined by the CSS specifications.

    > It doesn’t contain HTML errors, because there is no

    >defined error handling for HTML, so there’s nothing to check.

    That couldn’t be more wrong, Acid2 does contain HTML errors checks, at least for HTML comments-handling errors

    > many existing sites that rely on this bug will cease to display properly,

    > unless IE7 fixes all of the bugs with web page formatting (such as the box model).

    WTF? box model has been fixed in Strict mode ever since IE6, if you still have to hack for the box model in IE6 then you don’t know anything and you don’t have the qualifiquations to whine in this blog

    ( 😉 )

    This selector should now work in strict mode, and I hope it doesn’t since it’ll probably be unneeded

    > I do want to remind you guys, though, that IE won’t be forgiven so easily.

    The only reason they won’t is because IE6 is alive and kicking and the non support for running IE6 in W2K will give the former quite a long lifetime left.

    If IE7 complies to the web-crew demands only the ones that are full of hate won’t forgive.

    The reason why (quite a lot of) people currently loathe MS is because they keep the web from going forward and painfully toughen people’s jobs… if they stop doing that we’ll be happy.

    As a side note to the IE team:

    1- Will the HTTP Accept header be tuned to reflect the true capacities of IE7?

    2- Will we see any modification in DHTML/DOM Apis

    3- Would xmlHttpRequest be avaible as a standalone object instead of an ActiveX?

    4- application/xhtml+xml? pretty please?

    Anyway, thanks again for your work and thanks for this list, the only thing I regret now is that you didn’t disclose it earlier, and didn’t tell anything to the community, which triggered the reactions to the previous IE7blog posts. Unlock the Black Box and let us see inside, we probably won’t harm you and may damn well help. (it’d help Molly’s sanity too, I think)

  38. Anonymous says:

    Will overflow in <tbody> be supported?

  39. Anonymous says:

    If you have been wondering what IE7 will do for web developers then take a look at Chris Wilson’s post…

  40. Anonymous says:

    How about pixel font resizing? Whenever a font size is specified in pixels, it won’t resize in IE6. Is this fixed in IE7? I surely hope so!

    I am really happy with this post and it really shows that you guys are ready to commit! 😉

  41. Anonymous says:

    Two things that really annoy me on MSIE:

    1) If you click on a label of a checkbox, radio button or the like this should be treated like a click on the checkbox itself.

    2) Dropdown menus (select tags) are always on top, you cannot make anything appear above them. (Yes I know, because they use native widgets, but that´s a poor explanation!)

  42. Anonymous says:

    > Jim, it has an HTML comment error.

    > That couldn’t be more wrong, Acid2 does contain HTML errors checks, at least for HTML comments-handling errors

    Guys, check it again. That’s not an error. The idea is that broken user-agents will think ERROR is part of the content, thus displaying the text ERROR. Conformant user-agents will understand that ERROR is part of the comment and will not display it. It’s not an error in the markup, the text ERROR is there to show when a user-agent has made a parsing error.

  43. Anonymous says:

    I hope that, given the fix list, you will make sure that IE7 won’t support the Tan hack (the famous ‘* html’) otherwise we will be feeding our IE5 and 6 workarounds to IE7.

  44. Anonymous says:




    All have passed Acid2 if you the latest builds.

  45. Anonymous says:

    One thing hasn’t mentioned too much is the conflictive zooming strategy. There are two type of conflicts: 1) For IE itself. IE’s Text Size is one kind of zooming function, which works for some sites, but does not work on others, such as the site: , no effect when you change the Text Size. The other function is DHTML style.zoom function. Also it works for some sites but does not work “properly” on others due to complicated layout. That makes me difficult to supply a simple Ctrl+ ‘+’ and Ctrl+’-‘ for the zooming function in our tabbed browser.

    2) This zooming difference between different browsers, such as IE and FireFox. NO STANDARD. You can try by using Ctrl+MouseWheelScroll.

    Because the screen resolution varies a lot, some site’s text is too small to read. How can we supply a simple and unified way let user to use?

    Also, in DHTML style.zoom, the text is not rendered properly.

    Mike J

  46. Anonymous says:

    Here is the good and bad…

    Security SHOULD NOT be a concern but the chair is physically melded to the table because of the prior need to kill off Netscape… Had Microsoft not molded IE in to the OS and forced companies creating ActiveX controls to go through a screening process I would not be charging 85 dollars to remove 500 pieces of spyware from her 5 month old computer (yes she has SV1).

    Chris’s post does however shed some light in the very dark areas of MSIE’s existence. Head banging errors ARE most important and ACID testing is ~NOT~ mission critical. I’m using a Gecko build from last week, Opera 8.01 (8.02 is out at the moment) and I have yet to see ANY browser pass the Acid 2 test.

    I will be banning all MSIE versions below 7 as soon as 7 is released and making my site both fully XHTML 1.1 and work with full AAA accessabilities standards in mind.

    Testing MSIE7 I am disappointed at some initial GUI changes.

    Tabbed browsing is ALWAYS on by default. Honestly this feature is only really useful for MAC users who obnoxiously have no taskbar (total pain in the ass on a MAC) and tabbed browsing (while it should be on by default in IE7) is not needed to be always showing.

    There are WAY too many buttons in IE6. There are NO buttons in IE7…just a left and right button?

    Buttons should be made available..

    Back, forward, stop, refresh, home | Favorites, History, Print

    The favorite’s Link toolbar isn’t on or available.

    Things I’d like to see from MSIE as far as GUI enhancements…

    Something implemented to clean up the THOUSANDS of junk bookmarks just blatantly DUMPED in to IE’s favorites. People are unorganized enough as it is. They will not embrace technology if others are deciding what their personal settings are set to…they must set their preferences, not programs their kids installed.

    Where is the GO button? Non-technical people do not use their keyboards for anything but typing in the address bar for example.

    If there are important things you folks are working on (such as :hover) that drive us crazy don’t keep us waiting for the end of the world. Delaying the prosperity of technology until competition fires up is not something the history books will embrace MS for.

  47. Anonymous says:

    Great news! Thanks for keeping us up to speed.

    > 1) If you click on a label of a checkbox, radio

    > button or the like this should be treated like

    > a click on the checkbox itself.

    In my universe, this works fine in IE6 if you use an ID attribute on the form control and an identical value for the FOR attribute on the label.

    > I will be banning all MSIE versions below 7

    > as soon as 7 is released and making my site

    > both fully XHTML 1.1 and work with full AAA

    > accessabilities standards in mind.

    Please keep in mind:

    1) Many corporate users are stuck with older versions because IT departments can be very conservative (who knows what intranet apps the new version might not work in, plus non-standard configs are a pain to support).

    2) You better believe Grandma will see few reasons to upgrade to IE7. IE7 will probably see a slow uptake similar to IE6’s.

    3) Only WinXP SP2 and Vista users will even have the option of installing IE7. There are still many people (especially business users) on Win2K, and even some people using Win9x.

    "Accessibility" for people using special browsers (screen readers, mobile devices) at the expense of millions of IE5.x and IE6 users is not very good accessibility.

  48. Anonymous says:


    To the person requesting more then two downloads being allowed at once… MS ~IS~ following a standard (which is the HTTP standard) which states that there should not be more then two active connections at once to any given server.

    Paste this in to a file with a REG extension to overcome the two file download limit (but be aware that this does not follow the HTTP standard…


    [HKEY_CURRENT_USERSoftwareMicrosoftWindowsCurrentVersionInternet Settings]



  49. Anonymous says:

    By the way, is there any chance that the version of MSHTML in IE 7 will generate strict correct code? Quotes on attributes, lower-case tags, well-formedness etc etc?

  50. Xepol says:

    Yes, this is a great post. It finally gives details we all want (unlike the doc file which is suppose to say what is in B1 but really just gives vauge promises about what may or may not be in b2).

    Glad to see the MS team active not just in the blogs but the comments. You can’t imagine how reassuring it is to see the MS team actually leaving comments replying to comments! SOMEONE IS LISTENING!!! (Although the very on-topic blog messages kinda give it away, actual participation in the convo goes that extra mile!)

    On the topic of security, again : Passwords on settings and downloads and installing active X controls -> Help me secure my machine from kids, employees, my friend bob who THINKS he’s an expert etc etc etc all of it. If you put the right security tools in our hands, it will REALLY help. (imagine how many times I would not have had to reinstall my kid’s machine if he couldn’t download EVERYTHING on the planet)

    Soulhuntre : Glad someone else gets it, sadly, no one loves a winner. I personally liked how Netscape bitched about MS’s monopoly while Netscape still had 80+% of the market. Maybe if they had focused on a quality product, they wouldn’t have lost all that ground to a relative new comer in the market.

    Hopefully, that lesson isn’t lost of MS. Just glad firefox is there to re-invigorate MS’s IE efforts, because it is definitely needed.

    – Xepol

  51. Anonymous says:

    >> we’ve planned out what we can and can’t do in IE7.

    Any chance of getting a list of things you can’t do then?

  52. Anonymous says:

    Chris Wilson is talking about which CSS bugs Microsoft are planning to fix for IE7 beta 2.

  53. Anonymous says:

    Excellent news! I’ve updated my post to reflect this:

    Chris & IE Team:

    I am wondering, with the quirky percentages, are you talking about the fact that IE can’t divide by 2? Case: using a background image on the body to ‘fake’ a full-height column layout, but having it be 1 pixel off depending on the size of the scrollbar (17px makes it off, 18 doesn’t, etc.)

    Also, are you guys aware of this issue:

    I’ve had IE crash on text resizes for a while, but then after a restart we couldn’t reproduce it. It was narrowed down to left-floated labels with a specific width, and a certain order of CSS properties. This order, however, I can’t reproduce. 🙁

  54. Anonymous says:

    This is great, it is nice to see that so many of the issues will be resolved, at least for those who can upgrade.

    What I’m wondering is how the changes in CSS handling will affect the hacks that are currently being used.

    It would be useful to know if any of the hacks that we currently employ to make things IE only, or non-IE only will do anything unintended in IE7; for example, if a hack still works to correct positioning, but the positioning no longer needs correction are we going to see an over compensation the other way?

  55. Anonymous says:

    I love you guys! If I had the money I’d come over to the US and kiss your feet!

  56. Anonymous says:

    Good to see such a candid post on what’s going on.

    A comment though: The requirement to have the OS verified and SP2/winXP is a bit rich: this sort of thing reenforces microsoft as the "big bad ogre". There is no reason at all that IE couldn’t be separate from the OS (it’s called good software engineering), and certainly shouldn’t be forcing users to run validation routines for other software products (i.e. the operating system) just to install. There are no other browsers that enforce this.

    On a similar note: with MSN messenger can you make it so that IE being online won’t affect anything but web browsing? I would like to be able to have IE in offline mode while still able to use MSN messenger (this prevents other applications popping up URLs)..

    Those things aside: keep up the posting on how things are progressing..

  57. Anonymous says:

    John A. Bilicki III,

    If you are making your site fully AAA compliant, you won’t be banning ANY browser. That bugged me a little bit..

    Anyhoo, this is fantastic news. I’ve been irregularly following this blog and this is the first post I’ve really gotten a "woohoo!!" feeling out of. I’m very familiar with the bang-your-head-on-desk feeling when working in IE, and this is certainly good news for my desk!

  58. Anonymous says:

    Thanks for this post Chris, it has certainly put my mind to rest that the IE team are serious about upping their standards game. I have to admit from previous posts, I had started to lose faith.

    Could I ask though, if you’ll be doing anything (or have indeed dome something) with the star selector hack – where by:

    * html body div {

    width: 100px;

    height: 200px;


    … is only read by current IE browsers?

    For example, if it were to remain, along with the addition of selectors, it could be used to pass CSS just to IE 7, such as:

    * html>body>div {

    width: 100px;

    height: 200px;


    … as a fall back in case IE 7 continues to have major CSS issues.

    I’d just like to ask your views on this particular selector?

    Again, thanks for the very informative and well received posting.

  59. Anonymous says:

    Thanks for the update Chris. This is what we needed to hear. I can’t wait for Beta 2 now.

  60. Anonymous says:

    First thing: can we get some closure as to the support for :before and :after and the content property in CSS? I don’t remember if it’s CSS 2.0 or 2.1 or 2.01 or, but it’s something absolutely required to make quality sites these days. Microsoft has always been the leader as far as dynamic content on websites, so don’t skimp on this one.

    Aside from that, I’d like to hear more information about JScript. I know it’s not strictly in the IE team’s court, but it is something tightly integrated with IE.

    I have had problems recently with JScript, specifically in the area of functions. I don’t know the terminology exactly, or even what is going wrong, but there is something really wrong with the ‘var x = new Function() {…}’ …functionality. It just doesn’t work.

    Another thing about JScript: I seem to remember that in the past, if a JScript application was taking too long to do something, as if it was stuck in heavy recursion or an infinite loop, it would notify the user and ask if the script should stop running. This seems to still work for infinite loops, but not always for recursive functions. No idea how to fix that, but it’s something that should be. With my test code, the latest versions of Firefox, Opera, and IE all crash/freeze, and I assume other ECMAScript-compatible browsers will also.

  61. Anonymous says:

    Well this is certainly odd. One day everyone could cut your throats, and the next they’re kissing your feet. And yes, I have to say, reading this post really made me cheer. Wouldn’t have hurt to announce it earlier though. 😉

    The only thing I really miss in that list is min/max-width/height. Is it too late in the game to add this to the list?

  62. Anonymous says:

    This is great news. Hopefully we can look forward to a swift 7.5 (or 8?) release with even further improved standards support.

  63. Anonymous says:

    Anyone who has relied on the various "hacks" to stop different browsers seeing different rules should expect to have to tweak them each time a new browser is released. That is why they are called "hacks".

    If you don’t want to worry about the maintenance of your hacks, you shouldn’t implement them in the first place. Just deal with it looking a little different cross-browser.

  64. Anonymous says:

    Someone mentioned specificity bugs in CSS; from my memory it has to do with wrong calculation of specificity on elements with pseudo-classes (like :hover).

    Also the following is broken:

    element.class1.class2 { }

    this will get applied to elements that have only class2 set, the class1 rule is ignored. This is probably a leftover from the original CSS1 implementation where only one class could be defined on an element.

  65. Anonymous says:

    I guess it is good that IE team thinks about complying a bit to one man-made standards, but I still find it more than ridiculous that some w3c makes standards, and not standard sites. I find all who suffer, bleed, have fever because of poor CSS positioning, itioning implementation seem more whiners than web designers to me. Don’t get me wrong, there are good people making good sites using all the w3c zealotish requirements, but there are a lot more of those who never ever use CSS, PNG, ABC, DEF and still complain about "standards in IE". My point: no point, w3c are crippled whiners.

    Wagga. IE rendering engine is good. Shall be better. If not, oh well…

  66. Anonymous says:

    Congratulations guys!

    Im ashamed to admit that the first page i visited after loading the beta was positioniseverything. I was happy to see most of the bugs exposed on this site were fixed.

    Thanks for the transparent png support…

    Thanks for the :hover selectors…

    Keep up the good work guys. Looking forward to the full release, although it will be difficult to wean me off of Firefox by then.

  67. Anonymous says:

    in reply to the person hoping for font sizes to resize using a fixed dimension (px). that i and many other people consider to be a bug. pixels are fixed dimensions and so shouldnt scale, if you want a font to resize you should be using a relative dimension such as em.

  68. Anonymous says:

    what’s about supporting selectors like input[type="submit"]? Yes there are more important bugs out there that are waiting to be corrected, but this could be a good (solved) issue.

  69. Anonymous says:

    Ah finally. This is the type of post we want to see on this blog and from Microsoft: direct answers. It’s a bit unfortunate that it took so long and there’s still a long way to go to make developers happy but this is a good step in the right direction.

    I was disappointed to see no mention of making "dotted" borders actually appears as dots instead of dashes and any mention of contextual selectors. If these are soley a feature of 2.1 (can’t remember off the top of my head), then it’s a pretty lame reason not to implement them.

  70. Anonymous says:

    > in particular CSS 2 ( 2.1, once it’s been Recommended)

    CSS 2.1 *is* recommended. See

  71. Anonymous says:

    As for CSS2.0 vs 2.1 please read:

    CSS2.1 is CR; it is implied that vendors who implement CSS2 should follow the 2.1 specifications.

  72. Anonymous says:

    All HTML related here, but what’s about JavaScript.

    Great could be better JavaScript performance. Especially I would like to see if you the memory leaks in your engine. And please make arrays and object creation and fill faster. In my Firefox this runs with doubled performance. And IE seems to get slower with the growth of the javascript array. Firefox does not have this problem.

  73. Anonymous says:

    It is great that you are now supporting :hover correctly. Are you also planning to support :active and :focus (and maybe even :disabled from CSS3)? Using the CSS3 pseudo classes allows most of the UI handling to be moved from js to css where it belongs.

    It would also be nice if you fixed your get/set/hasAttribute bugs in DOM but I guess that might be a bit too complicated at the current stage.

  74. Anonymous says:

    Chris Wilson beklager i IEblog at IE7 beta 1 viser s&#229; d&#229;rlig fremgang i sin CSS-st&#248;tte, men lover at dette vil bli mye bedre i IE7 beta 2.

  75. Anonymous says:

    This is all well and good, but there’s 2 major things missing.

    1. display: table-*;

    2. Box-model tables

    Currently IE6 renders tables "specially", rather than treating each <tr> and <td> as "display: table-row;" and "display: table-cell;" elements respectivly. (Hence why IE6 doesn’t support those CSS display property values).

    "display: table-cell;" is a particullary useful CSS property value, since it means that a box "fills" its parent (much better than "height: 100%;"), Firefox and Opera both support the table-cell, table-row, et al. property values flawlessly.

    With "display: table-cell;" support, it allows CSS layouts to do everything tabular ones can, with "full height columns" being one of the reasons why CSS is having issues with getting converts.

    Does anyone care to comment on this?

  76. Anonymous says:


    It’s be a pain having to support another version of IE, but the fact that it is getting better standards support will make that soooo much easier. Thank you very much.

  77. Anonymous says:

    This looks really promising guys! One thing that concerns me is :hover on all elements. I’ve mentioned this in a previous post about IE downloading css background-image’s during each hover causing images to flicker. Can you make sure this is fixed? Check this link for details:

  78. Anonymous says:

    About the guillotine bug.

    You may have fixed the bug between releasing the beta and posting here in the IE blog, I wouldn’t know. But if you meant that the guillotine bug is fixed in the current beta, you’re wrong.

    As Dave Shea points out in his blog here: the beta doesn’t fix it. See the screenshots:




    But then again, you’ve hopefully already fixed this. 🙂

  79. Anonymous says:

    The only negative opinion I have about this post is that it was a long time coming. A perfect summary and a great answer to most of my questions as a web developer. Good work!

    I hope for continuous follow-ups as you fix more bugs and address more issues, and I hope that the end result is that when IE7 ships and falls for "html > x" selectors instead of "* html x" ones, it will render the rest of the page (or at least a great deal) similar to the other browsers. And if it doesn’t, I hope that conditional comments will work ;).

  80. Anonymous says:

    The big news of the week is, of course, that Windows Vista Beta 1 has been released. In addition, IE7 Beta 1 was released as well. While I haven’t installed either yet, this is definitely good news on the road…

  81. Anonymous says:

    @Chris Wilson: It looks _very_ good, and I thank you for the decision to post these details.

    Am I right that one of the "large and difficult features" you mention is genrated content?

    @Chris H, and all the people asking about the * html hack: Whatever happens for this particular hack, it’s not a good idea to rely on browser bugs. Are you aware that there is a fully supported, backward- and forward-compatible way to special-case IE versions? Conditional comments are a much cleaner solution than CSS hacks. You could use e.g.

    <!–[if lt IE 7]>

    <style type="text/css">@import "ie-fix.css";</style>


    and put all you hacks for previous versions of IE in ie-fix.css. Note that it doesn’t affect validity, nor other browsers (no matter how exotic they are).

    @Jim: XHTML support could be even harder to implement than many people realize.

    As you note, the DOM itself poses some difficult problems: SVG and MathML have their own DOM specification, derived from the XML DOM. Combine this with the requirement of extensiblility, this means that it should be possible to subclass COM objects in the DOM tree. I’m not sure COM is flexible enough for this. Possibly something like javascript expandoes could be used, but it doesn’t look very clean.

    What is worse is the kind of implementation assumed by recent W3C works (esp. XHTML 2). They assume that they are free to make changes in the language, since XML-enabled browsers will be flexible enough to convert it on the fly. This means that the browser will have to convert the document object tree into an internal "visual" tree which can be rather different, _but_ both scripting and prgressive rendering will modify the document tree, which means that both trees must be kept in sync. Add the previous remark about subclassing, and your browser is turning into a nightmare. Sorting it out will require a lot of work.

    (Note that this is very acute for IE, since it is built as COM objects, instead of simply providing interfaces for embedding).

    @People complaining about the "integration into the OS": adding an HTML renderer library to the standard set of tools provided by the OS, and using HTML for parts of the interface in several programs, doesn’t look like much of a security vulnerability. (Some, of course, since it’s additional software). This is what the much-overrated "integration into the OS" amounts to. It’s not a matter of special hooks into kernel mode, or whatever some people imagined.

    @People who want fixes for memory leaks: As far as I understand, they _cannot_ be completely fixed in the current architecture (which combined a garbage-collected environement in javascript, and reference-counted objects in the underlying COM implementation. The real solution to this problem would be to move to a fully garbage-collected implementation (.Net), but this is necessarily a _very_ long term project, requiring a complete rewrite of the browser, plus resolving the backward-compatibility issue. I think such a projet would be a good idea, but don’t hold your breath.

    Keep up the good work!

  82. Anonymous says:

    Hooray! Thanks for delivering 😀 And as has been said above – keep it going!

  83. flaimo says:

    it’s great that you support alpha transparency PNG pictues, but you should not forget to think about a way how webdesigners can distinguish between IE7 and older versions. right now i’m using the "html > x" selector since every browser out there who supports alpha transparency PNG pictues also supports this selector. example:

    body p {

    background-image: image.gif;


    body > p {

    background-image: image.png;


  84. Anonymous says:

    Will IE7 have a text zoom feature like Firefox’s?

  85. Anonymous says:

    Well.. this looks promising. 🙂 Now I only have to update my pages like this:

    <!–[if lt IE 7]>

    Get a decent browser. Your version of Internet Explorer can’t handle this properly.


    But this will only happen if max-width and min-width will be supported in IE7. Those two little thingies are my major wishes. Fingers crossed! Keep up the good work! 🙂

  86. Anonymous says:

    Oddly enough, although I *loathe* the placement of the menus, toolbars, and address bar in IE7 when running XP, in Vista it actually seems to work — the elements match better in some sort of vague and indefinable way…

  87. Anonymous says:
  88. Anonymous says:

    Wow… Next thing you’ll be saying is that Apple are going to be using Intel chips…

  89. Anonymous says:

    I’d like to see a *good* Javascript console in IE7. Debugging JS on IE is a nightmare because of the poor debugger. It can’t even get the line number correct.

    Glad to see that IE developers, if not the project managers, take the standards seriously. Nice one guys 🙂

  90. Anonymous says:


    > right now i’m using the "html > x" selector

    > how webdesigners can distinguish between IE7 and older versions

    IE already has this functionality. It’s called conditional comments.

  91. Anonymous says:

    How about adding the :last-child pseudo-class as well? I know, I know, it’s CSS3, but the selectors-module is already a candidate recommendation and hey, how likely is it that the specs for :last-child will ever change? Not very. It shouldn’t be that hard to add, either. I’ll live even if you don’t add it, but :first-child is quite lonely without :last-child 😉

    It would mostly benefit list-based navigation systems, but could, of course, be used for anything.

  92. Anonymous says:

    This is excellent news.

    I’d like to see, at the very least everything in Dean Edwards’ IE7 JavaScript in the real IE7.

  93. Anonymous says:

    Thank you for posting this, it’s almost exactly the information I was hoping for.

    > Improved (though not yet perfect) <object> fallback

    Can you explain what exactly you mean by that?

    Also, are you aware of this bug that causes content to disappear when the browser window is resized? Any chance of getting it fixed, if you haven’t already.

    Can you please also provide the following information:

    1. a list of which CSS properties, from those which aren’t currently supported in IE6, that will and will not be supported in IE7

    2. A list of the individual properties that are currently buggy in IE6 that will/will not be fixed. (e.g. ‘width’ and ‘height’?)

    3. Details about the effect of known CSS hacks where the result differs from IE6.

    With this information, it should make it easier for authors to start writing their stylesheets with hacks that are less likely to cause disasters when IE7 is finally released. For example, if ‘min-width’ is supported, ‘width’ bugs are fixed, yet * html {} still works, then this common fix for IE6 will likely cause a problem in IE7:

    #foo { min-width: 10em; width: auto; }

    * html #foo { width: 10em; }

    If, for example, that’s the case, then we can start to prepare by changing the hack in some way, such as adding another "be nice to IE7" rule.


    *>html #foo { width: auto; }

  94. Anonymous says:

    I could use some help from someone who has IE7 b1 installed. It’s basically no effort – all I want is for you to load a certain webpage with a number of different stats services, so I can tell how good they are at ID:ing IE7. Send me a line if you could help me.

  95. Anonymous says:

    Oops! Email: d dottt naylor at telia dott com

  96. Anonymous says:

    Well this sounds great but I honestly have subconscious Twilight Zone music in my ears, this is a lage step for M$ that many of us never thought we would see.

    If you pull this off then Kudos to you, I hope they pay you well, then you should get a raise.

    I do have one problem with all this: A Eternal IE6. Maybe this has changed and I have missed it… but my last info is that IE7 was earmarked for Longhorn/Vista only. Only The Beta IE 7 would be available for XP and nothing for earlier OSs. That means people whith rather new XP PCs are not gonaa go Vista for a few years, so all the Win 98 (still alot out there) as well as ME, 2K and XP will still be using a Buggy IE6 for years to come. So all these slick improvements in IE 7 will make little difference for us Developers as long as large segments can not use it. We all know IE 5.5 has not yet trully died out, so imagine how long it will take IE6 to die if IE 7 only comes with a whole new OS?

  97. Anonymous says:

    not sure if this is related to css but are you aware of this bug…

  98. Anonymous says:

    I am afraid I can become a devotee of MSIE until it rcognizes and processes of MIME type application/xhtml+xml.

    James Pickering

  99. Anonymous says:

    >> 1) If you click on a label of a

    >> checkbox, radio

    >> button or the like this should be

    >> treated like

    >> a click on the checkbox itself.

    > In my universe, this works fine in IE6

    > if you use an ID attribute on the form

    > control and an identical value for the

    > FOR attribute on the label.

    Maybe he means <label><input …> Text</label> ? This should work by HTML Standard, I don’t know for IE.

    But how about "BUG: The FOR Attribute of the LABEL Object Resets Focus" described at;en-us;314279

    (The title gets the terminology wrong. IE resets the selection in the select element when the select element gets focus via label.)

    Will that be fixed?

  100. Anonymous says:

    I’d like to request that a number of CSS properties in relation to tables be added or fixed as well:

    * add CSS property border-spacing

    * add CSS property empty-cells

    * add CSS support for border when applied to COL or COLGROUP

    * add CSS property caption-side

    There are a few other discrepancies that I’ve remarked on here:

  101. Anonymous says:

    Chris, an appreciative "thank you thank you thank you" for taking the time to write this. Measurable goals! CSS 2! (I mean, hopefully, 2.1!) Someone pinch me. (Ow.)

    I trust Acid2 will be tackled in time, and I appreciate that there’s a lot more to this release than just CSS. What you’ve posted here matters all the same, it really does. Keep it up!

  102. Anonymous says:

    Looks like I was right, beta 2 is the browser we’ve all been waiting for. Microsoft has released a list of fixes that will appear in the second beta of Internet Explorer 7. I suspect the backlash after beta 1…

  103. Anonymous says:

    It must be said, I find this really really good news !

  104. Anonymous says:

    Reiterating the request for JPEG2000 support. The ideal case would be that it’ll just be another image codec that needs to be added. But if <object> works as image host (since IMG is being deprecated with XHTML2), then this’ll be fine, too.

  105. Anonymous says:

    I’ve waited a long time to hear this.

  106. Anonymous says:

    Please excuse the early morning fuzziness — or old age creeping in — I previously wrote …..

    "I am afraid I can become a devotee of MSIE until it rcognizes and processes of MIME type application/xhtml+xml."

    ….. which _should_ read:

    "I am afraid I cannot become a devotee of MSIE until it recognizes and displays MIME type application/xhtml+xml documents."

  107. Anonymous says:

    I admire the honesty of this posting. No bullshit bingo, this is what we can do, this is what we cannot do. So simple and it is so helpfull. Keep on making this type of postings, it makes discussions very interesting because they focus on the message.

  108. Anonymous says:

    Yesterday, Chris Wilson announced that the next beta of Internet Explorer 7 will contain many bug fixes…

  109. Anonymous says:

    David "W3bbo" R has a point about the display: table-cell, table-column,… stuff.

    I’ve tried playing with it in Firefox and it’s really an easy and even intuitive way to display elements (like divs) in a tabular manner.

    It would speed up my work tremendous!

  110. Anonymous says:

    The installer on XP SP2 gives me "The data area passed to a system call is too small" error after the window validation have been completed

  111. Anonymous says:

    This is good news, thank you for providing specific details at last.

    Unfortunately, it will take several years for IE7 to crowd out IE6 (especially since you’ve decided it won’t be available for Win2k). Which means that developers will *still* have to undergo huge amounts of "bang-your-head-on-the-desk" agony when we try to design a site for the next several years, because we will *still* have to do huge amounts of hacking and tweaking to get things to work correctly for the people who will continue to visit our sites using IE6 and below.

    I understand why you have made the business decision to not release ie7 for older OS versions. You want people who are still using 9x and 2k to buy XP or Vista from you, and you hope that the new features of IE7 will encourage them to do so. But, people don’t care about things "under the hood" when they decide if they should upgrade or not. They care about tabbed browsing and popup blocking and spyware, not standards compliance. That goes for both Jane user and for IT directors.

    So, what are the chances that you will be able to bundle *just the rendering fixes* and release them as a "rendering engine update" for IE6? Then, instead of spending an eternity fixing our sites to work around IE6’s quirks, we could just tell surfers "to see this site as intended, you need to install this bugfix for IE6." It won’t have much, if any, impact on your sales, but it will save web developers everywhere hundreds of hours of work, and it will keep us from damaging our keyboards with our heads.

    Oh, and from your perspective, it would help keep you from losing web browser market share. Because right now, if we are not given the time or budget to tweak and hack a site so it renders correctly for IE6-, then the only thing we can do is put up a sign saying "to see this site as intended, you need to install Firefox."

  112. Anonymous says:


  113. Anonymous says:

    Now this is the kind of thing you should be posting about all the time – bugs fixed, improvements added.

    Congrats on bringing IE into the modern world – I can’t wait to get my hands on beta 2.

    Is Acid2 a priority for IE7.5/8?

    My personal pleas (some may be in beta 1 – I don’t have access). In no particular order

    1) Please turn off all mime-sniffing except when browsing quirks-mode pages (so text/plain .html files are text, stylesheets require text/css etc)

    2) :last-child should be a doddle to implement after doing :first-child…

    3) <q>. Pretty please?

    4) Support * > html – conditional comments are nice, but are overkill if you only need a single IE-hack. Will allow IE7-targeting, whilst being valid CSS.

    5) Turn off filter: in standards mode

    6) Don’t choke on application/xhtml+xml. I personally don’t care whether you treat it as xml or html – just display it!

    7) Stop width/height being treated as min-width/min-height

    8) Fix Z-index support. This is horribly broken in IE6.

    9) .innerHTML should generate valid code.

    10) When calculating widths with percentages, IE should always round down. (example – 8×12.5% floats on a window width of 1004px (maximised Win98/’classic XP’) results in 8×125.5px, which are then rounded up to 8×126, causing the last float to wrap to the next line. We should have 8x125px, or a mix of 125px and 126px (Gecko does this) Hope that makes sense.

    (Those quoting hixie on CSS2.1 being a reccomendation should note that CSS2.1 is currently undergoing further modification and is back to working-draft status.) CSS2.1 should still be targeted though

  114. Anonymous says:

    (Forgot) – any eta on beta 2?

  115. Anonymous says:

    Sounds like a step in the right direction! Great to see the IE team be specific about CSS fixes and support.

  116. Anonymous says:

    No matter how application/xhtml+xml may be hard to implement, it NEEDS to be added. Programmers have waited long enough and another 4 years or so of waiting for a new IE version to support it would be terrible. Please add full support to XHTML pages in IE7, thanks.

  117. Anonymous says:

    there will Be support for application/xhtml+xml content (xhtml 1.1) and support for xhtml 1.1 (image maps and ruby notation?

  118. Anonymous says:

    &lt;p&gt;As everybody and his dog know by now, IE 7 Beta 1 has been released. this good news is being overshadowed by an apparently quite vicious campaign against Molly Holzschlag, who seems to have had the temerity to suggest that the IE 7 beta is actually good news for web developers and the standards movement in general, a message that encounters quite a bit of resistance among web standards fascists.&lt;/p&gt;

  119. Anonymous says:

    Are there any plans to inmplement SVG support ?

  120. Anonymous says:

    1. postion:absolute + left + right with no width still doesn’t scale with Beta 1 which you say above is fixed.

    2. same deal with top + bottom.

    3. Printing with transparent pngs results in a mess of black and transparent still.

    4. <a href="…"><table>…</table></a> still doesn’t work right. Clicking on any section of the table should result in the link being followed. (Of course firefox has a bug where if you wrap the <tr> in an <a> it does really weird things, so even though IE does support wrapping the <tr> you’re still screwed so you can’t create nice image + text that doesn’t wrap under the image layouts that are links.)

    The absolute left + right and top + bottom stuff is the most important thing you could ever do for web design. Please please please fix it.

    For those of you that want a javascript that fixes it for now check out our site @:

  121. Anonymous says:

    All of the improvements here are welcome, but I notice that they all target rendering – no mention of the W3C DOM event model. Support for this would make development of complex applications much easier and more reliable.

    What are the arguments against publishing the source to the rendering and Javascript components, as Apple has done with Webkit and Webcore (but not Safari)? This might get IE out of its current third-world status more quickly.

  122. Anonymous says:

    > wrap the <tr> in an <a>


  123. UnexpectedBill says:

    Hey…as others have said, this is an *excellent* post from a member of the IE team. I’m glad to hear some concrete news on what rendering issues are being fixed, although they don’t really impact my sites at the moment.

    Something I’m curious about–so it’s been said that there is no IE7 going to be released for any OS prior to Windows XP Service Pack 2. How’s about giving us just a rendering engine update for IE6 Service Pack 1 on the platforms where it is still supported in some way–like Windows 2000, 98(se) and Windows XP Service Pack 1.

    This might be a lot of work for the IE team, but I assure you it would *not* go unnoticed. Many users just aren’t ready to upgrade for a while yet…myself included…

    Oh, and if anyone knows–is Windows Vista going to have an option to look and work in the way that "classic" versions of Windows do? I really don’t see myself liking, using or even taking advantage of the new look.

  124. Anonymous says:

    Could you tell us something about current CSS workarounds for IE that will break your layout in IE7.

    like the HolyHack

    * html .aclass {

    height: 1px;


    Should we add rules like

    *>html .aclass {

    height: auto;


    to make sure our layout stay working in IE7?

  125. Anonymous says:

    For IE team:

    It is widely known that IE 7 will be available only for XP SP2 and up, because it relies on some security related features of it.

    As it is mentioned earlier, rendering improvements on IE 7 will not be significantly advantages for web developers, since this will force IE 6 to remain around for some time (as IE 5.5 currently is).

    Question is: is there any possibility (or plan) as for backporting the improvements made for IE 7 (in the fields of (X)HTML and CSS) to a say "IE 6.5"?

    I guess many web developers like me would like to have this question cleared out, preferably in its own post.

  126. Wombert says:

    There still is something missing from CSS1!

    Maybe you didn’t mention it, but at least IE6 does not imply a width (or a height) if both "left" and "right" (or "top" and "bottom") values are specified. It would be great to have this in IE7, too.

    Oh, and if you could fix the background flicker on hover… that’d be cool 🙂

  127. Anonymous says:

    Is there a way we can weed out all the trolls?

    No matter how good a job you do, people just want to attack you. Please ignore them and carry on doing a great job that you are.


  128. Anonymous says:

    I really doubt that the IE team is going to backport anything. Look at the Lifecycle model posted a while ago. They have to support anything they release with patches and service packs, etc. Why would they release IE7 for Vista and XP, then an "IE 6.5" for others and then do what? IE8 for Vista and XP and nothing else? If 6.5 came out, wouldn’t they have to release patches for any security issues in it? If IE7 uses security features of Vista and XP, then there are likely to be more security issues on a 6.5 release.

    I don’t see it happening. XP has been out for four years this fall and Vista comes out next year, right? Windows 2000 is out of support pretty soon and the other operating systems are out of support. People complain about the IE team moving too slow but then want them to make frankenstein versions of their software for out of support operating systems?

  129. Anonymous says:

    I’d like to clarify something I said earlier here, saying that web developers won’t forgive Microsoft so easily. What I meant is this:

    Microsoft showed their business plan very clearly during the IE6 era. It became obvious that Microsoft had no interest in doing anything to help us unless they absolutely had to. Right now, it seems to us that the only reason IE7 is being developed in the manner that it is now is because Firefox has slashed through their market share so fiercely, especially on technology-oriented sites (Internet Explorer has under 50% usage share on my website, for instance). It doesn’t strike us that Microsoft suddenly cares about us; it just seems like they’re desperate to keep their customers.

    So right now, we do not believe that Microsoft’s business plan involves wanting to make us happy. It seems that they just care about making us "happy enough" that we don’t try to pull more people away from them. This is the mentality that makes us not trust Microsoft, because this means that if they manage to stomp out Firefox, they might just sit on their browser for five years again. As long as I think there’s a chance of this, I’m still going to dislike Microsoft.

    I do not dislike Microsoft just because they are Microsoft. I applaud many things that Microsoft does, and especially lately, it seems like they’re finally starting to clean up their ways. But they aren’t there yet. We don’t trust them like we trust other companies like the Mozilla Foundation, Opera, IBM, and Google. They aren’t yet at the point where they are willing to constantly go out of their way to do the right thing, unless it’ll earn them a direct profit. If Microsoft shows that they can consistently work for the better good without relying on all of these marketting gimmicks (communist Linux? Give me a break) and half-assed jobs, then things will turn around for them.

    If a serial killer says he has changed and donates some blood, it doesn’t mean that I’ll invite him over to my house. All I’m saying is that it’ll need to be a consistent and true effort to earn back respect, and it won’t happen overnight.

  130. Anonymous says:

    I have a suggestion for IE7. On the Favorites, I’d strongly suggest having a "Sort by Alphabetical" option, as well as possibly a "Most Favorites" feature similar to <a href="; target="_blank">the Maxthon browser</a>. I’d also, along similar veins, look into optional mouse gestures and, as has been mentioned before, allow browser windows to be closed by double-clicking their title.

    Finally, I’d have the toolbar settings be remembered on each launch, as well as stay the same from tab to tab unless otherwise enabled.

    Great startup, though, guys! Keep up the good work!

  131. Anonymous says:

    Excellent news! That second list reads very much like my own wish list, though there’s two sets of features I’d still like to see:

    1. min-width/max-width and min-height/max-height

    2. :before and :after generated content

    (I realize that to implement #1 properly means changing the way width behaves, since it currently acts as min-width.)

    Any plans for either of these?

  132. Anonymous says:

    Well, I didn&amp;#8217;t get around to downloading IE7&amp;#160;beta&amp;#160;1 yesterday, so I won&amp;#8217;t be able to check it out over the weekend. But it&amp;#8217;s become clear that, from a web developer&amp;#8217;s point of view, all the action is slated for beta …

  133. Anonymous says:

    Good news! And will the support of XHTML be improved ?

  134. Anonymous says:

    I’m so glad to read this and all the planned changes in beta2. It is sounding quite good. Will you please give IE7 it’s own unique hack signature… (as in the tantek hacks) as it will save our skin if there are any issues. But Chris thanks for being so honest about what is coming up. Much appreciated.

  135. Anonymous says:

    The very base of IE is wrong because it doesn’t pass the Acid2 test.

    Chris Wilson is promising they will work hard to make IE better. I don’t believe it. They don’t really love their product. Working hard means working day and night, releasing betas every week, not once in a year. If there is something diffuclt to implement – this means the basic part has run out of itself. It is as if trying to make a 3d game out of isometric engine.

  136. Anonymous says:

    This CSS1 bug has been mentioned at least twice already but I got a test case for it…

    "If both top and bottom css properties are set the height should become flexible. Meaning that if top and bottom are set to 5px then the height should be the content height of the parent minus 10px."

    Same applies to left and right

    The test case was made for another browser that also has some issues with this. Not as serious as the IE ones ofcourse 😉

  137. Anonymous says:

    Thank you SO MUCH for this post! It’s exactly what I’ve been waiting for since this blog was started. I’d be happy with this list for release, even, as long as the IE team is going to stick together and keep working hard on version 8 =) But one thing at a time.

  138. Anonymous says:

    Would you guys be able to arrange for the W3C to release a CSS 2.2 specification? Particularly to fix the height, min-height and max-height properties, which should be analogous to the width, min-width and max-width properties respectively, but aren’t. That causes problems for anyone who wishes to set a height percentage for anything that isn’t an absolutely positioned box or a box that doesn’t have an explicitly stated height. This should have been fixed for all boxes in CSS 2.1, but it was only fixed for absolutely positioned boxes (hence why I mentioned that this wasn’t a problem for absolutely positioned boxes). The syntax and data types section in the CSS 2.1 specification states that "Percentage values are always relative to another value," which forms a discrepency with "If the height of the containing block is not specified explicitly (i.e., it depends on content height), and this element is not absolutely positioned, the value computes to ‘auto,’" which is what the CSS 2.1 specification states in regard to percentage values in the height, min-height and max-height properties. Considering this, you should be able to slip the fix in before the W3C releases a CSS 2.2 specification (should you go ahead with my request) without having to worry about being at odds with the specification as with respect to these properties, the specification is at odds with itself.

    Also, would you prioritize fixing the stuff highlighted in Acid 2? It tests for support of things that web developers want most, and to ignore it is to ignore what web developers want most, regardless of what you say or think you’re doing.

    Lastly, I disagree with Jonathan Fenocchi’s statement regarding * html { … } as it is a hack, used to differentiate CSS for IE’s rendering engine, from CSS for everything else. It is used because IE’s rendering engine up to this point, stinks. If you guys fix the problems in IE7, then IE7 shouldn’t be using the legacy broken codepath that just so happened to render correctly in previous versions, but should use the correct codepath that was meant for standards compliant browsers.

    Besides, anyone using hacks should know darn well enough that a rendering engine update will probably break them:

    In other words, I agree with Martijn ten Napel, regarding the tan hack.

  139. Anonymous says:

    If security is a priority, please make the phishing detector NOT communicate to MS!

    Also, is there any official place to post suggestions and bugs?

  140. Anonymous says:

    I would love to see a small improvement to be made in the context menu of IE. Don’t laugh … please move the "print" command away from the "refresh" command (or let the user customize the order in the contextmenu).

    If I had a dollar for every time I wanted to refresh the page, and by accident pressed the print command I would have been a milionaire.

  141. Anonymous says:

    In my other post, "That causes problems for anyone who wishes to set a height percentage for anything that isn’t an absolutely positioned box or a box that doesn’t have an explicitly stated height." should have been "That causes problems for anyone who wishes to set a height percentage for anything that isn’t an absolutely positioned box or a box that isn’t in a container with an explicitly stated height." I retyped the section that section was in so many times that I had written it incorrectly by the time I posted it.

  142. Anonymous says:

    " "I do want to remind you guys, though, that IE won’t be forgiven so easily."

    Well, lets be honest. Most of the people making the most noise will >never< forgive MS simply because it >is< MS. Their whole personal self image is defined by how rebellious they are and how they can read /. and "stick it to the corporations!" by pirating movies and complaining about MS.

    MS could ship perfect CSS2 support and they would complain it isn’t Firefox "bug compatible", that it doesn’t retro-install into Windows 3.1. For an encore they would complain that because a fully compliant browser is in Windows that they need to fire up the whole monopoly machinery again because they can’t compete anymore on features. "

    what a bunch of crap! if m$ could make a half way decent product there wouldnt be this issue.

    And I doubt you’ll have to worry about "what if ms made a perfect bowser cause that certainly wont ever happen or even anything close. And your stereotyping was lame is that the best you can do too support your company who "inovated" technolgies and features that everyone else has been using for a long time? Personally I think it’s pretty sad when the biggest corperation in the history of the world can’t make something as good as a few ppl working for free can produce.

    AM i anti ms? you bet i am and I will continue to be as long as they make my job as a developer more difficult. But whats fun is when I have to explain to clients why their site is costing more due to the additional time it takes to hack it to death to work right in ie and the reasons behind it, how they react and how it opens their eyes and its real fun when they go download ff or better yet when they go buy a mac.

    And now we have to go back and modify our sites to work in ie7? And partial png support? damn, it’s only been what 8 years now and we only get partial support?

    I used to love ms but this kinda crap opened my eyes. Heres my advice to ms, If you can’t make it right then dont make it at all, let the competant ppl do it for gods sakes.

    And lets be real, windows will never be secure with the chicago code base, it’s just not gunna happen, ms please throw in the towel on this and try from the ground up.

    Thank god for Apple 🙂

  143. Anonymous says:

    Great job guys. As many have already mentioned, your list is a great start. I agree with others that other bugs, like min/max-width/height should also be added, and some serious thought needs to be put into how these changes will effect current hacks in use.

    I’m not really suggesting that all hacks should be supported, but at the very least some analysis on the impact of these hacks is warranted.

    As for others comments:

    CSS 2.1 is *NOT* a recommendation anymore. It was demoted to Working Draft and changes are once again happening. This puts *ANY* browser that supports the "old" CSS 2.1 behavior in a bit of a pinch, and illustrates why CSS3 behavior should NOT be implemented until the standard is set in stone.

    What we don’t need is another case of browsers implementing a non-existhing standard because too many people wanted to push the envelope.

  144. Anonymous says:


    Thank you for posting this update on your plans for CSS in IE7. I and many others have been waiting for a statement like this from Microsoft since IE7 was first announced. Seeing a list of items that will be addressed in IE7 is also very encouraging.

    Thank you not only hearing our concerns frustrations, but also working to address them. Before this post, I was very pessimistic about what Microsoft’s commitment to improving standards support in IE7. I am now actually excited about IE7 and looking forward to its eventual release. Keep up the good work.

    P.S. Add me to the list of people hoping that the min/max-height/width properties can also be added in IE7.

  145. Anonymous says:

    > If you guys know of anything else missing form CSS1 please post it here.

    !important rules are not considered correctly in IE6. Will this be fixed?

    Thanks for listening, finally…


  146. Anonymous says:

    You guys rock! I’m so happy that IE is going through all of these changes. Thanks for posting the update, for the transparency, and for the good intentions.

    If you haven’t seen this link already, it is a source of some great CSS2 tests to run IE against. Have you considered it?

  147. Anonymous says:

    Nanobot – I understand your sentiments in questioning long-term commitment to standards; but please don’t compare the IE team to a serial killer. 🙂

    As for the "* HTML" selector issue – actually, it’s currently fixed (that is, it no longer works) in beta 2; however, I’m on the fence as to whether we should ship that (it does help our appearance on the acid2 test), since it is in use in the web today for browser switching. I’d welcome feedback on whether we should fix it in IE7 or not.

    Minghong – the font-weight issue is an issue in the Windows font mapper, which I expect is why you see similar behavior in Opera also. And "blink" is (by design and demand, I might add) completely optional for a completely compliant implementation. We have no plans to implement it.

    Rosyna – you are completely right, the Acid 2 test also tests error handling.

    Ivan – input[type="submit"] is an attribute selector, which was on my list.

    Bjorn – we’ve fixed the Guillotine problem – we haven’t fixed the floating-element-next-to-element-with-layout problem. More on that in a later post.

    Lachlan Hunt – More on object fallback later.

    Dan – partial PNG support? Explain please.

    Christoph – do you have an example of how !important doesn’t work correctly?

  148. Anonymous says:

    Great news on future CSS support in IE7

  149. Anonymous says:

    One question. Any good reson why IE 7 won’t be availible for Win 2K?

    I’d also like to join the "petition" for application/xhtml+xml and XMLHTTPRequest()

  150. Wombert says:









    will this be possible with IE7? (left+right assignments should imply a width here)

    It’s very important IMO.

  151. Anonymous says:


    IMHO, the * html thing should be fixed (i.e. it should not match anymore). IE 7 will be much more standard-compliant than IE 6, and thus will probably have a better behavior with "standard" CSS than with whatever hacks were devised for IE 5-6.

    People who want to add IE 7-specific stuff should use methods that are documented to work, such as conditional comments, or request a proper selector for browser sniffing (which you could implement as soon as possible).

    That’s only an opinion, of course.

    W.r.t. application/xhtml+xml support, I want to point out that I think that it should be supported as soon as possible, too. My point was that it’s important to do it right, and I don’t want a clunky implementation that will cause more problems later. Make sure it’s on the list, but no dirty hacks.

  152. Anonymous says:

    That’s great to hear fellows!

    I, myself am not an Internet Explorer user nor will I ever be but that doesn’t mean I don’t care. Internet Explorer has a huge marketshare and every bug you guys fix has a lot of impact on the internet.

    Perhaps you won’t be aiming for ACID 2 for IE7 but I’m sure you will in the future. Each improvement for IE is an improvement for the internet!

  153. Anonymous says:

    بالاخره یه خبر مهم برای عاشقان سینه چاک Firefox بی ریخت. یکی نیست بگه آخه این موجود چی داره که اینقدر کاربر پیدا کرده. نمیدونم هیچ کدوم از کسانی که Firefox رو آخر برنامه میدونن تاحالا شده با Opera…

  154. Anonymous says:

    Just wanted to say thank you for addressing more of the compliance issues.

  155. Anonymous says:

    Internet Explorer 7: Now in beta testing for developers

    The first stage of the beta process for Internet…

  156. Anonymous says:

    Will the z-index issue be fixed for listboxes and drop down menus?

  157. Anonymous says:

    Chris and others:

    Regarding the "* html" hack, I think this should be left in.

    Practically all CSS hacks rely on either a) parsing errors or b) lack of support for selectors.

    It seems to me that these will be fixed in Internet Explorer 7. So there won’t be any ways to target Internet Explorer 7 with CSS hacks unless the "* html" hack is left in.

    Yes, there are other ways to serve CSS to Internet Explorer 7 alone, but all of them have downsides. In many situations, CSS hacks are the best solution.

    We know Internet Explorer won’t fully conform to the latest CSS specifications, so most people will want to use CSS hacks. The "* html" hack is relatively benign because it is used by choice, not triggered accidentally.

    It’s also able to be used in conjunction with hacks for previous versions – so you can serve rules to Internet Explorer 6 and below with "* html", serve rules to other browsers with "html>body", and serve rules to Internet Explorer 7+ with "* html>body".

    There’s really only one reason to take this away, and that’s because people are serving bad CSS to Internet Explorer with it today, when that bad CSS might not be appropriate for Internet Explorer 7.

    This is a problem. But it’s part of a bigger problem – all the other hacks will be going away too, so people using hacks will have to update their stylesheets *anyway*. The argument to take away the "* html" hack is an argument that prevents 1% of updates, when 99% of updates will have to happen regardless. The benefit of taking away this hack is miniscule.

    I think the benefit of keeping the "* html" hack vastly outweighs the downside at this time. The only people who would be negatively affected by the downside are the people using this hack but no others. The people who will be positively affected by keeping the hack is anybody who wishes to target Internet Explorer 7 with a CSS hack.

  158. Anonymous says:


    "Dan – partial PNG support? Explain please."

    Among the numerous reviews out there, there is mention to these 2 facts:

    1. PNG transparency is not correctly displayed when the image is selected.

    2. PNG transpareceny is buggily printed.

    By the way, another "memo": dotted != dashed … :/

  159. Anonymous says:

    I know chances are i’m going to get ignored, and this probably isn’t the forum for it, but on the off-chance some developer is reading this and likes the idea.. I have a feature request..

    I would like to see conditional comments that can be embedded directly into stylesheets, along the lines of:

    /*IF ie* … */

    IE has many interesting non-standard CSS properties that unfortunately do not validate (such as the amazing directx filters and gradient, etc), and this functionality would allow us validation-zealots to use them neatly without having to include extra stylesheets, and hopefully make these features something we see more frequently.

    It will also help organise the handling of the various rendering differences between the major browsers, and be an acknowledgement that there are problems, and a positive step towards helping cope with these differences, which I’m sure is something even the most hardcore standards advocates can get behind.. at least when they have a deadline coming up..

  160. Anonymous says:

    Sam Riley,

    > pixels are fixed dimensions and so shouldnt scale, if you want a font to resize you should be using a relative dimension such as em.

    That’s a common misconception, px units are not fixed dimensions in the context of CSS, they are relative units. Read the spec:

    James Hancock,

    > <a href="…"><table>…</table></a> still doesn’t work right.

    Yes it does. I pointed out last time you posted that, Internet Explorer does the right thing. Your code is broken, don’t blame Internet Explorer.


    > do you have an example of how !important doesn’t work correctly?

    IIRC, CSS 1 said that !important rules in user stylesheets were overridden by !important rules in author stylesheets, and later CSS specifications reversed that logic. I don’t use !important very often, so I don’t know if you ever changed from CSS 1 logic to CSS 2+ logic.

  161. Anonymous says:


    Might I suggest a possibility for the * html selector issue.

    Maybe you could provide some mechanism (registry entry, command line switch, whatever) to turn this feature on and off for Beta 2. This way, we as developers can test the effects of it both ways and provide feedback on which way is better.

    I think right now we really don’t have enough information about what the effects will likely be. I’d suggest enabling it by default, with the ability to disable it for testing purposes.

  162. Anonymous says:

    I’m not sure if this is possible, but once IE7 goes RTM, would you guys be thinking of making an update to the IE6 engine to provide 100% CSS 2.0 support as well?

  163. Anonymous says:

    Oh! excellent!

  164. Anonymous says:
  165. Anonymous says:

    Please support application/xhtml+xml MIME type. Microsoft claims to be a leaders in XML technology. Yet, the most common dialect of XML on the Web (XHTML) is not supported. This may be a difficult feature to implement, so take an extra 6 months to release IE 7. If you commit to this feature, I am sure most developers will gladly wait an extra 6 months to see this feature in IE 7. I urge you to demonstrate your leadership in XML technology by supporting application/xhtml+xml MIME type.

  166. Anonymous says:

    I’m not sure if some one has mentioned this already, and I will admit that I haven’t gone thru all 166 comments…

    What about the Box Model hack ( ""}""; )? I have to use this nasty thing with every new CSS layout I develop. It has to do with IE not calculating div widths correctly, when combined with margins, borders and padding.

    More detail on this here:

  167. Anonymous says:

    > Christoph – do you have an example of how !important doesn’t work correctly?

    It applies the declaration, but not the change in importance.

    body { background: black !important; background: red; }

    Will result in a red background.

    body { background: black !important; }

    Will result in a black backround.

    It’s what I use for IE workarounds, so fixing it would be high on my list if all the other fixes are complete. 🙂

  168. Anonymous says:

    Bruce, that extra six months to add XHTML support is an extra six months web developers won’t be able to use the new HTML, CSS and PNG supported by Internet Explorer 7. I’d much rather start using them earlier and wait for XHTML support. The lack of that stuff really slows me down as a web developer. The lack of XHTML doesn’t.

    Haydur, Microsoft fixed both the box model and the parsing bug that you exploit in that hack in Internet Explorer 6, released four years ago. There’s nothing more Microsoft can do, unless you think they should travel back in time and fix it sooner 🙂

  169. aeberhar says:

    Well, I’m disappointed to hear that IE 7 won’t pass ACID2. However, I genuinely appreciate the candor, tardy though it may be. I also appreciate the honesty it took to say that fixing IE to pass the test would be no small task. I don’t agree with the criticisms of the test levied by Chris Wilson, but c’est la vie.

    Focusing on what makes developer’s lives easier is a great approach, but I’m not entirely convinced it’s related to a better browsing experience for non-developers.

    In any event, with the ACID2 "litmus test" now comfortably dismissed, a little more formality with respect to exactly which CSS changes (apart from the aforementioned bug fixes) we should expect would be greatly appreciated.

  170. Anonymous says:

    I am glad to hear you are working toward CSS improvements. When the reviews of the first beta came in all the web developers at our shop got very depressed. Not a single pixel changed in the acid2 test.

    The time we have to spend making CSS render properly on IE is really disheartening.

    IE is a good browser because of the sweat put forth by web developers making sure their creations look good in IE as well as more "standards compliant" browsers.

    While you may argue about what validity ACID2 is, in practice other browsers require fewer tweaks to make CSS "work".

    Think of the all the developers you will doom to years of tweaks by not improving…

  171. Anonymous says:

    I hope you fix the follow bug..

    HKEY_LOCAL_MACHINESOFTWAREMicrosoftInternet Explorer

    and that’s value, "Build" is "62900.2180"..

    You know, Build must be written "2900.2180".

    And there is many many version bug in ie6.

  172. Anonymous says:

    Sorry about the serial killer thing. I was just trying to deliver the point. 😉

    Here are some issues that I haven’t already addressed or I feel need repeating.

    Gamma correction in PNGs needs to be supported. It can sometimes be difficult to fit PNG images into the design when the image is slightly darker or lighter than it should be. There are also other issues with the current alphatransparency support that are detailed on many reviews out there.

    Supporting the q element properly is an absolute must. I use the q element all the time, and I’m currently forced to remove the quotation marks with CSS and then surround the element with quotation marks in the markup if I need it to work right in Internet Explorer (and then non-CSS browsers who support the q element correctly get two pairs of quotation marks).

    The tabindex property needs to be properly supported.

    This one isn’t as important, but the noscript tag currently isn’t supported properly. The contents appear when JavaScript is disabled, but this is not the same as the correct behavior. It should display the contents whenever a script element with an unsupported or disabled language appears before it. But again, this isn’t as big of an issue as others.

    The value of 0 for the colspan and rowspan attributes needs to be properly supported. This value should cause it to span all columns/rows.

    The @import CSS rule currently doesn’t work when used with a media type.

    At the moment, the grammar for pseudo-classes (and I assume this would also apply to pseduo-elements, none of which are supported in IE 6) is incorrect. Pseduo-classes are currently ignored unless they are the last part of the simple selector. According to the specification, I should be able to do something like a:visited:hover{} or any other combination of any number of pseudo-classes/pseudo-elements, and it should recognize all of them, combined with a logical AND.

    Counters should be supported. Firefox will likely support it in 1.5, or else definitely in 2.0. Counters will be very useful.

    DO NOT release IE7 without support for CSS table displays and "inherit" for all properties.

    Width and height need to be handled correctly. When I say 100px, I mean 100px, not at least 100px.

    Support for the outline properties would be nice, too. I believe Firefox will also have this in 1.5, and Opera has had it for a long time.

    As for DOM support, all I can say (other than supporting the standard DOM event model, which will keep us from having to write two versions of a lot of our code) is to clean up some of this: <;. Internet Explorer’s DOM support really isn’t tragically bad like the CSS support, but we’d appreciate some tuning up.

  173. Anonymous says:

    The beta version of Internet Explorer 7, too old? According to GoDaddy… Click for larger and clearer version I might by the standards compliant argument for Beta 1 (as evidenced by this heated thread at IEBlog), but too old? It’s…

  174. Anonymous says:

    Will we see some form of new JavaScript capabilities? I’m particularly curious about getters and setters for all (native and custom) objects. Mozilla has had this functionality for quite some time now, and it is a feature that would be greatly appreciated.

  175. Anonymous says:

    > Focusing on what makes developer’s lives easier is a great approach, but I’m not entirely convinced it’s related to a better browsing experience for non-developers.

    The less time we spend working around Internet Explorer’s bugs, the more time we spend on actual features.

    > Gamma correction in PNGs needs to be supported.

    This would cause a lot of trouble for colour consistency between images and CSS. Read:

    Gamma is best left alone until it can be solved in a general way for all formats.

    One other thing that I just thought of that would be *really* useful is supporting prototypes properly for JScript, e.g. = … As far as I am aware, other browsers have supported it fine for ages, but I’ve not been able to use it because of lack of Internet Explorer support.

  176. Anonymous says:

    Thank you for this good news. I believe the general consensus here is that we all wish the IE team good skill in coming out with IE 7.

    I would just like to add that in working to address standards compliance you currently have a great opportunity to make IE shine. It’s certainly worth doubling the manpower of your team to make this happen (and I know adding more talent to the team is within your capabilities).

    Once again, good skill on the continuing development.

  177. Anonymous says:

    James Hancock <; wrote:

    > 4. <a href="…"><table>…</table></a> still doesn’t work right.

    That’s also invalid HTML and its behaviour is undefined in the spec. <a> is an inline-level element and <table> is block-level. Inline-level elements cannot contain block-level elements, so any behaviour exhibited by browsers for that invalid code is acceptable. You, as an author, should not write invalid code and browsers cannot be expected to interoperably implement that which is not defined.

    Chris Wilson [MS] <; wrote:

    > As for the "* HTML" selector issue – actually, it’s currently fixed (that is, it no longer works) in beta 2;

    Good news!

    > however, I’m on the fence as to whether we should ship that (it does help our appearance on the acid2 test), since it is in use in the web today for browser switching. I’d welcome feedback on whether we should fix it in IE7 or not.

    IE7 *should not* intentionally support any hacks that are dependant upon non-standard behaviour. Authors that make use of hacks should be very well aware of the fact that those hacks may no longer work in a future version of the browser, even though it’s equally likely the bug(s) they were used to fix may still exist. As a result, it has been long known that hacks should be used very sparingly and the decision of whether or not to continue with non-conformant behaviour should not be affected by irresponsible authors.

    Even if IE fully implements CSS2.1 selectors and there are still bugs that require fixes for IE7, there are already hacks that can be used, simply by making use of some of the implemented CSS3 selectors to pass styles to other browsers. e.g. Here are some hacks that will hide some styles from IE7, based on what we now know (or assume) will/will not be implemented.

    p { color: red; /* IE and Opera */ }

    p:not(foo) { color: green; /* Gecko */ }

    (I don’t know about other browsers)

    @import "data:text/css,html%3Ebody%20p%7Bcolor%3Agreen%20!important%3B%7D";

    /* There has been no indication given that IE7 will/will not support data: URIs yet,

    * so this assumes that it will not.

    * For Gecko, Opera and anything else that supports data: URIs:

    * html>body p { color: green !important; }


    html>body p { color: blue; } /* IE7 (at least) */

    * html p { color: red; } /* IE 5/6 */

    p { color: black; } /* Anything else that doesn’t apply the other rules */

    Note that I do not, in any way, recommend the use of such complicated hacks, unless absolutely necessary. I publish this only as proof that other hacks may still be usable in IE7, even if ‘* html’ is correctly fixed.

    Personally, I very rarely use CSS hacks and only occasionally make use of ‘* html’ usually for the purpose of substituting ‘width’/’height’ for ‘min-width’/’min-height’ in IE where necessary. Considering that min/max-width/height have now been implemented, that would mean most of my stylesheets will continue to render perfectly.

    (Although, I’d like to point out to several commenters above that ‘width’ and ‘height’ in IE6 are not exactly equivalent to ‘min-width’/’min-height’. It just so happens that the behaviour of ‘width’ and ‘height’ in IE exhibits one similarity to them, making them an acceptable substitute in some cases.)

    Barry <; wrote:

    > I would like to see conditional comments that can be embedded directly into stylesheets, along the lines of:


    > /*IF ie* … */

    Barry, as I believe you are well aware, conditional comments were discussed and rejected in a thread on www-style back in April 2005. If you are not the same Barry from that discussion, sorry, but the chances are highly likely considering your request. Conditional comments encourage the very bad practice of authoring for a single browser and it’s comparable with the evil practice of browser sniffing (though slightly more reliable, and their abuse is not quite so wide-spread). CSS hacks suffer from the same problems in some cases, although I strongly discourage the use of either unless absolutely necessary. There are many other reasons, please refer to this thread for more information.

  178. Anonymous says:

    That is absolutely splendid guys. I thought you wouldn’t have a complete support for CSS 2 and only very little support for CSS 2.1 but you have proven me wrong here.

    The fixes you say have already made to IE7 should help me and other developer a lot!

    Now I am only starting to wonder: What of you could be looking over att CSS 3? CSS 3 has two very interestings things as I can see it. It has the border-radius which helps making beutiful websites and it has a good way of doing columns which helps sites bringing a newspaper look to their site which ain’t really possible today. If you would add those two also then you would certainly has as good support as Firefox 1.5 when that one is released for CSS.

  179. Anonymous says:

    There is one bug you have missed in that list that I believe is extremely important.

    The bug is the flickering and disapearance of background images used on links.

    All you have to do to recreate this bug is to use something like this:

    a:link {background-image:url(image.gif);}

    a:hover {background-image:url(image_over.gif);}

    In Internet Explorer this causes the images to disappear completely for relatively long periods. It may also cause the images to flicker.

    This bug is even more apparent in IE7 Beta1. All other major browsers are free of this bug.

    Otherwise I’m extremely glad you’re addressing the other CSS issues. Just add this one to the list as well, thanks.

  180. Anonymous says:

    Here are a few bugs that need fixing:

    * xml prolog causes browser to go into quirks mode

    * A blank line causes the browser to go into quirks mode.

    Sometimes we are using tools that insert blank lines (like CMS systmes) at the beginning of the document. This makes it hard to run in strict mode.

  181. Anonymous says:


    When you open a new tab by clicking on the blank tab button hanging to the right of your open tabs, it goes to about:blank (annoying, would prefer the home page, but that’s not the security issue). BEFORE it loads the page however, the new tab displays the caption of some previously loaded page.

    Yes, the delete browsing history… does seem to clean this ‘memory’ out, but it nukes all my cookies, so I don’t ever forsee using it (now I have to go log back into a half trillion sites and type in passwords etc etc just to get the memory back) -> The delete browsing history is TOO efficent to be useful except when I wanna take a computer to the shop.

    Oh, and back to the toolbar issue. IE 7 doesn’t just nuke the google toolbar, it also whacks the MSN Toolbar with equal ease, in fact, it pretty much randomly destroys my toolbars as it feels like it (and turning on or off a toolbar -> OYYY instant randomiation of order).

    Oh, and you are loading the toolbar layout too late into the startup process. I should NOT see some default layout only to EVENTUALLY see it switch over to my toolbar layout once the page loads (might be when the page first starts downloading, can’t tell, my connection is too fast to be 100% sure)

    I REALLY wish you guys had a better feedback process like the Visual Studio guys, or if you do, I wish your next BLOG documented it and that your blog site prominently linked to it.

    I have a LOT of minor glitchy issues and some security stuff I wish was more solidly tracked than a few rambling comments to a blog.

  182. Anonymous says:

    Hi guys –

    I just want to say thank you so much for fixing the PNG stuff. For me, that was one of my biggest wwww pains. It’s fantastic to hear that you’re taking compliancy seriously, and I cant wait to see how IE 7 performs.


  183. Anonymous says:

    Chris Wilson wrote:

    "Improved (though not yet perfect) <object> fallback"

    It’s good to hear you’re also working on the <object> tag, and hopefully the fallback method will be working as needed when IE is shipped.

    However, would there be any chance of adding a ‘fallback step-through’ segment for this, where by web developers can test each fallback they’ve written in to their web pages?

    For example, if I had:

    <object "svg attributes">

    <object "flash attributes">

    <object "java attributes">

    <p>Multimedia is not available.</p>




    … and I had the Adobe SVG viewer, I wouldn’t see the flash or java alternatives by default, and hence couldn’t see how each fallback appears. I’d find being able to try out each fallback in turn would help developers build a better web site for visitors who don’t have the first multimedia option available.

  184. Anonymous says:

    Back from holiday to some very good news. No, Bush hasn&amp;#8217;t backed down on climate change, but IE 7 beta 2 will support CSS about as well as I&amp;#8217;d hoped. Thankyou, Chris Wilson and your team!…

  185. Anonymous says:

    Chris: here’s an example of the IE specificity bug:

    Here’s an example of the IE Z-index bug:

    Also the issue with the background-images getting reloaded when changed either through script, by some :hover pseudoclass or a classname swap is a large problem that really needs attention.

  186. Anonymous says:

    <<< For example, if I had:

    <object "svg attributes">

    <object "flash attributes">

    <object "java attributes">

    <p>Multimedia is not available.</p>




    … and I had the Adobe SVG viewer, I wouldn’t see the flash or java alternatives by default, and hence couldn’t see how each fallback appears. I’d find being able to try out each fallback in turn would help developers build a better web site for visitors who don’t have the first multimedia option available. >>>

    How about removing the external layers to test the internal ones?

    Sounds like rocket science doesn’t it?

  187. Anonymous says:
  188. Anonymous says:

    the list seems just too wonderful to believe 🙂 the only thing dark cloud in my mind for this is – how is ie7 going to handle the well-known css property hacks (underscore, backslash). is ie7 going to accept the definition ‘_width’ (rewriting the previous ‘width’ attribute) as well as ie6 does? or is this the kind of thing, that ie7 will require to be written according to standards?

  189. Anonymous says:

    Masklinn, if I’ve understood your comment correctly, your solution doesn’t actually test IE’s fallback abilities.

    If I were to remove the first <object> tag from my mark-up and reload, IE (and all browsers) would see the flash object as being the first object, so no ‘fallback’ would be taking place, as such.

  190. Anonymous says:

    Any chance of getting the border conflict resolution algorithm implemented properly for table-* borders? Here’s a few hundred test cases that show what I mean:

    Note that there’s an issue with a few tests in the first set (20050702), which are awaiting a response from the CSSWG to be updated and added to the official test suite. This set is also available as both HTML4 and XHTML1.1.

    The second and third set (20050703 and 20050705) need to have HTML4 versions generated for IE to test them.

  191. Anonymous says:

    IEBlog : Standards and CSS in IE Coming CSS support in IE7 – looks like they’ll keep tailing behind. Embarassing for a company of Microsofts stature and size (tags: css browsers Microsoft standards)…

  192. Anonymous says:

    A standards-compliant IE7 doesn’t help web developers one bit unless Microsoft patches the CSS bugs in IE6. Most companies still use Windows 2000 and see no reason for an XP upgrade, so it would be out of the question to drop pre-IE7 support for any professional website.

    Since these CSS bugs are not linked to any of the security enhancements code (hopefully!) it should be quite easy to backport the fixes. If not, something is either very wrong with the IE codebase or Microsoft wilfully refuses to fix their bugs.

  193. Anonymous says:

    Following from Henk Tiggelaar’s post, this update is not exciting unless M$ extends the upgrade to those people who don’t want to (or cannot) upgrade to XP/SP2. There will otherwise remain a large installed base of crappy IE5/6.

    Hopefully those people will migrate to Firefox or another standards-compliant browser.

    As for security, I guess it would be less of an issue if M$ didn’t insist on the browser being so integral to the OS (which is also why it’s difficult to test multiple IE versions on Windows.

    All in all, nice work Chris, but wake me up if it ever gets really interesting zzzzzzzzzzzzzzzzzzzzz

  194. Anonymous says:

    Chris H: Why not just disable the SVG plug-in via the manage Add-ons menu option? Wouldn’t that test the fallback, without modifying the code?

  195. Anonymous says:

    Jim, if Microsoft does not take say extra 6 months now to add support for XHTML, it will be another 5 years or so before IE 8 comes out. This will make the XHTML spec 10 years old before it is supported by IE. Lack of XHTML support in IE is hurting us all and is holding back development and adoption of new Web technologies such as MathML.

    IE 7 can be in beta right up to the release of Vista. If some customers need tabs or RSS or certain CSS fixes right away, they can use the beta versions. This approach worked for Firefox and it will work for IE. Microsoft has always said that they will release products when they are ready. You should release IE when it’s support for Web standards is as good or better than Firefox. Until then, we’ll use the betas.

  196. Anonymous says:

    THANKS A LOT. Allthough a Firefox user, knowing that IE7 will support CSS is a great relief. I still think it is a pity that IE7 won’t be broadly supported by Windows OS, but, still, that’s a very good point – the one all webdesigners have been waiting for.

  197. Anonymous says:

    "A standards-compliant IE7 doesn’t help web developers one bit unless Microsoft patches the CSS bugs in IE6."

    Well, we’ll just have to spell it out clearly to our visitors: "Either upgrade to Firefox, or if you’re on XP/Vista get IE7, or view this website totally muddled up – it’s your choice."

  198. Anonymous says:

    I just wanted to echo a previous comment about using conditional statements in CSS similar to the conditional comments in HTML. Possibly something like

    /*[[if ie gte 7*/

    css declarations here


    This would allow us to target ie7 and up with anything browser specific.

  199. Anonymous says:

    I’m not entirely sure how the CSS spec states that !important should act, but I think there is a bug in the IE implementation, rather than Firefox, Opera, etc.

    It seems that IE won’t make note of the importance of a property until the declaration has been closed, for example:

    p {

    background: #fff !important;

    color: #000 !important;

    background: #000;

    color: #fff;


    While Firefox and Opera will keep the background white and the text black (as these are set to !important) IE will overwrite (or cascade) the two properties with the two below it, setting the background to black and the color to white.

    If you split the declaration in to two, for example:

    p {

    background: #fff !important;

    color: #000 !important;


    p {

    background: #000;

    color: #fff;


    Then IE will respect the !important call for the two properties, so the second declaration will not overwrite the first.

    I have no idea how IE works under the bonnet, but I expect it is because IE doesn’t check for the !important keyword until after a declaration has closed, where as other browsers make a note as they come to it.

    I’ve put together a small example below. You’ll see important1.htm and important2.htm are the same in Firefox and Opera, but different in IE 6 (I don’t know about IE7 BETA, as I don’t have access to it).

    Like I said though, I don’t know how the spec defines the !important keyword, so it could be that IE is correct and Firefox and Opera are incorrect.

  200. Anonymous says:

    Visual Studio 2005 (ASP 2.0) is going to support XHTML. This means Microsoft development tools will be capable for producing XHTML Web sites that cannot be properly processed by IE 7.

    At the very least, Microsoft should make a statement as to when it plans to support XHTML in IE.

  201. Anonymous says:

    AndyC, thanks for the tip. I have to be honest and say I hadn’t noticed the "Manage Add-ons" option in IE. Was this added with XP SP2, or has it been available for a longer period?

    Anyway, thanks for the heads up, I would have thought it would allow the testing of <object> fallback as needed.

  202. Anonymous says:

    Features Request:

    1) A better bookmarks manager. Minimally, something that would make sure my bookmarks are upto date. Remove changed address. Maybe tag the bookmarks like

    2) Go button in IE 7. Lot of people still use that.

    3) Possibaly, some kind of API for the bookmark managers such as to have a direct connection with IE. So, people can switch to using instead of the IE bookmark manager.

  203. Anonymous says:

    Wow, compared to what I normally see in here, we have virtual "lovefest" going on….. Are my eyes deceiving me?

  204. Anonymous says:

    Would it be possible to displayed dotted borders as dotted borders? Right now dotted borders are displayed as dashed borders.

    "I’m not sure if this is possible, but once IE7 goes RTM, would you guys be thinking of making an update to the IE6 engine to provide 100% CSS 2.0 support as well?"

    They are, it is called IE7. :p

  205. Anonymous says:

    If this is true it will be very interesting to see. Right now i have quite a few tests that I really did not intend to have tests.

    I started with a set of css temaplates and css that even have the hacks needed to make ie work and it still mamanges to abnormaly mangle it. When I saw the ie7 beta I grabbed it and tried it out. I cant tell you how dissapointed I was when I saw nothing had changed.

    If you are truley going to truly work on the list above that would be great but as I have been having to program arround ie’s bugs for years now I’m not going ot hold my breath. I will believe it when I see the results in a browser that is truely usable.

    for those interested here as some examples of ie failing. They all work great with every other browser I tested with and not just firefox.

    I want to say again the above was soley for my won testing I am testing designs use source ordering and so far ie is the only browser that cant seem ot quite pull it off. I have many more like this on my hard drive that have never been published simply because they were for my own playing.

  206. Anonymous says:


    what be great if IE conditional comments would be allowed inside regular css files. This would mean easier maintainable code.

    The star hack: * html exampleDiv { some css declartions here} that only IE WIN sees. Will it stay on in IE7?

    The html>body selector, will it be available in IE7

  207. Anonymous says:

    some bugs I ran into in IE6

    – reduced font-size of text inside a <code> element

    – IE CSS ‘bugs’ like:

    * html {}

    {_property: value;}

    – no !important supported

    – in addition to Tino Zijdel’s element.class1.class2 { } also element#id.class {} and other permutations don’t work in IE6

    – furthemore I would like to refer to for a complete list of CSS-constuctions are IE doesn’t support

    PS elem.setAttribute(‘style’, ‘some css’) doesn’t work in IE6

  208. Anonymous says:


    what be great if IE conditional comments would be allowed inside regular css files. This would mean easier maintainable code.

    The star hack: * html exampleDiv { some css declartions here} that only IE WIN sees. Will it stay on in IE7?

    The html>body selector, will it be available in IE7

  209. Anonymous says:

    You are so far behind the game. What’s the point of trying to catch up when there are good solutions out there that you could slot in for free? I’m sure that you *can* catch up, but it is worth the money? Why not do something new rather than solving old problems?

  210. Anonymous says:

    Please leave "* html" as long as you keep hasLayout chaos and broken width/height. This trinity is known as "Holy Hack". It’s not that glorious, but helps a lot when dealing with broken float model.

  211. Anonymous says:

    A few comments regarding things I have read here…

    First of all, IE7 will not have 100% CSS support. Asking that is completely unreasonable, as no browser in existence currently has 100% CSS support or even 90%. Based on the progress given here, I’d expect IE7 to have somewhere in the neighborhood of 60% support. That’s about halfway between where it is now and where Firefox and Opera are at.

    In my opinion, no features should ever be added with the sole purpose of allowing browser-specific hacks. Webpages should be made to the standards, and it should be up to the browser developers to make sure those websites display how they should. A lot of the reason that people will have to scramble to get their sites looking right in IE7 is because people were using those hacks instead of writing their sites how they were supposed to. In my opinion, web developers should code to avoid the browser quirks, not take advantage of browser quirks.

    The ACID 2 test is important, but it’s not as important as a lot of people are making it out to be. Certainly, it doesn’t test some of the most annoying problems with IE. Passing the ACID 2 test should be a definite goal, like it is with all browsers, but it doesn’t need to be top priority.

    Gerben: It sounds like the <code> issue is just a matter with IE’s default stylesheet, unless you mean something that I haven’t noticed. So that wouldn’t be a bug, just a difference in defaults (the CSS 2.x specs don’t provide a normative default stylesheet). Also, !important is mostly supported, but there is that one bug mentioned in previous comments.

    As for the issue with IE6 not supporting these new CSS features, it doesn’t bother me that much. This can be considered simply as a case where a company stopped supporting an older product, and if users expect their unsupported browser to work properly in a world of ever evolving technology, they’ll have to upgrade (which in this case would require upgrading to a new OS) or what is presently the easier alternative, switching to another brand. In this respect, IE7’s release will finally give us a more practical excuse not to cater to IE6’s incompetence. It’s five-year-old software, it’s time to upgrade.

    XHTML certainly needs to be supported. My main concern here is that Microsoft has a history of not fixing old design flaws because it would break applications. If Microsoft is to give us partial XHTML support in IE7, they’ll have to follow through this time and get it implemented correctly later, even if there are webpages that would break from the fixes. We’re talking about XHTML here; authors need to start doing things the right way or else expect total failure to display.

  212. Anonymous says:

    @Lachlan Hunt & Nanobot:

    About the * html issue, as I already said, I agree with you: CSS hacks should be avoided. However, assume that enough web developers *want* some simple way to feed customized CSS to IE. Whether it’s the right thing or not, the IE team would eventually have to give them such a tool. I think it would be better to give them clear feedback on this subject.

    My opinion is that they should first target maximum standard conformance, i.e. fix all CSS bugs, including those used in hacks. Ideally, this should be all, and web developers should rely on CSS’ fallback mechanisms for graceful degradation. If they *must* provide a special syntax for IE-specific CSS rules, then I vote for some kind of dedicated syntax (whether it’s a kind of conditional comments, a proposed CSS @ rule, or anything else) rather than keeping old quirks. I also add a request that, if such an extension is defined, the documentation contain a notice advising to use it as sparingly as possible.

    The rationale for this is that such an extension should:

    1) be as orthogonal as possible with current CSS specs

    2) be more expressive than a simple "is it IE?".

    Again, I’m not suggesting such an extension, and personally I’d think the current mechanisms (CSS fallback, the CSS rules suggested by Lachlan Hunt, HTML conditional comments) are more than enough for IE 7.

    If you assume that CSS browser sniffing has to be added, what do you think?

  213. Anonymous says:

    No browser should ever have to be sniffed. As one of the lead devs involved in postnuke and the one who added css sniffing in the first place a coule years ago I can say that this should never ever have to be like that. The browser should either conform to standards or it will simply go away.

    As others have stated I simply refuse to program any longer and hack around for hours to fix simple things that the browser should have done properly to begin with. These kinds of issues are why in the web developer community there has been a mass exodus to other browsers that are more standards compliant already. On my site for instance were I get mostly web developers I see about %40 firfox usage. It is to the point were firefox, konqeror and safari pretty easily outnumber ie. This is for not the one I listed before. My site does not promote any browser in particular and I dont have a bunch of propaganda either way.

    This is just traffic from all over that is mostly web developers. I might also add that even though it is all mostly about php programing that my usage does not favor linux either and most of my users are windows users. This says that if ie/micrsoft does not do something to change thier way of doing that they have already lost the game. By far the majority of the mind share right now is on firefox and it shows in the development community. Once the developers are mostly using other solutions the users follow to maintian compatability.

    To have mechanisms in place to allow such behavior is redundant and also a waste when it could be coded properly to beign with.

  214. Anonymous says:

    Nanobot, you said:

    "As for the issue with IE6 not supporting these new CSS features, it doesn’t bother me that much. … if users expect their unsupported browser to work properly in a world of ever evolving technology, they’ll have to upgrade (which in this case would require upgrading to a new OS) or what is presently the easier alternative, switching to another brand. In this respect, IE7’s release will finally give us a more practical excuse not to cater to IE6’s incompetence. It’s five-year-old software, it’s time to upgrade."

    Unfortunately, we will have to cater to IE6’s incompetance, because people will not be able to upgrade, and some may not feel comfortable ith, or know about, alternatives. I guess all we can do on our sites is to promote those alternatives and hope that people move away from IE.

  215. Anonymous says:

    As one of the participants in the W3C MathML working group, Microsoft should add MathML support to IE. In other words, besides third-party add-ons that render MathML, IE should have its own MathML support.

    It’s a really good thing you’re fixing all those bugs and incompatibilities in IE, and making security improvements, but don’t forget new features 🙂

  216. Anonymous says:

    Thanks for the CSS enhancements Chris !

    I don’t know if are already aware of this bug, but it should be quickly fixed:

    When clicking on the label of a select element, the first element contained in it is selected. This is not very important, but using this bug you can even select optgroup elements !

    I have to use Javascript on my sites because of this bug, please fix it in IE7, it would be great.

  217. Anonymous says:

    I agree with Lionel here: IE-specific hacks should not be necessary. If IE7 will fix both the parsing bugs in CSS as well as implement the specifications for CSS level 2 (mostly support for child-selectors as well as min-max height/width) and fix the current CSS bugs for which the hacks were necessary most sites would be just a-ok.

    If not; you could use any of the other methods mentioned of which I feel the HTML conditional comments is the best candidate since it is IE-only, allows for targetting specific versions, is a documented feature and keeps documents valid.

    Meanwhile I also found another specificity bug that is not just IE-only (also Firefox and Opera get it wrong):

    most browsers seem to confuse a pseudo-element with a pseudo-class attribute in CSS.

  218. Anonymous says:

    This is really surprising and wonderful news. Thanks for making Web developers’ lives better even though it will probably end up helping the competition (when some Web developers drop IE6 compatibility and Win2K/Win9x users upgrade to Firefox).

    I also feel personally satisified since forcing Microsoft to make IE better is one of the my goals in working on Firefox.

    Rob (Gecko developer)

  219. Anonymous says:

    There’s another bug which I haven’t seen documented, which affects text selection.

    When you use { position: absolute; } on an element, it becomes impossible to select text within that element using a mouse.

    If you try to select a small amount of text – for example, just one or two words – you instead get a selection from your cursor position all the way to the start or the end of the absolutely positioned block.

    Removing { position: absolute; } or changing it to another value such as relative corrects the behaviour. So too does changing the DOCTYPE – I think this issue occurs only in standards mode, not in quirks mode.

    I’ve thrown a quick test case online:

    Same HTML, different DOCTYPE, no problem:

    You asked for feedback on * html support. I say turn it off. As several commenters have already said, we knew when we used * html that it was a short term strategy, and we accepted that. No browser should be deliberately supporting broken code.

    Conditional comments are a much cleverer solution to the compatibility problems we face and so long as they continue to be supported we’ll get by without * html just fine. In fact, if you pick up on the suggestion of conditional comments within CSS files, IE7 support should be simple.

    I’d also like to back the calls for min/max/plain height and width to be correctly supported. Together with the other new support you mention, this would make IE7 a much more attractive browser to work with.

    Finally, I’d like to echo the praise for this post and others like it, such as the new one on printing. Clear statements of support, or lack of support, are much more useful than vague generalisations. The quality of the comments on this post reflects the quality of the post IMO. Well done.

  220. Anonymous says:


    Chris: great list, I’m looking forward to the next beta, including all additional things that will be implemented/fixed and are not on this list yet.

    HTML in Acid2: I’d say that’s an SGML error, not HTML.

    application/xhtml+xml: no please, unless it’s done well. Just accepting it and treating it as tag soup HTML would be the worst.

    Go button: put it back :).

    Ecmascript: will be see improvements here as well? I’m thinking memory leaks and improved DOM standards support? Also, when creating lots of JScript objects, IE slows down a lot (this problem doesn’t exist in VBScript).

    What I’m still missing, and would like to see fixed as well:

    * Font-size inheriting into tables

    * no * html in Strict mode

    * Multiple classes support

    * <q> support

    * images flickering when dynamically moved with cache disabled


  221. Anonymous says:

    @Rick: indeed selection of text is a problem with positioned elements. It seems to differ though with the mousedriver used sometimes; i know of simular problems with position: relative and some none-standard mousedrivers. It could be API-related…

    For those pleading about XHTML support: I don’t feel it’s that important since for most real-live situations HTML will do just fine. Most people use XHTML with a text/html mimetype anyway, which in essence is just (mallformed) HTML to the browser, so what would XHTML mimetype support make for any difference?

    Also most sites using proclaimed XHTML don’t respect appendix C ( ).

  222. Anonymous says:

    I am happy to see Internet Explorer’s rendering engine being updated again. I am thankful for the bug fixes, but most especially these two: line-height and double-float.

    I am also enthusiastic about the new features. Fixed CSS positioning, variable-transparency PNGs, :hover on all elements, abbr, and fixed backgrounds. Yummy. I do have two questions though. Are there any plans to include :before and :after psuedo-element support in CSS? If so, would :hover work on them? I’ve found that combining those three things can lead to a LOT of awesome possibilities, and it would be nice if the answers to those questions are "Yes."

    Anyways, thanks for the update! I can’t wait to hear more 🙂

  223. Anonymous says:

    While Beta 1 isn’t going to stop the bad words bein spoken about IE over here, It’s great to hear that you’re working on all the stuff that we want you to fix.

    Thanks for the honest insight.

  224. Anonymous says:

    I hope you guys at the development department understand, but I think many webauthors will rather wait and see before raving on the new IE. Unfortunately MS has too often never delivered what they promised. Though if all those bugs were really fixed we’ll have to rewrite webpages; all the present hacks for those bugs will trigger unwanted rendering. But I guess we should be able to make just one html and one css page for modern browsers, bravo.

    And as far as the ACID2 test is concerned; No-one really obliges you to get IE7 to pass the test, but it may be a good idea nevertheless if only for the status it brings. It sounds even like you dismiss the test as just a wishlist of a few elitist webgeeks.

    The test may be a wishlist of functionality and standardcompliance, but it is a much a wishlist as say, the eland-test for cars. While the eland-test is not compulsory in most countries, carmakers will see to it that their cars do pass that test. Just ask Daimler-Benz

  225. Anonymous says:

    Lots of improvement it looks like. It’s amazing what a little competition from Firefox/Safari can do 🙂

  226. Anonymous says:

    Thank you, thank you, thank you! I think we all greatly appreciate this list on CSS standards. I was angry and frustrated about reports from IE7 beta 1, but now I have more patience and look forward to future updates.

    I few things that I would really like to see:

    – min-width / min-height support

    – fix for the flicker problem when using :hover on background images

    – no * html in strict mode

    – user font resizing for px specified font-size

    Thank you again for giving some insight into standards.

  227. Anonymous says:

    System Information program displays msencode.dll is dll of IE. But you open the file, and open file property, you can see this file is dos-file. Can’t you fix it really?? The whole MS employee can’t fix it?? Are you stupid?

  228. Anonymous says:

    Please remove the * html thing from standards-compliant mode. This shoyuld have never been used in the first place, as there is a perfect solution for specific IE feeding: conditinal comments.

    Since there is a solution that works 100%, there is not reason to keep the buggy behavior.

  229. Anonymous says:

    This is great news!

    But how are you going to handle compatibility issues with sites that rely on IE6 bugs? For example, sites generated by some WYSIWYG editors rely on the double margin bug and other IE quirks. Will there be some backward compatibility mechanism, along the lines of doctype switching, to address this issue?

    Also, support for the table CSS properties should satisfy the purists on the table vs. CSS issue.

  230. Anonymous says:

    Sounds great!! Can’t wait to try the beta!

    I’m curious about "display: inline-block;" – will it work on all objects (block and inline) on IE7?

  231. Anonymous says:

    It is good that you guys finally started to come around and support standards to some extent at least, the bug fixing is welcome for web developers but still this is really too little too late.

    Think about it, when you fire up IE7 and go to a webpage that’s not standards compliant because it was designed for previous versions of IE (you’re fault) IE7 will simply revert to the older rendering mode, perhaps some sort of quirks mode again or something of the sort. Am I right?

    Otherwise older sites that are poorly coded (majority of websites sadly) won’t work in IE7, so you have to do this.

    But then, how does this solve anything?

    Think about it, if I made my website for IE6 not supporting standards and not wanting to spend the time working out all the bugs and workarounds usually needed to make a properly coded website work (more or less) in IE6 (and earlier versions), and it still works exactly the same in IE7 what reason do I have to change my website? None.

    Furthermore, IE7 will not have a bigger share then earlier versions for quite some time, the fact that it will only come with Vista and as an update to XP doesn’t help either because there are hundreds of millions of people with 2K, and 98 still.

    So as far as standards adaptation go, it will take a lot of time for that to happen. Thanks to the pseudo-standard you guys created earlier.

    Seriously, let’s say now I can make a nice standards compliant page that’s properly coded and works in Firebox and Opera and IE7. It still doesn’t work in IE6 and I still need to do all the old workarounds just like before.

    So really what’s changed? I don’t even know what at this point you can do.

    All I can say is that at least you should make IE7 a critical update and push it with other security updates so it have a greater market share.

    And from this point on accept the standards.

  232. Anonymous says:

    @ Bruce: user font resizing for px specified font-size is already possible in IE6. Unfortunately the default setting is ‘off’ and it’s hidden in a hard to find place (Tools > Internet Options > Accessibility > Ignore font sizes).

  233. Anonymous says:

    > A standards-compliant IE7 doesn’t help web developers one bit unless Microsoft patches the CSS bugs in IE6. Most companies still use Windows 2000 and see no reason for an XP upgrade, so it would be out of the question to drop pre-IE7 support for any professional website.

    > Since these CSS bugs are not linked to any of the security enhancements code (hopefully!) it should be quite easy to backport the fixes. If not, something is either very wrong with the IE codebase or Microsoft wilfully refuses to fix their bugs.

    What utter nonsense. Have you any idea of the amount of refactoring some of the above bugs will have required to fix?

    In any case, having now two versions of IE6 will make CSS hack management utterly impossible.

  234. Anonymous says:

    Regarding the * html bug; I’m still in two minds about this one. For one, it is a bug, and should be fixed to work as documented, but for the other, it has the potential to generate a possible IE 7 exclusive hook, which could be used down the line if the need arose (*>html { ie-7-only }).

    As Chris has said "I’m on the fence as to whether we should ship that since it is in use in the web today for browser switching", and while I agree it is a handy bug to have, I think that at the end of the day, it should be fixed, and developers should be encouraged to make use of ‘hacks’ a little as possible.

    I’d suggest leaving the fix in, as with the promise of all the current bugs being reviewed and corrected, its use shouldn’t really be needed.

    "But what if later on, we find IE7 has more CSS issues, but our hook is gone"?

    Because I feel the question above it valid, I would like to put forward a vote for a new conditional comments system within IE 7’s CSS engine, as with the case with the HTML parser. Rather than relying on hacks and bugs to filter styles as needed (IE 7 bugs could exist), using a fully though out, standards compliant, extensible method would benefit both developers and users (or so I believe).

    Rather than just throwing together a quick replica of the HTML conditional comments, perhaps you could start a stand alone blog entry so that you can put together everybody’s view in on place, rather than in this thread, as it is quite obvious there are people both for an against such a method.

    Personally, I think the addition of conditional comments in to CSS has no real downsides, if done correctly. What’s more, with a proper definition and documentation, other browsers could implement the same thing, so that down the line, there will always be a back up method to setting styles for one particular browsing device … as a ‘just in case’ fallback.

  235. Anonymous says:

    Re the * html question – whether to leave it out or in:

    Like most others here, I say get rid of this, i.e. so that IE is compliant.

    The use of this hack has been to get round all the problems which you’ve now listed as fixed.

    While IE7 may well still have CSS rendering bugs in, there exists the solution

    <!–[if lt IE 8]>

    <style type="text/css">@import "ie-fix.css";</style>


    which lets people get round this.

    So Chris, please DO ship the "* HTML" selector fix. Otherwise you’ll never get rid of it.

  236. Anonymous says:

    This is fantastic news! I feel like the forehead-shaped depression in my desk is well-earned, if IE will in future be easier to work with.

    I’d like to recommend that Microsoft introduce a "Deprecated/Obsolete" way of dealing with old quirks.

    Someone in the comments above recommended that IE still render the old bugs so as to make sure that badly coded sites still work. I think this is a Bad Idea, and that it would be better that each new release of IE have a list of Deprecated items, and some Obsoletes as well.

    A lot of sites out there are still using really bad elements such as Font, etc. Those developers will continue to use those deprecated items as long as browsers continue to support them. Weighing the threat of stopping support for deprecated items will force those developers to buck up their acts and read standards beyond HTML 3.2.

    A request:

    Fix the damned ECMAScript debug alert! It would be great to know exactly what is wrong with some code, instead of trying to guess what the error is.

  237. Anonymous says:

    That’s all great except it’s too bad that all of those enhancements for web developers aren’t going to be useful until more of the world is on XP and can use IE7. I see that as still years away. It would be nice to see some sort of update to IE 6 that addresses these issues even if all of the security enhancements can’t be retrofitted to it. I’m still going to be developing for an IE6 world no matter how great IE7 may be because too many people won’t have a reason to go to XP until they have to.

  238. Anonymous says:

    Regarding web sites which aren’t coded to exact standards (which run in quirks mode):

    Sorry if I have missed your commenting on this, but I recall a while ago a post stated that any CSS alteration would only be applied to standards mode, has this been the case?

    For example, if my web site was written with all these hacks and in quirks mode, will that web site appear the same in IE 7’s quirks mode … if in fact quirks and standards mode still exist in IE 7?

  239. Anonymous says:

    "I want to be clear that our intent is to build a platform that fully complies with the appropriate web standards, in particular CSS 2"

    Brilliant – that’s just what I and a million others wanted to hear 🙂

    Let’s hope your intent is matched by your product!


  240. Anonymous says:

    I don’t think I will ever install MSIE7 if proper XHTML support (including proper handling of application/xhtml+xml MIME type of course) and proper CSS2.1 support, including generated content (:before, :after, content, counter, url etc.) and selectors will still be missing.

  241. Anonymous says:

    Regarding the * html hack, I’d like to see it dropped in IE7. If IE7 really has fixed the massive CSS problems that IE6 had, then it also shouldn’t read the hacks some of us have had to use for IE6. In other words, if IE7 gets things right the first time, it could be a disaster if it also reads the incorrect values we’d been feeding IE6 through such hacks.

    I’m looking forward to a lot of these bug fixes. It looks like IE may finally be catching up with the times, which should really make things easier for developers. Just as long as it isn’t another five years before the next release…

  242. Anonymous says:

    This sounds promising ;c)

    About the * html hack : if IE7 is to comply with web standards, let it comply fully ;c)

    And that way hacks already in use shouldn’t break layout in IE7 (and not need a rewrite, not always that easy on production sites that have been handed over to the client).

    For ex. if min-height & * html work in IE 7, the background on the following box may not display properly if the content is higher than

    325px :

    #wrapper {min-height : 325px; height : auto; background : url(i/backgrounds/home_blooming_tree.jpg) no-repeat}

    * html #wrapper {height : 325px}

    Get ride of * html and all is fine for IE<6, IE7 & other compliant brothers

    Keep * html and things will brake.

  243. Anonymous says:

    Thanks for clarifying what we can expect. This type of post helps developers/designers the most. Since you solicted feedback on specific features, I’ll ask this: What will you _remove_ from IE7?

    I would like to see cursor:hand deprecated in IE7. This is IE-only and not a CSS standard. If someone (or a third-party app) uses cursor:hand they may believe that it "works" in other browsers but it won’t — and they’ll blame the other browsers for not supporting it. What should be used is cursor:pointer, which you also currently support.

    I can understand if you wish to provide backward compatitibilty for non-IE7 DOCTYPES but I would request that the IE7 DOCTYPE drop support for cursor:hand to assist developers/designers in creating cross-browser CSS.

  244. Anonymous says:

    Last week Microsoft made the first beta release of Internet Explorer 7 available to partners and MSDN subscribers. I’ve been keeping up with Chris Wilson and the IE Dev Blog, and I’ve been doing my damnedest to give them the…

  245. Anonymous says:

    Fixing specific bugs is good, but I’d rather hear about blanket solutions to the broken box and float models.

    CSS support is one thing, how about implementing complete HTML 4.01 support, or even better, full XHTML 1.0/1.1?

  246. Anonymous says:

    XHTML support would be nice, of course–preferably including XML prologs.

    I would also like to see Internet Explorer preloading Flash content under valid HTML 4.01 Strict. See for an example of how IE 6 fails to do so.

    Support for "display: table" and such would be nice. So would full (read: better than the current version of IE or the current version of Firefox) support for "display: inline-block".

    It would be nice if IE offered some method to block unwanted toolbars such as the "Norton Antivirus" toolbar that Symantec installs without permission and for which removal is not straightforward.

    Of course, there are many (many, many . . .) other issues I would like to see addressed in IE 7, but I am not certain that enumerating them here would not be a waste of time as your goals for IE 7 seem rather low.

  247. Anonymous says:

    I’m still digging through comments, but I wanted to address one comment I saw.

    kiji – I do NOT want to give the impression that I think of the acid 2 test as "just a wishlist of a few elitist webgeeks." That is explicitly NOT what I said. I think it is a decent wishlist of functionality – I understand why nearly everything in that test is something web developers would want in practice (data url support is a notable exception, but I won’t go into it here – and the WSP has released a version of the acid 2 test without data urls). My point was that it is not intended to be a compliance test, and many comments on this blog (and, for that matter, the original press article by Håkon Lie) seemed to treat it as a litmus test of whether we care about standards. We DO care about standards support, and in particular the pain web developers experience today in working with IE; to the end of making that better, I don’t think focusing on passing the acid 2 test is the right thing for us to do first.

  248. Anonymous says:

    Oh yeah: thanks for the work you have been doing even if it does not result in Internet Explorer supporting standards as well as Gecko/Firefox and Opera and Safari (and iCab . . .). Slow progress is still progress and even if Internet Explorer 7 sucks, it will doubtless suck less than Internet Explorer 6, which admittedly, sucks slightly less than Internet Explorer 5.5 did (not enough less, in my view, to warrant a full half of a version number, but that is beside the point).

    I would very much like to see a truly great Internet Explorer (regardless of who likes or dislikes the browser or its primary distribution method, Microsoft determines the lowest common denominator for Web browsers by shipping the largest number of pre-installed Web browsers with its operating system), so I do hope you keep improving it even if you release Internet Explorer 7 without fixing all of its bugs and fully supporting the most relevant current Web standards.

  249. Anonymous says:

    I think the biggest problem occurs when you put a lot of code (or a big Array) an IE comes slow and slow. This is our nightmare when we use AJAX!!!!

    Is this fixed in IE 7?

    Thyago (

  250. Anonymous says:


    What do you mean about data: URLs? It looks useful in some circumstances. Do you see some hidden problem?

  251. Anonymous says:

    I would like to see the layers/DHTML/native controls working correctly. Right now, only IE prevents the correct rendering of DHTML higher-Z-order over lists and selection boxes. This makes it rather hard to have DHTML menus on pages that have such elements on them and still be usable by IE.

  252. Anonymous says:

    How about fully supporting HTTP1.1 WITH pipelining? Currently this is broken in IE. Full support of the RFC2616 spec for HTTP1.1 would be a huge benefit to the users who use satellite or other high latency networks. Right now mozilla supports HTTP1.1 completely and is the recommended browser for people who connect to high latency networks. If IE wants to play in the satellite ISP industry, this is an absolute must.

  253. Anonymous says:

    How about the webpages that are using the current bugs to their advantage? Will it break them?

  254. Anonymous says:

    This new CSS support sounds great! I’d love to see XForms 1.0 or 1.1 support, even as a one-click option as Firefox is doing. That would really let cross-browser XML take off, without requiring a google of JavaScript programmers!

  255. Anonymous says:

    Can we expect SVG in the IE7 roadmap at all?

  256. Anonymous says:

    Whilst it is very nice to see these long-standing bugs fixed it’s actually suprising to me that with so little increased support that this is 7.0

    It almost feels like a 6.5 release with new GUI and back-end security fixes, rather than brand new support for standards which I’d expect from a 7.0 release.

    Do you have any thoughts on this? Or am I just projecting?

  257. Anonymous says:

    Support for combo boxes would be a useful and sorely missed capability.

    And I second the earlier comment about list boxes always appearing in a layer in-front of every other layer.

  258. Anonymous says:

    Editable content without an undo capability makes it completely useless for all but the most trivial of applications.

  259. Anonymous says:

    Do you even know about the OnClose event handling bug? If a page is renedered using XSL the OnClose event does not get fired when the page is closed. It’d be good to get these basic bugs fixed.

  260. Anonymous says:

    what about DOM support?

    IE is sorely lacking in this area as well.

    And what of the select menu bug?

    select menus appear on top of other elements like css/dhtml drop down menus and there is no current way of eliminating that issue.

  261. Anonymous says:

    > Since these CSS bugs are not linked to any of

    > the security enhancements code (hopefully!) it

    > should be quite easy to backport the fixes. If

    > not, something is either very wrong with the IE

    > codebase or Microsoft wilfully refuses to fix

    > their bugs.

    And this surprises you?

    One purpose of Microsoft establishing a virutal monopoly on Web browsers was to require people to purchase an entire new operating system in order to upgrade their browser. It would defeat that purpose if Microsoft allowed customers a path to the browser upgrade that did not also involve an OS upgrade. I have no doubt that the same technique will be used to leverage XP users into buying Vista.

    Like those multi-function printer/scanner/fax/copier machines that office equipment vendors keep trying to sell me, if you want to upgrade any part of it, you have to buy a whole new package. Even if you’re perfectly happy with the printing, copying, and fax functions you have now, you have to buy them all over again to get a better scanner. This is why in a cubicle-sized office I allocate space for a printer, scanner, and actual physical fax machine.

    You’re talking about the good of the customers, the website developers, and the Web as a whole. Microsoft doesn’t have the slightest trace of concern for any of that; their sole concern is for the profits of their stockholders. This is not the place to debate the rightness or wrongness of that, or of business ethics as a whole. Just accepting it as a given will save much time and speculation about what MS will or will not do. If it makes the stock price go up, MS will do it. Forcing customers to buy a new OS to get a browser patch will increase profits, which will increase the stock price; therefore, MS will do it.

  262. Anonymous says:

    First off, I’ll admit that I am no longer a windows user and never will be again, just to give what I plan to say some context. There was a time when I liked ie better than other browsers, but not now.

    I am *very* pleased to see IE7 starting to catch up with everyone else, when it comes to standards support. If only common cross-platform standards were genuinely supported in everything MS did.

    Keep it up!

  263. Anonymous says:

    i hope you wont implement the !important tag so we can always override your bunk settings with out own.

  264. Anonymous says:

    Just wanted to clarify my previous comment, as I lumped two questions in one –

    Is there any chance that there will be support in IE7 for the application/xml+xslt media type?

    Supporting it just means that you would consider it to be an alias for "text/xsl", which is a Microsoft-invented media type that ended up not being standardized. It’s not a showstopper, of course, but it does unnecessarily complicate the use of ‘xml-stylesheet’ processing instructions in XML docs that are used on different platforms.

    This has nothing to do with XHTML, DOM, or otherwise. I don’t think there would be any harm or complications in implementing it.

  265. Anonymous says:

    Please excuse my bad English, it is not my first language …

    In _my_ opinion, conditional comments are not a better way than exploiting errors in the CSS selectors (such as ‘* html’).

    First of all, a comment is a comment – it’s content is not to be interpreted as anything else but a comment. Thus, the whole idea of conditional comments is exploiting an error in IE – even if it is a documented error. Thus it is not better than using selector syntax errors.

    Second, the conditional comments can only be placed in the document, not in the CSS (or does /*[if IE lt 7]*/ … /*[endif]*/ work in CSS files?. Thus, for adding CSS specific to a new IE version, you have to alter the _documents_ instead of the CSS files. Changes to documents should only be necessary if the content of the document or the structure of the document change – not if a new IE is released.

    Third, it spreads the CSS over many files: either you put style elements in the Conditional Comments for each IE version in every document, or you link different CSS files for each IE version – in both cases the CSS code is spread over a lot of files instead of just one when using CSS selector hacks like ‘* html’…


    One wish I have (as I have not yet seen an WinXP system as stable as the Win2K systems I see – usual up time for the Win XP systems I know is about 3 days, for the Win2K systems several weeks) is that IE 7 will be available for Win2K too, not just for WinXP …

  266. Anonymous says:

    Well, if you’re still looking for a list of things in IE that could be improved, try this:

    It’s a good list of the things IE does wrong that tend to get on designer’s nerves, other than bugs. Some of it is on the list already, and some of it has been requested, but some if it also hasn’t been mentioned (most of the JScript problems).

  267. Anonymous says:

    To Simon Wilson: Here are some really interesting IE Win css tests documented, a must see. The links are included.

  268. Anonymous says:

    Links from the desktop: A lot of what I read online affects me, but I don’t want to flood the aggregators with lots of items. Here’s some stuff that I want to get into this weblog’s database, and in the…

  269. Anonymous says:

    You’re fixing all of the worst incompatibilities? And actively attempting to ship a standards-compliant browser?

    Who are you and what have you done with the real Microsoft?!?

  270. Anonymous says:

    Thank you! This is the first time I’ve ever seen anyone from MS commit to standards compliance. Is there somewhere we can report specific bugs to you in the hopes of getting them fixed?

  271. Anonymous says:

    Its about time you start integrating the CSS standards. IE is a nightmare when developing web apps.

    So finally after years of developers bitching about IE you’ve finally got it into your head that they are bitching for a reason.

    I will be alot happier when i actually see it. I gotta see it to believe it. 😀

  272. Anonymous says:

    This was never much of an issue in the past, but the advent of Wikipedia and other international applications has resulted in an increased demand for supporting the display of characters from multiple scripts (writing systems) in a single document. Other browsers are surpassing IE’s ability to deal with multilingual content. The basic idea is that if a web page that is mostly a Latin-based script also contains a bit of Greek, as does, then the browser should do its best to use Greek glyphs, regardless of what font is preferred for Latin-based scripts. But IE6 (on typical North American Windows XP setups, at least), regardless of font settings, is not able to render all of the characters on that page, and it seems no amount of browser configuration or tweaking of the HTML will remedy it. It’s particularly strange because the characters that are showing up unrenderable *are* available in the default fonts; you can load up the text in Notepad with, say, Verdana, just fine. Only in IE is there a problem with it.

    Since IE5 there has been a "font linking" technology in place that supposedly helps with this somewhat, but for reasons no one quite understands, it does not work very well, even when many fonts are available and HTML lang attributes are used. Font linking was never very well documented, and seems to require scripting an Mlang object and/or tweaking some registry settings in order to associate a font with its fallbacks, so it is not really automatic and I don’t think anyone has yet figured out how to make it work on the web.

    Is there any chance that IE7 will show any improvements in font linking? It would aid in the development of content for international audiences greatly.

  273. Anonymous says:

    I haven’t read all the comments, so I don’t know if this has been suggested already – I hope so… It’s good to see that Microsoft are finally listening to developers – the deficiencies of IE has cost my clients about 30% more than simply designing to W3C standards.

    My question is: since the Gecko rendering engine (along with the KHTML engine employed by both Konqueror on Linux and Safari on OS X) are open source, and available under very favourable licenses, why is Microsoft insisting on releasing an inferior rendering engine?

    Why not work with the community instead of against it?

    Doing otherwise will continue to embarrass the world’s most profitable monopoly and frustrate the whole universe of developers seeking to adhere to W3C web standards?

  274. Anonymous says:

    It’s good to hear that Microsoft has finally decided to pay attention to the webdeveloper community after making so much noise about only being concerned with security issues and the user end of the browser.

    About the "* html," I will take the position that removing "* html" will cause less overall breakage of webpages then keeping it now that Microsoft is committed to correcting many of the problems that forced webdevelopers to use "* html" in the first place. If it is still the IE Development’s teams goal to produce a browser that better supports the W3C specifications for CSS without creating havoc for webdevelopers, then removing "* html" becomes a necessity. From now on, webdevelopers can use conditional comments to work around any IE7 problems that we encounter.

    Now there is one thing that I am concerned about, that is which Windows versions IE7 will support. Windows XP SP2 doesn’t go back far enough. The issue I have here is that webdevelopers should be free to recommend visitors to upgrade their version of IE to version 7 without having to tell them that they must also update their OS. This is why support going all the way back to Win98SE should be desirable, Win2K at the very minimum. The idea here is to get users of older versions of IE to adopt IE7 quickly with the minimum of fuss. Otherwise, you are opening the door and inviting a competitor to fill in the gap you have left behind.

  275. Anonymous says:

    I think the biggest CSS2 issue that needs to be added/fixed is full support for the display attribute. Currently "display: table", "display: table-cell", and other similar properties do not work at all. This presents a MAJOR PROBLEM when working with XML with CSS as you cannot display anything in a table format without using XSLT. I personally I find this to be the biggest issue with IE.

  276. Anonymous says:

    As long as the BOX MODEL gets fixed I’ll be happy.

    The only advice I could give is to take your time to release it and make a good standard compliant program instead of patching it a thousand times. This way we won’t end up with a mess for each and every different version.

    I’m one of those who hate IE but I can change, as long as IE proves me wrong.

  277. Anonymous says:

    Regarding PNG Alpha support;

    Please bear in mind many images don’t need RGBA 32bit, and that PNG supports paletised alpha images.

    A random image with alpha channel from a site I work on is 8k as a 32-bit PNG, but only 3.2k if 32 colours are used, with no loss of quality. A significant saving.

    IE6 does not support paletised alphas, even with the DirectX/CSS hack.

  278. Anonymous says:

    IEBLOG reports that IE7 will fix a number of old css bugs. one person working hard to achieve this is Molly. weird thing is she’s been criticized for it. oughtn’t everyone be jumping with joy that that darned browser finally…

  279. ieblog says:

    Dave Lane,

    We don’t believe that we have the inferior rendering engine. We have a technology deployed by fortune 500 companies for line of business applications and used by consumers all over the world. We plan to continue to develop and improve our rendering engine and the browser as a whole.

    Microsoft isn’t going to rebuild IE on open source software. I think every here knows that and no one realistically expects us to do so. We actually like our technology and think that it has a lot of benefits for our customers.

    Al Billings [MSFT]

  280. Anonymous says:

    I am one of those who switched to Firefox and I don’t plan on going back, but if the next release of IE is at least as the equal to Firefox, then I will stop bitching (not that I bitch publicly, much, but I’ll stop kvetching and grinding my teeth behind your back).

    Please, save me the dental bills.

  281. Anonymous says:

    @ IE team

    Thank you, thank you, thank you! Finally!

    I don’t want to sound whiny but this doesn’t solve anything if Win98 and Win2000 users (oh yeah, and WinMe) are stuck with a broken browser! As in, I cannot ignore them, so I’ll still have to hack for IE! Not fun, and that’s an understatement! However, this is a great first step. Thank you.

    PS I promise to stop hating IE if you release a huge update (CSS, PNG, etc) for Win9.x/Win2000 users, preferably IE7 without the security updates. Promise. I’d even stop pushing the Fox on my friends 😉

    PPS Will palette-based PNG alpha work? I sure hope so.

    PPPS Oh yeah, update the HTTP accept header, and accept application/xml+xhtml. Thanks

    @ Chris Wilson

    If you fixed IE, fix (drop) "* html".


    <b>* html was used to write code for IE’s bad CSS engine ONLY. Standards-compliant browsers never see it. If IE7 becomes largely standards-compliant, then it in effect becomes Gecko, per se, and SHOULD NOT be subjected to the BAD CSS delivered to prior versions of IE</b>.

    Also, from my experience, fixing IE but still allowing * html would break a lot of my CSS work. Throwing out * html would not negatively affect the CSS on my websites.

    Thank you for asking, Chris.

    About CSS support, <b>I’ve noticed that foo a:link { color: inherit; } does not work.</b> It would’ve been really handy in at least one instance. So please fix "inherit".

  282. Anonymous says:

    The Acid test IS pretty harsh, but it does address many concerns that us web developers have. And it is a milestone to hopefully hit one day. The thing is why not hit those standards of compliance? Sure it’s going to take the IE dev team longer to do it, but think of how many hours of researching and bug-fixing and css hacking WE do to fix problems in IE. And not even that, how about just for the sake of saving face. I mean what web developer that uses css doesn’t have issues and anger towards IE and the endless hours of hacking for it. I am glad IE is finally coming around a little bit, however, I’m not sure if it’s enough. Just keep workin guys – we are 😉

  283. Anonymous says:

    Great news, glad to hear it.

    Was worried by initial feedback on CSS support in Beta 1 but this has put my fears to rest.

    Good work guys!

  284. Anonymous says:

    I?m very happy that we?ve shipped IE 7 beta 1. I wanted to make it clear that we know Beta 1 makes l. . .

  285. Anonymous says:

    After seeing this blog entry referenced in Slashdot, I am pleased that active steps are being taken to make IE7 (shipping with the new Microsoft OS Vista, aka Longhorn) more CSS compliant. With IE7’s catching up in the features game…

  286. Anonymous says:

    Why do you say you’ve fixed things that you haven’t fixed? The :hover pseudo-class is apparently supported for all elements, which its not, and child and adjacent selectors are apparently supported, which they’re not. What’s going on?

  287. Anonymous says:

    I’m not sure if this request has been mentioned, but this would be really useful.

    With CSS, I’d like to do something like this



    top: 100px;

    bottom: 0px;

    left: 100px;

    right: 0px;


    Logically, this CSS *should* make a box that’s 100px from the top of the page, and 100px from the far left of the page. The box itself should expand as the right / bottom part of the browser is resized.

    This doesn’t work in IE, but does in Firefox. Please address this, if possible 🙂

    If this bug is fixed already, thank you very much. If it’s a duplicate of another bug, sorry for being redundant.

  288. Anonymous says:

    Finally: Internet Explorer version 7 will allow you to
    use the CSS :hover pseudo-attribute on all elements.
    This is a notorious IE deficiency. If I, as a web designer,
    want to make something happen when you hover your mouse over,
    say, a graphic, for any b

  289. Anonymous says:

    Finally: Internet Explorer version 7 will allow you to
    use the CSS :hover pseudo-attribute on all elements.
    This is a notorious IE deficiency. If I, as a web designer,
    want to make something happen when you hover your mouse over,
    say, a graphic, for any b

  290. Anonymous says:

    Al Billings,

    I hate to break it to you and Microsoft, but you do have the "inferior" browser. Inferiority is based how well it can do its primary purpose, the downloading and rendering of webpages from the Internet, and its ease of use. And frankly, the other browsers are beat the pants off of IE in both areas. It’s this kind of hubris that infuriates people when the folks at Microsoft are not willing to admit the obvious.

  291. Anonymous says:

    I’m sitting here as I write this in the Apache Testing tutorial and the speaker is saying that IE Digest authentication is broken because it doesn’t follow the RFC and doesn’t include the query string in the Digest material…

    Are you planning on fixing these sorts of bugs as well?


  292. Anonymous says:


    In one of your comment posts you mentioned problems with data urls? I’m curious: What are these problems?

    I think data urls are great for adding small items like bullets, image buttons etc. to web pages without having the overhead of doing many HTTP requests – especially when delivering content over "expensive" HTTPS connections and with large RTTs.

  293. Anonymous says:

    @IE team, Chris:

    I forgot.

    If 99% of the problems with IE are fixed — CSS, PNG, positioning, etc — then please drop <?xml> quirks mode switching. i.e. now all XHTML pages will use Strict rendering.

    You’ll be just as good as other modern web browsers then.

    The benefits are huge:

    Nothing will be new to web designers, who have been (or should have been) designing to accomodate Safari for Mac users and Gecko for the 12% of the Net that uses Firefox, and Opera for those who prefer that.

    If any lazy web developers have been IGNORING the standards and Safari/Mozilla/Opera, this will force them to stop relying on incorrect (quirky) behavior (good for the ‘net).

    Faultily formed HTML documents (legacy documents) can still display as they would in IE6.

    Pages which were marked as quirksmode for IE6 would now benefit from a more powerful, correct IE7 AND still retain their IE6 quirkiness in IE6.

    Some web developers don’t know about DOCTYPE quirks mode switching. I didn’t for awhile. This will spare them the agony of trying to figure out why their standards-based XHTML doc won’t display correctly in a standards-compliant IE7 browser (if thy have the DOCTYPE without understanding its consequences, as I did for a time)!

    Thank you for your time.

  294. Anonymous says:

    I will beleive this when I see it. It’s fine and dandy to make a bunch of promises, but with the new WGA scheme I see less users making the move to IE 7, and frankly another IE browser that supports yet ANOTHER degree of standards just makes the need for "hack-on/hack-off" a bigger issue.

    When you guys figure out how to make IE lightweight and offer IE7 to the general windows using public (yes many people surfing the intarweb are still on Win98) then maybe we can get excited. As long as I still have to backwards support IE5.x, I am not going to offer any high fives.

    Fixing IE is one step. Getting older IE browser users to update is the other. Make it easy, make IE relevant (and available) again.

  295. Anonymous says:

    Could you explain how IE internally handles CSS?

    Is there a function/method that I can call to show the actual code of the "Presentation (CSS/XSLT)" stage, like document.documentElement.outerHTML for the "Content (XHTML/XML)" stage?

    Thanks, Chan.

  296. Anonymous says:

    This my seem like an odd request but I want better support for icons in web pages. See link.

  297. Anonymous says:

    &lt;p style=&quot;font-style: italic;&quot;&gt;Just nu r jag vldigt glad. Det r just denna nyhet jag har vntat p de senaste ren…&lt;/p&gt;&lt;p&gt;Nr Internet Explorer 7 kommer ut s kommer mnga av deras designbuggar att vara borta. Det innebr att man ntligen kommer att

  298. Anonymous says:

    As a Web author my main concern is with Web standards *not* security. IE is part of the shell so its Windows that must be updated.

    I had to set up a Win98 machine last week (and a 2000 machine yesterday) and was so sad I had to use IE6. They’d use firefox if they didn’t insist on using that stupid SBC Yahoo! browser. People will use their machines until they no longer work, don’t care about Web standards.

  299. Anonymous says:

    Awesome! Finally some real words on CSS standards support. Your goals are admirable – I hope you achieve them and more!

    I really hope that the IE7 rendering engine will be nicely wrapped as a .NET control… better than MSHTML and the like, maybe with hooks to embed browser windows into WinForms and pick up what link was clicked in the page cleanly from the WinForm??? 🙂

  300. Anonymous says:

    Lachlan Hunt:

    For the record: different Barry.

    However, I disagree with your point. Conditional comments are available already in IE, and it’s IE i’m talking about here, i don’t expect this to form part of a standard, a standard’s job is to describe the best possible case, striving towards all browsers rendering (almost) identically*. Where browser-specific functions can come in handy, is with overcoming their shortcomings. As conditional comments in HTML (and embedded script) are already available, it’s becoming almost standard practice to include an extra stylesheet for IE hidden withing conditional comments in order to overcome it’s flaws/differences. What I am asking for here is a way to do that in a single CSS file, in a clean manner.

    The alternative is, of course, to get IE’s CSS support to follow the standards completely and accurately. But it doesn’t look like this is going to happen soon, so this would be:

    A) An acknowledgement that browsers ARE different, and that IE isn’t the only browser out there

    B) A practical tool to help us make our pages work in IE quickly and efficiently

    C) A ‘gift’ that will be loved by the css/validation-zealots

    * Yes, I know, in an ideal world we should make pages that are flexible enough to look good with whatever level of support is available, but years of table-based design have lead clients into expecting pixel-perfect webpages… Point is, that’s another topic for another day 😉

  301. Anonymous says:

    Thanks for this. It is much appreciated.

  302. Anonymous says:

    Please consider rolling these bugfixes out in the form of a patch for IE6! It would make a <strong>huge difference</strong>.

    Well done to the new IE team – you’re on the right track.

  303. Anonymous says:

    Hey Al or Chris…

    Why didn’t you guys listen to us for IE6?

  304. Anonymous says:

    Potential Bugs/Feature Requests

    Bug (I think): When disconnecting from a VPN that doesn’t allow me to contact my intranet proxy server IE requires a restart in order to get it to contact sites again. I didn’t have to restart IE 6 after disconnecting from a vpn, it would just go back to working again.

    Feature Request: Confirmation before quit if more than one tab open – by default I click the X when im done with a window and I forget about tabs 🙂

    Feature Request: Save open tabs as favorite folder/open favorite folder as tabs

    Feature Request: Modifier + Click for open in new tab like ctrl-click opens in new window.

    Thats all that i’ve arrived with so far…

    Gareth Evans

  305. Anonymous says:

    This has probably been mentioned already, but a MAJOR pain in the ass missing from that list is the z-index problems with select boxes.

    Please help me regain my sanity by fixing this one.

  306. Anonymous says:

    Ah, good. It’s nice to know that you are working on the bang-your-head-on-the-desk problems, as my forehead is looking a little flat these days.

  307. Anonymous says:

    Fiery Kitsune,

    >Hey Al or Chris…

    >Why didn’t you guys listen to us for IE6?

    Can you be more specific?

  308. ieblog says:

    Fiery Kitsune,

    Since you are addressing me, I’ll respond personally. I didn’t work on IE6. 🙂

    I worked on IE4 and IE5 (and the variations of both such as dot releases). I worked on IE6 for XPSP2, which is the closest that might qualify but my test team was focusing on shipping the Information Bar, Popup Blocker and a few other things. I didn’t work on the IE6 original release, though Chris did. I was off at other projects at Microsoft.

    Al Billings [MSFT]

  309. Anonymous says:

    For Al Billings [MSFT]:

    I’d appreciate it if you could explain to me how a broken rendering engine (IE) is preferable to one that works (Gecko,KHTML,Opera’s engine, etc.).

    Perhaps you can also explain how a rendering engine can be anything less than inferior if it doesn’t do what a rendering engine is supposed to do: efficiently turn W3C standards compliant content into consistently/predictably presented web pages.

    In what way does the fact that fortune 500 companies deploy IE support your assertion that MSFT’s rendering engine is superior? An equally valid explanation for their use of IE is that it’s another example of MSFT leveraging its criminal monopoly with impunity.

    You say "We plan to continue to develop and improve our rendering engine and the browser as a whole." To what end? If W3C standards compliance isn’t the first and foremost goal, what is? Will IE7 be <a href="">ACID2</a&gt; when it’s officially released like a quickly growing number of open source browsers? If not, then what value could IE7 possibly offer MSFT’s customers??

    As for your comment "Microsoft isn’t going to rebuild IE on open source software. I think every here knows that and no one realistically expects us to do so." that simply reflects the attitude which will eventually bring about MSFT’s demise – the bottom line is that the open source community is producing better code at a faster rate than MSFT – how can any betting person possibly put their money on MSFT in the long run? (Unless, of course, MSFT manages to make open source illegal by exerting the kind of influence on the US government that got it let off its criminal monopoly conviction scott-free… Of course, even that would only slow things down in the US and perhaps its puppet nations. The more savvy, hungrier nations who don’t kowtow to the US will blow right past)

    And for this one: "We actually like our technology and think that it has a lot of benefits for our customers." perhaps you could enumerate some of those benefits – or are you really talking about non-standard features that serve no purpose other than accomplishing lock-in…?


    Dave Lane

  310. Anonymous says:

    I am very pleased to see the list of bugs that will be fixed in IE7. This is a great start, but I hope we don’t have to wait another 4 years for the next big update.

    As far as things that need fixing, the one thing I haven’t seen mentioned is correcting the "dotted" attribute for the BORDER property. IE currently displays "border: 1px dotted black" as a dashed line.


  311. Anonymous says:

    Whatever IE is doing, it is just playing catch up. Whatever is being revolutionised on the web is not happening thanks to IE. The W3, Mozilla, and many other organisations promoting open source and web standards are paving the way.

    You’ve really got out of touch with the web guys. Your only saving grace is your user base. If you were a smaller organisation like Mozilla, and you released this, you know it would not be as popular.

  312. Anonymous says:


    One very very simple bug to fix, one that would help us a lot, would be to allow text before the DOCTYPE declaration while retaining "standards mode". This is especially important for allowing the XML declaration.

    Plenty of people think IE6’s box model is non-standard, for the simple reason that they include an XML declaration before the DOCTYPE, IE goes into "quirks mode", and the IE5 box model is used.

    I did standards-based web development for three years until someone told me the box model was fixed in IE6, and all I needed to do was remove the XML declaration…

    Maybe you’ve fixed it allready. In that case: cheers!

  313. Anonymous says:

    the sad irony is that this doesn’t really matter, when conjoined with the fact that IE7 will only be available to users with legitimate copies of WinXP.. out of the hundreds of machines i’ve personally worked on in the last year, i’d easily say 15-30% are swiped copies..

    (sure, part of that is the higher than usual density of college students around here.. but still.) and i highly doubt that more than a quarter of them will "upgrade", even with the new WGA program.

    That means that web developers can no longer recommend with much of a degree of certainty ‘just upgrade’.. unless of course they mean upgrade to firefox.

    so between the fact that a) people don’t know they can or really care enough to upgrade, b) a significant chunk won’t be able to, and c) i’d bet on an uptake curve of Vista even shallower than that of WinXP, I don’t really think this is going to have much of a dent in browser usage statistics or easing of developer pains.

    Which isn’t to undermine your work… I’m quite glad it’s happening, and that you’re sharing the results with us.. It just won’t make much of a dent for at least several years.

  314. Anonymous says:

    nimdoc and all the others talking about the XML declaration and quirks mode: You are out of spec.

    According to RFC 2854, the only form of XHTML you are permitted to send as text/html is XHTML 1.0 documents following Appendix C.

    The first thing Appendix C tells you is that you need to avoid the XML declaration.

    If you are including the XML declaration in documents that you are sending to Internet Explorer, then you are writing non-standard code, and Internet Explorer is doing the right thing by assuming that it’s buggy and using quirks mode.

    If you want Internet Explorer to follow the specifications, then follow the specifications yourself. You can’t deviate from spec and then complain about Internet Explorer doing so too.

    PS: IE developers – the new colour scheme and layout is nice, but stop messing with the font size! I have a perfectly good font size set in my browser already, and you are making it less readable by shrinking the text.

  315. Anonymous says:

    Hyatt pulled off Acid 2 in a coupla weeks, and all of the King’s men at IE can’t do it. He even documented the progress through the process. Goes to show doesn’t it.

    Also, when I click "Comments", I have to scroll all the way down to add a comment. The new comment textarea should appear before the comments already placed, not after.

  316. Anonymous says:

    This is all good news. Fewer headaches.

    Sad that the decision was made years ago to

    bake IE into the OS. If instead IE were a

    regular app, these fixes would work for

    Win 98/ME/2K/XP SP-1. Still sad/funny that

    the baking was announced as a technical

    goodness, when it was clearly technically


    Anyways, good moves. Keep going in the

    right direction. MS will win if they

    help developers win, which means increasing

    standards compliance.

    — stan

  317. Anonymous says:

    IE needs to follow the standards. It’s the only way to catch Mozilla/Firefox and keep developers and users happy.

    I really hope Microsoft doesn’t dis-ban the IE team after the release.

    Right now, I only use IE when I have to and that is happening less and less.

  318. Anonymous says:

    Jim wrote: "nimdoc and all the others talking about the XML declaration and quirks mode: You are out of spec."

    I don’t think so. If you read RFC 2854 (, it says in section 5: "XHTML documents (optionally) start with an XML declaration which begins with "<?xml" and are required to have a DOCTYPE declaration "<!DOCTYPE html"." So it mentions that the XML declaration is optional, which is fine.

    Now if you look at appendix C of the XHTML 1.0 specification ( you will first see two things: this appendix is informative, not normative and it presents compatibility guidelines for rendering XHTML in old HTML browsers. Section C.1 explains that you may want to avoid the XML declaration in legacy browsers that are rendering text found before the HTML header (are such browsers still in use today?) or interpreting such documents as unrecognized XML rather than HTML. Neither of these applies to IE6, which incorrectly switches to quirks mode instead of displaying the XML code.

    We should also balance that with the fact that the XML 1.0 specification states: "XML documents SHOULD begin with an XML declaration which specifies the version of XML being used"

    So even according to the IETF and W3C standards mentioned by Jim, there is no good reason for IE to switch to quirks mode when the XML declaration is present.

  319. Anonymous says:

    I understand that prioritisation is essential and presume from the silence that SVG doesn’t yet make that cut, but I just wanted to echo the other people here asking for SVG support in the browser.

    Support of a vector graphics standard is not the sort of thing that end users will spontaneously clamour for, but it’s too important and fundamental a technology to be left to a third party plug-in that users may not choose to – or be permitted to – install. If there’s any doubt about that, surely all that’s needed is to make the comparison with <a href="">Avalon/Windows Presentation Foundation</a>?

  320. Anonymous says:

    I see via that IE 7 Beta 1 is out, and support for CSS 2.1 is in the works. For comparison, see what Deer Park Alpha 2 (the next version of Firefox) includes. It looks like IE 7 is…

  321. Anonymous says:

    If all CSS errors that are currently dealt with by:

    * html div { property: IE/Mac value; /* */property: IE 5.x value; property: IE6 value;/* */ }

    are fixed then that hack can be ignored. And about time too. Also about time too is ABBR.

    Q with quotes would be nice (in the HTML4 spec so not exactly a recent addition), as well as CSS quote definitions so we can change them (along with content: so we can assign them).

  322. Anonymous says:

    ECMAScript prototypes is still lacking, this would be a nice way to extend the the toolboxes of clientside-scripting people. And it is supported by all other browsers already… I’m really happy about the mime-type for xhtml. Keep up the good work…

  323. NikolayMetchev says:

    There is a nasty minimized script tag bug which is explained in more detail here:

    Please fix it. It is very difficult to know what has gone wrong when internet explorer presents you with a blank screen.

  324. Anonymous says:

    I understand that your work on IE is very difficult. It is very complex application with many more problems that I even cannot imagine.

    I know, guys, that you have difficult position when developing IE but believe me – so do I when designing pages for IE.

    I hope that in the future I’ll be able to say: "I created page according to standards and I don’t need to spent any time by checking it against widely used browsers, because I’m sure it will work!" That’s my professional dream!

  325. Anonymous says:

    The latest CSS support sounds very encouraging!

    Since there are currently over 320 comments, I’m going to admit that I don’t have time to read through all of them (hopefully you do); hopefully, I’m not being redundant. I just wanted to throw in my suport for keeping the IE underscrore hack in IE7 (and not removing it in any service pack upgrades to version 6).

    While there are certainly deficiencies in IE’s current CSS rendering, I find that using the IE underscore hack (_margin: 10px;) is very helpful in creating a unified cross-browser look. This hack will be especially useful since you will be improving first-child selectors, which many of us have used to fix rendering bugs.

    There will most likely always be subtle rendering differences; the underscore hack is a farily simple means of fixing these issues.


  326. Anonymous says:

    Although the bugs that are/will be fixed in IE7 are very welcome, I don’t think it will make any difference because the percentage of users that will be using IE7 in 3 years from now will be very small since only Win XP SP2 can be updated to IE7. So, if we can’t tell people to "upgrade to IE7" without them having to upgrade their OS, it’s all pointless. Therefore, at least the CSS bugs should be made available as a patch for older IE versions.

  327. Anonymous says:

    re: XP / Vista only… i have no bones with this, (an IE6 renderer upgrade would be nice though) after all when ford brings out a new model everyone doesn’t bitch and yell that they arent releasing an "upgrade" for the rest of the ford owners to get the nice shiney sat nav unit, do they?

    If MS is to construct a less buggy more secure OS then it needs to wipe the slate clean with Vista and have zero quirks for the sake of backwards compatibility.

    It’s a no win situation. they make it backwards compatible then everyone complains about the bugs, miriads of config options and slow execution of software. make it brand new and everyone complains about the lack of support for the old bugs and options!

  328. Anonymous says:

    As my title specifies, IE7 isn’t going to pass the Acid2 Test that a few browsers have now been tweaked to pass correctly. This isn’t all bad though, since a member of the IE team has posted on the IE…

  329. Anonymous says:

    I add my voice to the requests for application/xhtml+xml support, but more importantly, I would be delighted to hear a definitive answer one way or the other.

  330. Anonymous says:

    Though it has been said before, I want to add my voice to the chorus of ‘no hacks’: IE7 should not support things like ‘* html’ or ‘_property: value;’.


    – Many CSS designers have used them. If you keep the hacks, but fix the actual bugs they are used to work around, you might break many designs that would work if rendered with standards-compliant CSS. If IE7’s CSS support is good (and the list of fixes suggests this), then if IE7-specific tweaks need to be made, a designer will want to make them relative to the standards-compliant CSS, not the IE6-hacked CSS.

    – You already have a back- and forwards and compatible and officially supported method for designers to use: conditional comments. I’ll admit I have used hacks before, and IMO with good reason: IE5/6 required tons of fixes, so using hacks allowed me to keep the related standards-compliant and IE-specific CSS close together in the stylesheet. But if we don’t require obscene amounts of IE-specific CSS anymore for IE7, this excuse is pretty much lost.

    – Companies want to make their site IE7-compatible quickly and without paying a designer for days on end. It might not even be the same person who did the original design: forcing them to dig through rule after rule of IE6-hacks only slows down the ‘healing’.

    – In the end, they’re still bugs.

  331. Anonymous says:

    To Al Billings [MSFT]:

    – quote

    "We fully recognize that IE is behind the game today in CSS support."

    – end quote

    You most certainly do have an inferior rendering engine, what in the wide world do you think this whole blog is about!!!!????

  332. Anonymous says:

    I’m very curious on the next IE. For me the support of XForms 1.0 / 1.1 would be very helpful. The support of SVG is not with priority, it’s pluggable.

  333. Anonymous says:

    Chris W., this is off topic but we need to know where we can report real bugs in IE 7 (not feature requests).

    We are a plug-in vendor and we noticed that some behaviour in IE 7 is breaking some of our functionality. We are not sure if this is a bug or an intentional change in behaviour.

    In IE 6, the ActiveX container called both OnDragEnter and OnDragOver events when a mouse was clicked. In IE 7, OnDragEnter event gets called and only if the mouse button is held down longer does the OnDragOver event gets called.

    We need to know if this behaviour is a bug or an intentional change.


  334. Anonymous says:

    Chris W, what I meant was this: Why wasn’t there any community influence in the quality of IE6 during its development?

    Also, did this blog get a makeover yesterday?

  335. Anonymous says:

    What about supporting pseudo-classes for elements other than <a>? e.g. Firefox supports td:hover as well as a:hover, but IE doesn’t and you have to use Javascript to work around this.

    I would also like to see the DOM implemented correctly please. But well done for all the hard work.

  336. Anonymous says:

    This blog translated:

    "We have never and never will implement standards competently (we’re Microsoft… we don’t have to… we make our own standards), but we are striving to implement them less incompetently. Thanks for dealing with (though you have little choice) our quasi-competence."

    I’ve been in the MS camp for a long time when it comes to web dev, since pre-ASP v. 1.0, but enough already. Just implement a damn standard already, lest more of us will move onward to the standards-compliant (and, ironically, poorer) technology shops.

  337. ieblog says:

    Mike, you wrote:

    > I downloaded the IE7 Beta from my MSDN

    > account and I want to submit a few bugs I

    > found. I couldn’t find a way to do it

    > through MSDN. Is there another way to

    > submit bugs and be able to track them like

    > I normally do on betaplace?

    The page where you downloaded it on MSDN has links and information for reporting bugs. Please report them there.

    Thank you,

    Al Billings [MSFT]

  338. Anonymous says:

    Al Billings [MSFT], I was not able to find any links to report bugs either on the MSDN IE 7 download page nor in the release notes document en_IE7_B1_4_XPSP2_RelNotes.txt.


  339. Anonymous says:

    I don’t know if the following is "on the radar"

    for IE7, but I’ll mention it here for posterity:

    Javascript Benchmark – w3c DOM vs. innerHTML

    I have personally hit some severe performance

    penalties in IE6sp1 when manipulating large

    HTML tables (hundreds of rows) via Javascript and DOM references.

    The same exact Javascript code is several orders of magnitude faster in Mozilla 1.7.10 …

    To be fair, I don’t know if this is an IE problem or a JScript problem…

    Just thought I would mention it here. 🙂

  340. Anonymous says:

    Here are 2 features I think are critical for IE7.


    It looks like this isn’t in beta 1 but planned based on "Fix :hover on all elements". I would like to see all the psuedo tags used on the <a> tag be availble for all elements.


    It looks like this isn’t in beta 1 but planned based on "Background-attachment: fixed on all elements not just body".

    One last thing. Were do we provide developer feedback?! This states that the release was intended to provide feedback, but I cannot find anywhere for proper feedback like a news group or bug log.



  341. Anonymous says:

    Достаточно интересная информация для тех, кто посмотрев на Beta 1 начал топать н

  342. Anonymous says:

    Will IE 7 be "unbundled" from the OS? It seems to me that is a primary design arbitrary, which if rectified would make all of this a whole lot easier. I can’t really say that having a browser "built in" to an OS fulfills any of my needs or wants…..

  343. Anonymous says:

    Please please please. pretty please. Fix your tables so we can have fixed headers/footers with scrolling rows <thead> and <tfoot>. Now I have to use the following <a href="; >Hack</a>. Which causes issues.. Thanks again for tackling all the rendering issues!

  344. Anonymous says:

    I was kinda suprised of the harsh words of Paul Thurrot.

    Anyone of the IE team that will comment on (t)his column?

  345. Anonymous says:

    I’m so confused.

    You explicitly said: I want to be clear that our intent is to build a platform that fully complies with the appropriate web standards, in particular CSS 2 ( 2.1, once it’s been Recommended). I think we will make a lot of progress against that in IE7 through our goal of removing the worst painful bugs that make our platform difficult to use for web developers.

    This means that IE7 will NOT be standards compliant, but instead will "make a lot of progress against that" goal?

    Why would you even bother then? How could you possibly compete in terms of quality against standards compliant browsers? Browsers that DO pass the Acid2 test? Why have you taken so many YEARS to go most of the way, only to stop just short of the goal?

    Please, some clarification: Will IE7 be standards compliant? Yes, or no.

  346. Anonymous says:

    Chris, you misunderstood what the Acid2 test page meant by "Acid2 does not guarantee conformance with any specification".

    What they’re saying is that because Acid2 doesn’t test every single aspect of the standards involved, it doesn’t confer conformance.

    Acid2 is a necessary but insufficient condition for a browser to be conformant. A browser that fully conforms to the listed standards WILL pass the Acid2 test without any problems. A browser that fails Acid2, can NOT be conformant to all of the listed standards.

    We all understand that certain bugs and parts of the standards are more important than blind conformance to the standards, but the Acid2 test represents a weighing of those things as done by a knowledgeable member of the web community. Its fine if you make different priorities and choose different areas to work on, just give us good reasons for them if they don’t match with what the users want.

  347. Anonymous says:

    Can IE just come up to date with All w3c standards?

    I mean stop doing stuff that only IE will support, dump JScript in favor of JavaScript, Use The DOM for crying out loud get rid of document.all & just go with standards so the programmers out there don’t have to dual code.

  348. Anonymous says:

    Its dissapointing that the IE team has no plans for full standards compliance, and in fact you gurantee that it won’t happen.

    I think that this is a great oppurtunity for IE7 to jump ahead and take over marketshare as well as developer mindshare. Microsoft has billions of $$, and more importantly the brightest people. You guys should be leading the way in innovation by implementing full support for things like Acid2 and Css3.

    It seems that the security issues with IE will be a thing of the past (they already are a lot better in Sp2), so the only area where it lags behind is technology. Firefox, Safari etc are all actively working towards implenting the latest standards and once again IE7 will be left behind.

  349. Anonymous says:

    Just fix that zombie quickly and quietly. Nobody is excited about IE. There is absolutelly nothing to be excitied abouot it anymore. Majority developers hate it. And If you guys do not feel as janitors cleaning up old mess and feel eager to share your work excitement with tired community that is also pathetic.

  350. Anonymous says:

    I have to say this, in all respect to the whole IE development team, you are all probably very nice people but,. You do not listen to what people want, but what you guys want, you think just because you work for the supposively largest public corperation in the world you can set your own standard, you cant! What is exciting about IE7?? PNG transparancy! Wait, all your competitors already have that. :(, Wait! A new UI! Sorry, Firefox has the same one. Tabbed Browsing! ALL of your major competitors have that.

    Security and Rendering should be your first priority. Scrap the IE rendering engine, rebuild it from top to bottom, make it standards complient and then some, take a step ahead of the competitors. Maybe use the Gecko as reference. Because, you are losing the market in IE, because you do not listen to the community, Paul Thurrott is even giving in, telling us to boycott the IE system.

    I do graphic work and code systems for people everyonce and a while, and so far, I cannot get an all CSS design to work in IE without a use of tables for positioning. IE has given so many people SO many headaches, if you guys would visit some community websites and find what WE want, you could save a failing browser.

  351. PatriotB says:

    "Sad that the decision was made years ago to

    bake IE into the OS. If instead IE were a

    regular app, these fixes would work for

    Win 98/ME/2K/XP SP-1."

    IE was "baked in" to the OS starting with Windows 98/IE4. It was "baked in" through Windows 98 SE/IE5, Windows 2000/IE5.01, Windows Me/IE 5.5, and Windows XP/IE 6.0. The baking in (integration) is itself not a problem–from IE4 through IE6, each version was made available for previous versions.

    The problem really began with XP SP2 and the IE 6 that came through that. If IE 6 SP2 would have been developed in the same means that IE 6 SP1 was developed, it could have been made available downlevel just eas easy. Someone specifically had to have made a decision that the SP2 improvements would only be available in XP. The internal build process that IE goes through was undoubetbly revamped prior to SP2. Unfortunately, it seems to be too late to revert that and go back to the way it was.

    However, that said, I can’t think of any reason that a new version of MSHTML can’t be made available for Windows 2000 at least. It would have to be a distinct version, say IE 6.5. But really, all that would accomplish is having yet *another* version of IE out there to support, since even with this, IE6 isn’t going away.

    It’s really unfortunate that the integration was "tightened" post IE6 SP1. It would be nice if someone from the IE team could comment on what really happened regarding this.

  352. Anonymous says:

    How does that exactly relate to the ducks mating rituals?

    Don’t get me wrong here, I’ve read the text carefully and I haven’t noticed even a single hint regarding ducks, not even a duck stew.

    Personally I wouldn’t care as much, but the mentioning of so many bugs clearly requires balancing with ducks (which like bugs, are quite useless).

    I’d appreciate your clarifications on the issue.



  353. Anonymous says:
  354. Anonymous says:

    I’ve been working with IE for quite a while now in developing the Echo/Echo2 web frameworks, and wanted to point out a few specific significant CSS and general rendering issues which are present in IE7 beta 1 (which are carried over from IE6). I’ve written an entry in the Echo2 project blog about these issues here:

    While this blog entry is written from the perspective of solving CSS/rendering issues that specifically affect Echo2, the issues listed will equally affect anyone attempting to develop a web-based application using CSS and dynamically modified HTML.

  355. Anonymous says:

    I have to disagree with those who say that it is pointless for Microsoft to update IE if it isn’t going to 100% adopt the W3C specifications. No browser currently supports the W3C specifications at 100%. IE is improving its support for CSS and fixing all but the most obscure bugs with the CSS that it currently support. For this and the next version, IE is in "catch up" mode with the other browsers. We should be encouraging them through this period instead of complaining about it.

    As for not passing the Acid2 test, well Firefox 1.1 isn’t going to pass it either. But we must remember that Acid2 is just one goal post, not a minimum requirement.

    As I said earlier, my biggest concern with IE is the how far back the development team will backport it. Having XP SP2 as the cutoff is going to seriously hamper IE7’s adoption.

  356. Anonymous says:

    How can you say in one paragraph that you’ve failed to make any headway with standards and in another paragraph that ease of development is a priority? This is a MAJOR contradiction. As a web developer, having a big player ignoring standards is my biggest pain.

    Come on, it’s not that hard.

  357. Anonymous says:

    I knew it. I knew not to get my hopes up. In one instance I’m happy to see some of the bugs fixed, in another I’m disappointed that we’re not seeing advancements over the norm. I feel melancholy about this new operating system more and more…

    I know that no browser can be 100%, but I see a handful of people get further along than a billion+ company. How can you not argue for open source?

    As soon as your product is ready for primetime, it will be years behind other browsers. Two years after, we’ll be on a new blog, complaining about the lack of progress from IE compared to the rest…

    You’ve cleaned up some of your mess, and I’m happy for it, thank you. But it just seems like under achieving again, and I have little excitement hearing about anything from this camp any further.

  358. Anonymous says:

    With a simple xsl hack, I can make xhtml display correctly in IE when parsed as xml, so I don’t think IE is very far from supporting XHTML.

  359. Anonymous says:

    — Quote —-

    1) If you click on a label of a checkbox, radio button or the like this should be treated like a click on the checkbox itself.

    2) Dropdown menus (select tags) are always on top, you cannot make anything appear above them. (Yes I know, because they use native widgets, but that´s a poor explanation!)

    — Quote —-

    1) And exactly how does the checkbox magically know what its label is? Is it the text to the right, or maybe the text to the left, above, below?, in column with hundreds of other little checkboxs? is it an image to the right, or maybe something embeded they are selecting? Maybe an Iframe, or it could be positioned anything.

    So I would suggest you figure out how to make the checkbox that magically knows all this, while the rest of us continue to use that really hard to understand concept of … label.

    2) There is two very very simple way around that – that works in Moz with no problems – and takes about this much code …




    <iframe src=invis.html id=a style=display:none;position:absolute;top:0;height:0;width:100%;height:100%;z-index:-1 allowtransparency></iframe> <img src=invis.gif> for other browsers

    Its sloppy, though I cant say I would care much – it works, (you would need to move it or make it as big as the full page to cover the scroll – either way, who cares – nobody can see it)

    oh – thats a hack, my bad – toss it out – who needs those things, even if its gonna get me over a windowed object, make it so I can drag an object across another window in an iframe, get above flash, java – etc etc

  360. Anonymous says:

    — Quote —

    Will IE 7 be "unbundled" from the OS? It seems to me that is a primary design arbitrary, which if rectified would make all of this a whole lot easier. I can’t really say that having a browser "built in" to an OS fulfills any of my needs or wants…..

    — Quote —

    Um, where you been, and do you have any logical circuits working?

    You should have just asked, hey, any chance you guys are gonna jump in front of a bus tommorrow?

    Same diff – and if you have no clue why – then you have to be pretty much clueless about what is going on in the world around you. (Might want to catch up on the 2000 Election, 9/11, Iraq, and steriods while you are at it)

  361. Anonymous says:


    1) And exactly how does the checkbox magically know what its label is? Is it the text to the right, or maybe the text to the left, above, below?, in column with hundreds of other little checkboxs? is it an image to the right, or maybe something embeded they are selecting? Maybe an Iframe, or it could be positioned anything.


    Zach, please read: <a href="">HTML”>">HTML 4.01 Forms – Labels</a> (URL: – Jump to the section on Labels)

    And perhaps adjust the level of your arrogance to be more in sync with your understanding of the issue.

  362. Anonymous says:

    Come on! Firefox, Opera, Netscape, Safari, Konquerer, they all are beyond CSS 1 spec. Why can’t IE7 be? Are you incapable of making IE as compatible as those browsers, or is it part of Microsoft policy that dictates that IE be "different" – Is this an edge that you purposely exploit? Pass the ACID2 test and I’ll forever switch back to IE.

  363. Anonymous says:

    Personally, I don’t have a problem with the way IE6 handles [b]most[/b] CSS2 issues. The problem is that much of it has been subject to interpretation (the most famous of which being the whole argument of whether borders should be caluclated as part of the dimensions of an object or as an addon to said object).

    But since people are putting wishlists, I figured I’d put a few of my own:

    1) The min/max-width and height support. But that’s been mentioned a few times now.

    2) Height inheritance in terms of percentages for nested tables within table cells. i was trying to work on a custom calendar for a client that required vertical borders, and because IE didn’t inherit the height of the child table from the parent table cell, I ended up having to spend 6 hours and 3 tries writing a Javascript that would calculate the appropriate heights and assign their styles.

    Needless to say, I found that to be a bit of a pain in the you-know-what. If you want to see the calendar, I can include the link in a post. (Doubt you guys would make it a priority issue though, since I’m the only one bitching about it, and justifiably so.)

    3) Keep the filters. I’d imagine you would anyway, but just in case they were coming out, they can be rather useful. I especially like the gradient filter, although it would be nice if you could do a 3-or 4-colour gradient as well.

    If possible (and I doubt this is even going to be), it would be nice if there were a coding standard you guys could develop that would be cross-browser and cross-platform. This would eventually (hopefully) make the gradient a standard and save load time and hassle slicing up an image in Photoshop when we can simply adjust a few values in a CSS file.

    4) A better Javascript debugger. Ideally, along the lines of the Mozilla/Netscape/Firefox debugger, but as long as it’s accurate and reports the correct error on the correct line that it’s on, as opposed to the line after, that would be good enough for me and I’m sure most others.

    As far as everyone talking about how IE is failing and how all the other browsers are going to take over and kick IE’s ass because it’s so far behind, I have three very simple words for all of you:

    Get a clue.

    Seriously. Get a clue.

    Why get a clue? For three reasons:

    1) You guys complain about support for a free product. I’m not saying MS shouldn’t support it, but if you’ve got 1 million customers spending $100 a pop for Product A, and another million spending $0 for Product B, which customers are you going tu support first? Product A. And since MS has a number of products that fall under the Product A category (Office, Anti-spyware, Windows itself), guess where the support is going to go, as opposed to Product B (IE).

    It’s not necessarily the answer we as designers/developers would want to hear, but we’re asking them to do stuff for us when they really don’t have a lot to gain from doing so, for reasons below.

    2) IE still has the vast, VAST majority of the browser market. I can accept that part of it is from lack of knowledge of other browsers. Some people don’t know or are scared to install alternates.

    Buuuuuuuut..I maintain over 20 networks of varying sizes (3 to 50 computers). And I’ve let everyone know that there are alternatives to IE, and in some cases about FireFox specifically. I’ve even sat there, downloaded, and installed the browser for some of them. None of them were geeks; they were regular, ordinary 9-5 working stiffs.

    And to a man (or woman), they all wanted FF uninstalled. Too much resource was being used up. The additional search engines were too confusing (where the G is for the Google search, that is). Pages that were fully loaded still had "waiting for…&quot; to load (a leftover from the old Netscape days, I guess). And a couple of people hated the tabbed browsing. (I’m not a big fan of it myself, but I can see why others are.)

    Every one of them wanted FireFox uninstalled and to go right back to IE. And to be perfeclty honest, I couldn’t blame any of them.

    The design community in general often makes one very big mistake: we assume that the non-design community will follow just because we speak. The problem is that it’s kind of hard for other people to follow us when we’re talking to ourselves.

    3) FireFox is a volunteer, open-source project. That means that people will not be as likely to devote time and energy to it (see point 1 for an explanation). And with that comes the same errors, bugs, and nasties (eventually) that IE is subject to, especially if the code is published for all to see.

    I’m just waiting for the day Claria manages to get its ugly hands inside of FF and run its unique brand of havoc. And that day will come. It’s just a matter of time. Those Gator people are as ubiquitous as Joan Rivers’ stretched-out plastic surgery face at an awards show.

    You guys seem like you’re on the right track, Chris. Even if you don’t get everything solved in IE 7 (which would be the ideal), at least there’s some semblance of progress which in turn should lend itself to future versions.

  364. Anonymous says:

    "Acid2 does not guarantee conformance with any specification"

    I think what they mean is that rendering acid2 properly does not neccessarily make the browser compliant with the standards. However, not rendering it properly means that you aren’t complying completly to the standards.

    Although Wilson has noted that acid2 is mostly a wishlist, acid2 does contain alot of tests for HTML, CSS, and PNGs in it that should be supported.

  365. Anonymous says:

    If the IE team stopped improving CSS now, I’d totally be fine with your beta fixes!

    This was great news to hear. Keep up the awesome work you guys.

    …on with the DOM now 🙂

  366. Anonymous says:

    IE should be standards compliant i think by introducing CSS 2.1 etc it will bring the platform farward. This should have been done along time ago tho i feel.

  367. Anonymous says:

    Regarding the * html {} hack…. Fix it!

    No hacks should be left in intentionally!

    If web developers want to target IE 7 we can use this:

    <!–[if IE 7]>

    <style type="text/css">@import "ie-7.css";</style>


    If the box model and all the previous IE bugs we’ve been using the star hack for are fixed then there should be no need for it.

  368. Anonymous says:

    I suggest another correction. In IE7 beta1, the accesskeys on the links only select them and don’t open them.

  369. Anonymous says:

    S&#229; var sommeren over for denne gang – tydeligvis. Etter en uke under Hellas’ brennende sol er jeg n&#229; tilbake i gr&#229; og kalde Norge. V&#230;rgudene vet tydeligvis at jeg har fri en uke til, men jeg skal ikke klage…

  370. Anonymous says:

    well sad but true, as we all web developers expected, ie7 wont compromise in follow the standards, they will do their own politic (as microsoft has always done).

    Even these days when CSS are now very important to just not think about.

    IE sucks someone post and he’s not far from reality…

  371. Anonymous says:

    All "sounds" very promising!

  372. Anonymous says:

    Now this is the kind of post that fills me with encouragement. Well done for the openness. So IE7 won’t be completely Standards Compliant… at least it will be a lot better than IE6, and lets just hope IE7.5 will be hot on IE7’s heels.

  373. Anonymous says:

    i wish you all to develop IE-compilant websites for the rest of your lifes. i bet that is the worst nightmare.

  374. Anonymous says:

    What about making scrolling tables work properly? (the idea is to be able to have a fixed footer and header and a scrolling table body)



    <!– Insert <TR><TH> tags here –>


    <tbody style="height: 300px; overflow: auto;">

    <!– Insert Many <TR><TD> tags here –>



    <!– Insert <TR><TD> tags here –>



    The idea is to have a table with a hundred rows and it will only be 300 pixels tall (plus the header and footer). When you scroll through the table the header stays put.

    In theory this should work, I’ve found nothing in the CSS declaration that says it shouldn’t. I see it as being an incredibly useful feature and hopefully you can adopt it into IE.

  375. Anonymous says:

    Well, uhm, it’s about time now wouldn’t you say?

    Microsoft showing off with functions and support that has been on the market for basically ages. Internet Explorer is at best a slow webbrowser with little functions and as always, tends to be so far away from the standardized methods and functions as possible.

    Very annoying, yet I am curious to see what MS has to offer against Opera and Firefox etc.

  376. Anonymous says:

    Kind of slipped in along with the whole announcement about Windows Vista and the beta release, was news of Internet Explorer 7 and its own beta release. Now, IE6 was the most advanced browser available when it came out in…

  377. Anonymous says:

    i *think* this may have been mentioned before but can you guys *please* fix the awful autoscroll image you get when you click the mouse scroll wheel. it’s currently all jagged and aliased.

    see this:

  378. Anonymous says:

    I’m trying out the IE7 beta and I found a webpage that has a bunch of red boxes on it that shouldn’t be there (just like in IE6):

    But seriously, there is more than just CSS2 that web developers are hoping to see supported in the next version of IE. Having to work with two different range/selection models, and different ways of attaching events (and while we’re at it please please find a way to get rid of the memory leaks here) is a serious pain.

    Another thing that would save a lot of headache would be to eliminate the need to employ the "IFRAME masking" technique to get floated divs to not be obscured by select boxes and active-x controls. IFRAMEs bring their own bugs to the table that make even this workaround cause problems, like misflowing the document and causing automatic rescrolling of their container at odd times.

  379. Anonymous says:

    I agree, that just W3C-recommendations should be implemented.

    But not a single word about XHTML2??? Will IE7 implement XHTML2, if it is recommended timely?

    Btw, as a leading software company, I am awaiting leading-edge products from Microsoft. That means, you HAVE to keep track, what the latest developments [in W3C] are and implement them.

    Otherwise, I think you will get hard times competing with Firefox, even, if the latter has his security probs, too.

    Another question is, will IE7 be .NET-based? Will it provide sandbox-capabilities?

  380. Anonymous says:

    CSS is a beautiful thing for both users and developers, but it is currently being completely destroyed by Microsoft’s utter lack of concern for standards. As a developer, I’m forced to abandon using CSS for layouts solely because of Explorer’s complete and total failure at rendering even the simplest positioning tasks anywhere close to accurately. Sure, we can work around most of them, but it’s too simple a task made too difficult (if not impossible in some cases), and there are better things to do with one’s life than trying to fight against a Goliath’s negligence, time and time again.

    Even more frustrating is the fact that this situation doesn’t exist because of some upstart no-name developer with no money for proper development. It exists because the largest, most powerful and influential software company (though certainly not even close to the best) on the planet simply feels no PRESSURE to do things CORRECTLY.

    At first, developers just assumed that CSS must be difficult – after all, if MICROSOFT can’t get it right, there must be something going on behind the scenes that we’re not seeing. Then Firefox comes out of nowhere, and – with a fraction of Explorer’s resources – gets CSS rendering 99.9% right 99.9% of the time. You just code correctly and it works. Hell, even Mac IE 5.2 is better than IE 6. Same story for Safari…

    The only conclusion one can draw is that Microsoft is INTENTIONALLY botching it’s CSS rendering, for reasons known only to themselves. A cynic would say they’re up to something similar to what they did with Netscape ("if we make it so developers have to write code for one browser or the other, their clients will always opt for IE because we’re dominant, and the other browsers’ methods will fall away, followed by the browsers themselves"). It doesn’t make any sense – and it certainly won’t work this time – but that’s how monopolies work: Destroy everything that stands in your path, regardless of costs because so long as you win, the end justifies the means.

    On the other hand, I’m completely wrong and Explorer is simply written by morons who don’t have a clue how to do anything correctly (see: Security). That could be…

    Either way, and once again, a monopoly has proven destructive to innovation. Developer’s really CAN’T use CSS much until Explorer is on board because no client is going to pay 30-50% more for a website that looks exactly the same as one built using older methods. We can’t sell it, so we have to abandon it. Thanks, guys. Great job…

    One can only hope that the tide is turning against Microsoft and that their day in the sun is fading out. And when they’re gone, the open sourcers will inherit their throne and do things the right way: correctly.

  381. Anonymous says:

    "Chris Wilson is promising they will work hard to make IE better. I don’t believe it. They don’t really love their product. Working hard means working day and night, releasing betas every week, not once in a year."

    Exactly, but I bet it has less to do with love and lot more to do with old white men in suits saying "We have 80% market share on a free product. What’s to gain by spending money to support something that only benefits web developers?" in behind-closed-doors board room meetings.

    Seriously, imagine this conversation between Chris and his corporate task masters:

    "Excuse me, Taskmaster? Uh, yes, we really need to get our CSS support working."

    "How much will this cost us, Peon?"

    "Oh, a few man years worth of labor and recoding…"

    "And how much will this increase our market share?"

    "Well, we have about 80%… But since virtually no one will notice or even care that is being used… Maybe it’d bump up to 85% ?"

    "Well, a percent is a percent. They all count in the bottom line. And how much does your product sell for?"

    "Ummm…it’s …it’s free, sir."

    Makes a bit more sense, doesn’t it?

    The old dinosaurs just work that way. It’s all about the bottom dollar, not innovation, beauty or doing things the best way possible. And we’re stuck with the consequences of their stifling inertia.

    Thank YOU, Department of Justice and your oh-so-effective antitrust laws!

  382. Anonymous says:

    It seems that IE7 will be good for web developers who supports standards. Thank you for your effort.

    And I would like to say that I also will be looking forward for MathML support in IE.

  383. Anonymous says:

    kidnap me at red deer alberta A#9 20 alford avenue for more imfo contact me at

  384. Anonymous says:

    Matt Ward – please re-read my post. I said these items are fixed in internal post-beta 1 builds, not the beta 1 build that has been released.

    Chan Xie – no, there is no exposure in the object model as you suggest.

    Fiery Kitsune – There WAS community influence in the quality of IE6 during its development. And yes, the blog did get a makeover Monday.

    Dirk – no, I would prefer not to comment on Paul Thurrott’s column, other than to say I do not believe my statement contradicts the Web Standards Project’s Acid 2 Test Guide, I believe it is in accordance with it.

    Leo Shklovskii – no, I don’t think I misunderstood what the Acid 2 test guide says at all – I agree with your statements, except that the Acid 2 test is a weighting of FEATURES, not a prioritization chart of features and bugs for Internet Explorer, specifically. The Acid 2 test doesn’t test, say, the Peekaboo bug. Does that mean we should have not fixed that until we’d implemented data url support? My point is not to denigrate the Acid 2 Test’s value as a list of features that users [web developers] want; my point is that Acid 2 should not be confused with "what users [web developers] want FROM IE." As a poster named "Farix" says, "…we must remember that Acid2 is just one goal post".

    Joe Developer – thank you. Your imagined conversation gave me a chuckle. I think I have some old business cards with "Peon" printed on them around here somewhere…

    …but it’s a shame you missed the whole history of IE and standards PRIOR to 2001.

  385. Anonymous says:

    IE7 security changes: Rob Franco of Microsoft provides guidance on some of the security work being done in IE7. The first beta, now in private release, adds additional constraints on some uses of URLs and browser scripts. Rob also describes…

  386. Anonymous says:

    They need to FULLY support CSS2 and the other stuff, why would they care so much about backward compatibility? I hope one day Microsoft will realize that they need to get to the standards and stop being behind everyone else, Mozilla Firefox > IE7 still to me, I know it’s not the final, but when it gets final, it better be something AWESOME to switch me back, not some kind of crappy browser that can’t support much of the standards.

  387. Anonymous says:

    I strongly support Microsoft software products. I use them on a daily basis, and I deploy them in the enterprise.

    However, I am confused and baffled by your approach to this browser market. You write, "We fully recognize that IE is behind the game today in CSS support."

    You speak of deadlines in getting this product out the door. With browsers that are publically available for free that offer a standards-compliant browsing experience, how you can release a product that does not comply to these standards? Its not a matter of Acid Tests – its a matter of standards. How can Microsoft release software that is continuously behind the game?

    You’re going from IE6 to IE7 – a major product revision, and still, features are "too difficult to implement"?

    This internet browser roadmap is, frankly, unclear to me. The market means nothing to you now, and IE 7 is simply a maintenance release. Microsoft has decided that this browser game is long over. It no longer matters. Correct?

    If that’s the case, stop kidding yourself. This is simply IE 6.01 or 6.1 at most.

    Not IE 7.0. You’re bug fixing and making minor feature enhancements. You’re not releasing a browser that actually follows the standards that the rest of the world believes in.

  388. Anonymous says:

    At first read it sounded promising. After I read it again, I started to note the weakness in the tone of intention:

    "our next major priority is removing the biggest causes of difficulty for web developers"

    Shouldn’t your priority be removing all of the issues, not just those which cause a headache for developers? It may well be a time/effort ratio but you’ll forgive me for asking the obvious — who’s fault is that? I know it’s not mine.

    "I think we will make a lot of progress against that in IE7"

    So what happens after IE7 is released? Abandoned again for another five years? You’re not exactly helping matters here, with your non-committing rhetoric.

    I truly would love to believe that you (and MS in general) give a stuff about anything but the stock price and the PR, but experience tells me that just isn’t the case.

    Even your much trumpeted anti-phishing solution is going to contact your own servers. You can turn it off, I hear. Yes, but as you know and I know, a great many hundreds of thousands of casual users won’t.

    "our top priority is (and will likely always be) security"

    I’m sure it will, as that’s where people are the most insecure and gullible. Tell them they won’t have their Credit Card number, identity, small puppy, etc. stolen and you can pretty much take what you want — such as, for example, having a handy list of which URLS people are visiting through your anti-phishing "solution". Nice way to calculate where best to spend the advertising $.

    Now for the standards.. <Rolls up sleeves>

    I’m intrigued, why now? Why not three years ago? Do we now have to wait for another company to come out with a browser which starts to hit your market share in order for you guys to get off your backsides and help the development community?

    You have no idea how much I would love to believe you guys have seen the light and realised just how important it is for MS to accept and embrace standards. However, given the MS track record, I can’t help but smell a rat and ponder the real motive behind this new enlightenment.

    As I see it, you have a golden opportunity to restore some faith with an ever-increasing array of disillusioned users, both domestic and corporate. How well you capitalise on that remains to be seen. However, going by the "aim to’s", "what we can and can’t do", "will fix as many of" and the comparative lack of promise, certainty and words of committment, I have a feeling we’ll be going through this again in another five years from now. Assuming your users are that forgiving — again.

  389. Anonymous says:

    Let’s get into standards now. Guess what, Paul? Your site, currently has 124 validation errors, according to the W3C’s Markup Validation Service. Even worse, the page which contains your “Boycott IE” story currently has 207 validation errors. Both pages don’t even define the page’s doctype, which is almost always the first line of the web page.

  390. Anonymous says:

    It is good news to read that better CSS support is planned in IE. Anyway take in mind that (possibly unlike Firefox) every IE release stays a fact for at least 10 years. Windows 98 has not disappeared yet, and XP and Vista are expected to have a similar lifetime. So every mistake you make now, and every feature of CSS 2.1 you don’t fully support, will produce headaches in 10, 15 or 20 years. That’s the weight of your responsibility, and under this point of view your posting is not too good news…

  391. Anonymous says:

    Just wondering if anything with regards to


    will be added?

    I don’t know if this is css1/2/3, but it’s somewhere. 🙂



  392. Anonymous says:

    Finally a IE with standards. It’s been very difficult to we, webdesigners, have to devolope a kind of "two versions" of a website. One for IE and another for the other browsers.

    Keep the good job.

  393. Anonymous says:


    I hope you keep on the good work until the end.

    Btw, the IE7beta1 doesn’t even render your own blog sites correctly!! ( eample: , compare in IE7 and FF)

  394. Anonymous says:

    Too many are too quick and too desperate to bonk Microsoft on the head. While being critical is fine (and I have been in certain Microsoft blogs), its also important to give credit where credit is due.

    Microsoft is fixing a LOT of headaches with this release. They may not be passing the Acid test, its true, but then, neither is Firefox. This IS a step forward, and I think its time we admit to ourselves the need to concede that.

  395. Anonymous says:

    could you add ways to vetical align content in block elements like positioned divs

  396. Anonymous says:

    am i dreaming!? 😉

    will "min-width" now work for <body>?

  397. Anonymous says:

    Bout time for some standards in IE. I don’t even waste my time developing for IE anymore. I’ve gotten to the point where I just say screw it, it’s not worth it for me.

  398. Anonymous says:

    "Disillusioned" – _NEXT_MAJOR_ priority. Next. Major. Not only. Yes, obviously, removing all causes of difficulty for web developers is something we want to do. Unfortunately, neither I nor anyone on my team has figured out how to warp space and time.

    As for our anti-phishing solution, obviously not everyone will be comfortable using it – which is why it is explicitly an opt-IN feature. If you have a magic-bullet solution to the phishing problem that doesn’t require central validation, I expect my security team has a job open for you.

  399. Anonymous says:

    I don’t have access to beta1, but have read several review of it, and seen some screenshots. The order of the toolbars seems counter-intuitive to me, but no matter…

    "…our next major priority is removing the biggest causes of difficulty for web developers."

    Personally, I think you spent too much time on tabbed browsing and not enough on more pressing issues, like broken rendering.

    While I do applaud efforts to fix those major CSS bugs (which have given me plenty of grief in the past), there is one item you have overlooked: the ability to have multiple IE versions *natively* co-exist peacefully. This is a major PITA for developers. FWIU, downloading IE7 blows away IE6. That does not help developers in any way–quite the opposite.

    Still, I await a public beta with reserved optimism. Just remember that, even though IE7 may have bugs fixed and better standards support, other versions of IE will be out there for many years to come. Just don’t close all the back doors that allow developers to both test in older versions, and work around their various CSS bugs.

  400. Anonymous says:

    :before and :after should be implemented, too

  401. Anonymous says:

    please implement border-spacing so we don’t have to use <table cellspacing="0" …

  402. Anonymous says:

    So I need to update &lt;ewelink pageid=&quot;-290f74e2af9877abf0d8e2dc6992a896&quot;&gt;my post from yesterday&lt;/ewelink&gt; with a day’s worth of perspective (and additional reads from my RSS feeds).

    Saner and clearer heads other than Paul Thurrott’s – and certainly s…

  403. Anonymous says:

    Currently, IE falls back to quirks mode if an XML pi (<?xml …?>) is placed before the doctype. I’d like to see this bug fixed.

    Anyway, this list looks really promising. I hope you’ll be able to stick to your plans to implement all these fixes and features.

  404. Anonymous says:

    "neither I nor anyone on my team has figured out how to warp space and time."

    I think I I dealt with that predicted response before you even gave it by asking who’s fault it is that you are in this position.

    "If you have a magic-bullet solution to the phishing problem that doesn’t require central validation"

    Why can’t the user download a copy of the ‘suspect’ URLS locally from the MS server when they want instead of sending Lord knows what data to Microsoft?

    Further, won’t having to validate each and every URL with another remote server cause an extra hit on load times as another hop is added to the MS server? Yes, I know it would be tiny, but these tiny hits all add up.

    Now, who do I send my CV too? 🙂

  405. Anonymous says:

    This is extremely encouraging news. I am well aware that I could find lots of things to moan about, as other posters have, but my overall impression is very favourable.

  406. Anonymous says:


    I would prefer to see * html preserved. It is a useful, minimal-effort way to distinguish IE from all other browsers. Yes, it’s preservation will break some existing pages, but in my view the advantages of being able to select IE quickly and easily will outweigh the disadvantages.

    Either way, the decision to support or not support that hack should be made soon and announced loudly. Regardless of which way you go, there will be a large impact.

  407. Anonymous says:

    We have a web app that requires a lot of data dictionaries to be loaded; unfortunately, the only mechanism provided is through the web GUI itself.

    We’ve used VBA in Excel to control an IE instance by setting values of objects on the page and triggering requisite events; it works pretty well.

    Until you get to a page for a dictionary that will, based on circumstances, present a web dialog. The web dialog may only contain buttons; or in more complicated scenarios, a list. In any event, because the IE6 DOM does not expose the properties of objects in web dialogs, we’re forced to resort to SendKeys to push buttons; of course, we’re out of luck in making list selections.

    Is there any chance that, in IE7, there’s more exposure for objects in web dialogs?

  408. Anonymous says:

    Disillusioned –

    The user can’t just download a copy of the ‘suspect’ URLS locally from the MS server because the database changes constantly, and could grow extremely large. If it takes a day before the scheduled update download of the database, the user has probably already clicked on the "eBay account problem" link that you were spammed with. Plus the network load from each individual user increases dramatically.

    >Further, won’t having to validate each and every URL with another remote server cause an

    I believe that’s DOMAIN, not every URL.

  409. Anonymous says:

    "I believe that’s DOMAIN, not every URL."

    Semantics as well you know. My point stands that you know, and I know, those who are gullible enough to send unknown data to MS are those who will get caught with Internet nasties just as easily via other methods. Those who aren’t would rather sell you their parents than entrust their personal data to Microsoft.

    Sceptics could argue, of course, that it’s a way for Microsoft to be seen to be doing something constructive, which, in actual fact they know will be largely ignored because of the lack of trust in Microsoft that has arisen over recent times. Anyway, I’m shifting way off topic now and for that I apologise.

    As I have stated already, I’d love to be won-over with a 100% standard compliants browser and hope that I am. I can download software in a second to make me more secure, I can’t do a damn thing to make IE standards compliant.

  410. Anonymous says:

    Hi, Chris. You say : ""Disillusioned" – _NEXT_MAJOR_ priority. Next. Major. Not only. Yes, obviously, removing all causes of difficulty for web developers is something we want to do. Unfortunately, neither I nor anyone on my team has figured out how to warp space and time."

    I’d like to know how many people are working on IE7 now and when you start to work. You seem to say that you have not enough time to fix all the "biggest" bug…I’m quite surprised. You’re so a little team or you start to late to develop IE7 ?

  411. Anonymous says:

    I hope these can be supported in beta2:

    – margin: auto;

    – display: table-*; /* here * is cell, row-group, etc.*/

    – content: "string";

    – list-style-type: decimal-leading-zero;

    – caption-side: bottom;

    – cursor: x, y /* here x is a not-working uri def and y is a valid resource */

    (and some non-CSS off-topic stuff)

    Also, the IE is still complaining the XHTML 1.1 DTD is "not valid". And I think IE7 *should* support Punycode. And finally, the tab doesn’t work quite well: when I drag a link to the ta panel, it just load the link in the current tab.

  412. Anonymous says:

    Good post Chris, with some promising changes to come in beta2.

    Not knowing about the internal politics of MS, it’s hard to know why we’ve found IE in such a parlous state, but it’s good to know that somebody (.ie. your team) gives a sh*t.

    Unfortunately, it’s probably too little too late to win back the developer community (and i’m not just talking about the anti-MS devotees), and it will take a version of IE that’s distinctly ahead of the game, not close behind it, to win people back.

    Still, keep pushing, and remember that many of us developers do appreciate your efforts!

  413. Anonymous says:

    I heard a chorus of angels singing while reading this article. Thank you.

    Thank you.

    PS you should have comment numbering on this post 🙂

  414. Anonymous says:


    yeah it’s getting there slowly but surely.

    Guys 1 more things I think is useful.

    Why don’t you implement functionality for the ‘accept’ attribute for the input tag type=’file’.

    No browser supports it yet (that I know of), so be the first to do it.


  415. Anonymous says:

    Great post. Looking forward to many of these improvements. I’d also love to see the background-image flicker (on hover) issue addressed.

    Thanks so much for all your work.

  416. Anonymous says:

    There&amp;rsquo;s a reason I use Thunderbird&amp;rsquo;s weblog-reading component: there&amp;rsquo;s just too much good content on the Web these days.

  417. Anonymous says:

    These CSS updates will be great.

    This may have already been mentioned, but there are too many comments for me to read! My IE7 wishlist would also provide support for:

    Multiple class selectors:

    .red { color: red }

    .blue { color: blue } { color: purple }

    <p class="red blue">This text should be purple.</p>

    Different classes for containers with an ID: { color: red } { color: blue }

    <p id="container" class="red">This text should be red.</p>

    <p id="container" class="blue">This text should be blue.</p>

    And, of course, a combination of both: { color: purple }

    <p id="container" class="red blue">This text should be purple.</p>

  418. Anonymous says:

    Paul Festa says it quote well in his article yesterday (…

  419. Anonymous says:

    Currently, when writing the XHTML/CSS for websites, I use two style sheets – 1 for Gecko browsers, and 1 for IE – using conditional comments for filtering.

    With this new support for CSS coming in IE7, would it be prudent to add another conditional comment directing IE7 to make use of my Gecko styles, so that it doesn’t barf all over the styles written for IE 5.x/6 (the styles that are meant to negotiate many of these bugs)?

    If this would be prudent, will IE7 handle it properly if it is written as "…if IE7…", which is how I currently write it for IE5.x/6?

  420. Anonymous says:

    want to use CSS2 with ie6 and ie5.5 ?

    then look at this:

    this think just make my life better :-p


    and here is a bug that i dont thik that u know abut:

    if i have body{direction:rtl;) or <body dir="rtl">

    then the scroll bar move to the other side of the window and body{margin:auto} (for centering the content) dosent work so good…

    to fix that i use:



    you can please fix that thing?

    i’m very hope thet ie7 will finnely support css like FireFox and Opera.

    FireFox RULS !

  421. Anonymous says:

    want to use CSS2 with ie6 and ie5.5 ?

    then look at this:

    this think just make my life better :-p


    and here is a bug that i dont thik that u know abut:

    if i have body{direction:rtl;) or <body dir="rtl">

    then the scroll bar move to the other side of the window and body{margin:auto} (for centering the content) dosent work so good…

    to fix that i use:



    you can please fix that thing?

    i’m very hope thet ie7 will finnely support css like FireFox and Opera.

    FireFox RULS !

  422. Anonymous says:

    I wish you people all the luck in the world and I honestly hope that IE will support the official standards one day. The great lack of support is a major thorn in the eye of every goodwilling webdesigner, to put it lightly. (Who knows, you might even ditch ActiveX one day as well)

    I’m a webdesigner which uses Firefox himself, but that doesn’t keep me from wishing you all the best in developing IE 7.

  423. Anonymous says:

    I’d echo the requests for :before and :after as well as min/max width and height. I’d also like to see all the display: table (cell, row, etc) implemented. I think this’ll go a long way to allowing really great CSS layouts without some of the problems of them breaking on window resize.

  424. Anonymous says:

    "Disillusioned – The user can’t just download a copy of the ‘suspect’ URLS locally from the MS server because the database changes constantly, and could grow extremely large."

    put the phishy domain/url list into a bloom filter.

  425. Anonymous says:

    found a funny (and yet unkown?) bug in ie6

    check out the url!


    Tobias 🙂

  426. Anonymous says:

    Sorry if this has already been asked and answered, but are there any plans for a more advanced view / edit source method, which could perhaps include code highlighting?

  427. Anonymous says:

    Good to see Microsoft is finally getting serious about standards. After the (at the time) fantastic CSS support that IE 3 had, Microsoft has been content to rest on its laurels ever since. I hope this focus lasts, as it makes the Web a much more dynamic and cool place for everyone.

    Some requests:

    * Please do not intentionally include CSS hacks. <!–[if lt IE 7]>@import url("ie6.css");<![endif]–> exists expressly for this purpose.

    * Please include a DOM tool similar to Firefox’s, with DOM node/JavaScript object inspection. I can’t tell you how many times I have wished for this in IE when developing complex JavaScript applications and classes.

    * As others have said, please find a way to get the new rendering engine into the hands of users without Vista. Web standards compatibility is not going to sell Windows, but it will make our lives easier, especially going forward.

    * Please fix the autoscroll icons. They look awful. Three years ago, I created the Firefox scroll icons in less than half an hour. Maybe someone can do it on their lunch break or something. 🙂

    Also, there seems to be a bug with insertBefore(). I wrote an application that incorporated the ability to dynamically reorder list elements and I had to prevent IE from being able to drag beyond the last element. The item would simply disappear. Worked properly in all other browsers.

    Anyway, thanks for taking the time to listen to our suggestions, even if it’s somewhat belatedly into the product lifecycle.

  428. Anonymous says:

    You addedd transparency support for pngs, but the color which IE7b1 displays is not the same as photoshop displayd on creating, and firefox on viewing the images… you can see it here:

  429. Anonymous says:

    That’s all that I want as web developer, But I agree with the way you are doing things.. the worst problem with IE right now is Security. That is a MUST.

  430. Anonymous says:

    Something that bugs me about IE is the Windows Update feature. Obviously security is #1 on your list, as it certainly must be, but I think #2 should be new -or improved- features. You have Windows Update for security largely, but rarely ever is it used to provide users with new features. It would be great to go to the Windows Update site and see optional downloads for IE. I just wish it were utilized more for user experience improvements every once in a while, that’s all.

  431. Anonymous says:

    "As for our anti-phishing solution, obviously not everyone will be comfortable using it – which is why it is explicitly an opt-IN feature. If you have a magic-bullet solution to the phishing problem that doesn’t require central validation, I expect my security team has a job open for you."


    I’m not privvy to all the conversations that your team has had internally, but it would make sense that allowing 3rd parties to establish databases would go along way in thwarting peoples discomfort with sending information to the database. Ideally, 3rd parties could download and reuse the MS database, as well as suggest entries for the master database based on reports from users. It would seem obvious that a method to maintian public validatation of 3rd party vendors would be in everybody’s best interest as well.

  432. Anonymous says:

    Disillusioned –

    It is most certainly NOT just "semantics" on domain vs. URL – you were attempting to make a point about hit on load times per page. If you only have to check per domain, that load time effect is not as significant, correct?

    As for those "gullible enough to send unknown data to MS" (and it’s not unknown – read the EULA) – if, as you say, those are the demographic who will get caught with Internet nasties, then it’s a good feature for them, no? And those who won’t can just "download software in a second to make [them] more secure"?

  433. Anonymous says:

    There is a rather frustrating bug in IE’s form handling that no one seems to mention very much. When you press a button, IE posts the text on the button, rather than the value attribute of the button. It also posts the text for all buttons, not just the one that was pressed. This makes it impossible to use a form with more than one button that has the same name.

    I’ve put together a testcase

    One instance where you may want to use multiple buttons with the same name but different values is if you have a list of pictures (or anything else) with captions, and for each picture you want the option to edit the caption, or delete the picture and the caption. Then you might have two buttons below each picture called "Edit Picture" and "Delete Picture", both with name="action" and values of "edit" and "delete" respectively.

  434. Anonymous says:

    Exciting news so far. There is one issue which I’ve seen mentioned once, but there’ve been no responses from Chris. This involves image caching problems when a user has changed the browser cache setting from the default "Check for newer versions of stored pages: Automatically" to any of the other choices. Is this issue going to be addressed?


  435. Anonymous says:

    white-space: nowrap still doesn’t work. W3C says that you MAY support this and MAY not, but still nice to have feature.

  436. Anonymous says:

    I would like for you guys to fix HTTP compression. I’ve seen several sites that say IE6 SP2 is supposed to support HTTP compression but it doesn’t work right. I believe it’s a combination of IE’s internal caching and compression that truncates remote javascript and css files after a certain point causing javascript errors that come up sometimes. Pressing reload generally fixes it but I’d like to not have to tell clients all the time that sometimes the site will break and if it does, press reload. I’ve read in many areas where people would definitely like to use compression as it would save them large amounts of bandwidth. Please fix this bug.

  437. Anonymous says:

    The Mozilla team seems to have the whole standards issue sorted. As Firefox is open source why don’t you just use their code and re-jig it for your needs. I appreciate that IE is heavily embedded into windows and many of it’s applications rely on the IE HTML engine.

  438. Anonymous says:

    It’s so SIMPLE you cannot imagine:

    in fact it needs to make ONLY WRAPPER to make existing native table rendering engine able to render the SAME tables but on basis of ANY elements instead of only native TABLE/TR/TD elements.

    There is only one real difference compared with traditional tables — slightly another padding implementation: equal padding values for all virtual cells in framework of precise virtual table (how we can see what way it’s already realized in other browsers). That’s all!

    Please add support for it. It’s AMAZINGLY NEEDED for all web developers in the world.

    Thanks in advance.

  439. Anonymous says:

    After reading a dozen or more posts recently on Microsoft’s direction with IE7 and the apparent hard-nosed opinions of many I decided to write a post of my own. I will start by saying I’m not sitting on either side…

  440. Anonymous says:

    border: 1px solid transparent;

    Please fix that too!


  441. Anonymous says:

    Please add the CSS properties min-width, max-width, min-height, and max-height to IE7. (For the record, it’s the lack of max-width that I bang my head against the wall about.)

    These properties would let developers to pursue elastic layouts, which create scalability for low screen resolutions and preserve readability for high screen resolutions.

    But don’t just listen to some random guy’s blog comment. Here are my sources:

    And my apologies if this last one is a bit of a sore subject. 🙂

  442. Anonymous says:

    Why wasn’t this done 3 years ago?

  443. Anonymous says:

    In addition to fixing the HTTP compression problem, could you also fix the progressive jpeg/image interlacing problem? I find it rather annoying that a non-interlaced image displays interlaced and a interlaced image displays non-interlaced. All other browsers comply and display interlaced images properly.

  444. Anonymous says:

    In addition to fixing the HTTP compression problem, could you also fix the progressive jpeg/image interlacing problem? I find it rather annoying that a non-interlaced image displays interlaced and a interlaced image displays non-interlaced. All other browsers comply and display interlaced images properly.

  445. Anonymous says:

    Generated content. Period. Exclamation!

  446. Anonymous says:

    Please, please, please get dotted lines working. It’s insane that this works in all the other leading browsers and not in IE.

  447. Anonymous says:

    I shall now troll…

    A little too little and much too late.

    To say that I’m disappointed about the failure of standards recommendations support for IE 5.x thru 6.x is a severe understatement on my part.

    Although, what you’re promising on this blog is a good thing. Whoopee, what made you finally give a crap? I guarantee that you did it because you were pressured, not because you guys gave a crap about standards (open source tickling your keys, eh?).

    Regardless, for those that will use your product and have to code for it… good. For the team fixing it… good.

    For me however, I’ve had to suffer greatly because of your company and it’s management teams and engineers ridiculous excuses for products and no longer wish to deal with it.

    If your company and it’s so call products (most of which you purchased and modified for your own needs) were to fall off the face of the earth, I would only notice because I no longer had to suffer coding to work around your painful and ever persistent issues.

    As much as this sounds like MS bashing – it is not entirely. I simply do not care for companies (with all the resources in the world – literally) making inferior products, forcing them on us and doing little to fix them in a reasonable amount of time. If you made cars or large appliances, you would certainly been out of business many years ago.

    It’s too bad for us, that bill g. and crew picked technology for their living, were good at bullshit marketing, deception and strong arming vendors to force such inadequate products into so many people’s hands. In a way, your consumer sheep deserve what they get. Oh how many of your flock shitcan their machines after they’ve been rendered useless from spyware, etc.

    However, I for one will not use or purchase your products. Every chance I get, I proselytize and convert one more person to another platform. Thus, reducing their pain and mine in supporting them.

    Believe me – this has worked extremely well for me and my customers.

    Good bye, good riddance and FU for your audacity in now telling us – oh boy look how happy you’re gonna be – we fixed our crappy browser. Now our stuff works. Please use our crap again.

    No thanks. I can only hope you people will have less and less to do with the Internet as time goes on. No matter how much you’ve fooled people, eventually the truth rises to the surface and you’ll be found out for your crimes.

  448. Anonymous says:

    Its about time.

  449. Anonymous says:

    Some days a go I was creating CSS2 IE version (just by w3c standart written CSS doesnt work fully in IE) for one web page and hit a simple IE crash case. Simple CSS and HTML crashed IE 5.0, 5.1, 6.0 and, as on IRC some tested, 7beta1 also. Nice.

    I dunno if it is some well known bug or new one. Dont blame me, if You know it already, blame IE 😉

    Heres example:

    <style type="text/css">

    #foo p:first-letter { margin-right:1px;}

    :link, :visited { background-color: gray; }


    <div id="foo"><p><a href="">MSIE sux</a></p></div>


    <style type="text/css">

    p span:first-letter {



    :first-line {




    <p> <span style="display:block;"> <a href="/">f</a> </span> </p>


  450. Anonymous says:

    How about printing iframes?

    There’s probably been i comment about this somewhere that i’ve missed, however if your frame spans over a printed page, IE seem not to bother printing anything on the second page. Is this going to be dealt with?

  451. Anonymous says:

    Call it wishful thinking, but I would love to see integrated SVG support, failing that it would be nice to have IE7 bundled with Adobes SVG plugin.

    I was dismayed when I tried out Beta 1, but if you even manage half these fixes in Beta 2 it will start to renew my faith in IE.

    I’m not switching back to it, though. Safari 2.0/Firefox suit me fine.

  452. Anonymous says:

    In case this haven’t been report before. The document.getElementById is not case sensitive as it should be, getElementById("A") able to return element with id="a".

  453. Anonymous says:

    I think it is great that IE is going to be supporting CSS to a large extent. And yes, security should be a higher priority. I hope you will add to your list of CSS stuff to add (though I think it’s CSS 3): transparency. Anyhow, good luck programming.

  454. Anonymous says:

    I agree with Philip that integrated SVG support is needed soon.

  455. Anonymous says:

    Just wanted to let you know that after being IE loyal for more than 10 years, I have decided never to use it again except to test. This is in response to learning xhtml/html strict/css. Your browser sucks so much that I am frustrated enough to stop defending it to my friends and I’m helping all of my browser-illiterate friends switch over to Firefox.

    Thought I’d let you know… not that you care.

  456. Anonymous says:

    "Thought I’d let you know… not that you care."

    Just my two cents, but lets all remember that there’s a person, or rather group of persons, at Microsoft on the other end of these comments. There’s no need to be rude. I share and completely understand the frustration that IE’s quirks and bugs have caused web developers (I am one), but lets keep our disdain pointed at the company, not the group of people commissioned to fix the problems. Simple courtesy.

    "…I have decided never to use it again except to test."

    As have I; sorry Chris (and team), but I doubt I’ll ever use IE as a regular browser again. But remember folks, the better this team fixes the product, and the more we provide feedback to assist them, the less testing we’ll have to do in IE, and the faster we’ll produce results for our clients/customers/selves.

  457. Anonymous says:

    "Thought I’d let you know… not that you care."

    Just my two cents, but lets all remember that there’s a person, or rather group of persons, at Microsoft on the other end of these comments. There’s no need to be rude. I share and completely understand the frustration that IE’s quirks and bugs have caused web developers (I am one), but lets keep our disdain pointed at the company, not the group of people commissioned to fix the problems. Simple courtesy.

    "…I have decided never to use it again except to test."

    As have I; sorry Chris (and team), but I doubt I’ll ever use IE as a regular browser again. But remember folks, the better this team fixes the product, and the more we provide feedback to assist them, the less testing we’ll have to do in IE, and the faster we’ll produce results for our clients/customers/selves.

  458. Anonymous says:


    I agree with the rest of the gang who appreciate your honesty and "up-front-ness" about all this stuff. As web developers, we’ve been crying in our beer for years about CSS2 and when you would actually pay attention, and this is the first glimpse of sunshine in our world…maybe you really are listening…the most glaring problems are being addressed!

    Thanks for your hard work, and I look forward to more "notes from the trenches" on this exact kind of stuff in the future. It’s really gone a long way toward building confidence.

  459. Anonymous says:



  460. Anonymous says:

    We would welcome a *complete* support for the <label> element. In HTML 4, a <label> tag can be associated with a control either explicitly (via the "for" attribute) or explicitly (if the <label> tag contains the control). The former has been supported since IE4, but the latter is still not supported. (explicit association) (implicit association)

  461. Anonymous says:

    — quote

    # Improper Button Handling

    Monday, August 08, 2005 9:54 PM by Jason

    There is a rather frustrating bug in IE’s form handling that no one seems to mention very much. When you press a button, IE posts the text on the button, rather than the value attribute of the button. It also posts the text for all buttons, not just the one that was pressed. This makes it impossible to use a form with more than one button that has the same name.

    I’ve put together a testcase

    —end quote

    I would like to reiterate this problem. w3c specification is that the value attribute should be returned ( MSDN documentation is that the innerText value is returned (, which becomes a particular problem for multi-lingual applications where the button text is presented in the user’s language whilst its value attribute is a specific value recognised by the form processor and independent of the user language.

  462. Anonymous says:

    Hooray! Looks great. Thanks guys!

  463. Anonymous says:

    I think there is nothing great abt internet explorer 7 . You people still havent developed it enough to support W3C CSS.2 .. Its still the old stuff with no much improvments what the big deal with desiging new interface with tabs and logos if you guys cant fix the major issues. check out the review internet explorer has got in

    Personally i feel microsost doesnt give a damn about the W3C standards since 80% of the people world wide use it.

    Its nothing but a headache for webdevelopers and designer.

  464. Anonymous says:

    Thank you! Now, what about DOM events model?

  465. Anonymous says:

    I’m happy to see that condsiderable progress is being made to improve IE’s standards-compliance.

    I would like to point out that Paul Thurrott has posted a follow-up to the oft-referenced "boycott" article:

    And, I would like to reitterate the commonly held desire that the inclusion of an <?xml prolog not throw IE into quirksmode. This problem, while perhaps more easily overcome than others, strikes me as particularly ironic, in that the inclusion of the prolog during page creation as part of an attempt to fully follow the recommendation in fact causes the entire page to be rendered in quirksmode. Further, I agree with the authors at that the policy of doctype switching itself deserves to be reevalutated.

    By allowing people to get away with bad code for so long, IE has become obligated to support bad code for the forseeable future. What will it take for IE to change course and begin insisting that code be brought up to an acceptable level of standards-compliance?

    Thank you again for your efforts,

  466. search-engines-web says:

    STYLE FILTERS are not being developed enough.

    They are a GREAT resource for Designers –

    fast loading, attractive, and (text)read by Search Engines.

    They also, are an asset for enhanced DHTML effects.

    More emphasis should be placed on REALLY developing them further – including TRANSITIONS.

    More work should be placed on Marketing them to Designers and Developers, as an option to Web graphics.

  467. Anonymous says:

    search-engines-web, you are joking, aren’t you? IE needs to follow agreed standards, not carry on down the path of setting its own. The style filters, while initially a fun toy to play with, deviate from the goal of haiving one code base to deliver to all (current) browsers.

    I’m sure most people would prefer the CSS3 opacity attribute to IE’s alpha(opacity) filter (and to the -moz-opacity attribute, for that matter.)

  468. Anonymous says:

    Here’s a behavior I’d really like to see fixed in IE7: when a page contains a character that’s not in the font being used to render it, IE should <b>not</b> display a little box. The CSS 2.1 spec says:

    <blockquote>Although many fonts provide the "missing character" glyph, typically an open box, as its name implies this should not be considered a match for characters that cannot be found in the font.</blockquote>

    This comes up all the time in my field, linguistics, when we use characters in the International Phonetic Alphabet (Unicode 0250-02AF and various other places) in web pages. When Firefox displays a page with IPA characters, it finds a glyph for them even if the current font doesn’t contain one. I’m not sure *how* it’s finding a font (my experiments suggest it first looks through all the fonts in the CSS font-family list, and if it doesn’t find a glyph, it looks through all the fonts on your computer, but that’s just a guess), but it does find one, and that means users are (almost) never shown a "missing character" glyph.

    It’s worth pointing out that the IPA range is a particularly tricky one for the way IE has handled fonts in the past (see: Tools, Options, General, Fonts), because the IPA characters are (correctly) not part of any language. That means any solution that involves specifying default fonts by language won’t help.

    I suggest making IE7 search character-by-character through the font-family list like Firefox does, and then perhaps have a way to specify a default font for a given Unicode range.

  469. Anonymous says:
  470. Anonymous says:
  471. search-engines-web says:

    STYLE FILTERS are a god-send – they are an AMAZING feature.

    Opacity has Many. Many tweaks available to refine their look – and because of VECTORIZATIOM the effects can be stunning.


    There are many other features availble – that can add design enhancements –

    e.g DHTML drop-down menus get an added dimension with some of these effects.

    Also, they disappear gracefully in other browsers – so standards are not a factor.



  472. Anonymous says:

    If you need to play with toys that only IE users will see, and thereby limit the experience for other users, rather than viewing sie devcelopment from a standards perspective, that’s your perogative I guess. I’m certainly not going to get into an argument as to how you should dvelop your sites.

    However, where IE filters duplicate CSS attributes, I believe that the CSS should at least be supported in tandem to, if not in replacement of, the IE filter. That way, those of us who have standards in mind can develop one site for all, and those that need to add little extras for a specific browser are free to do so.

  473. Anonymous says:

    Antes que nada he de decir que este art&#237;culo est&#225; inspirado en este otro: diez razones para no confiar en internet explorer 7, y sobre todo en estos: La verdadera raz&#243;n por la que Internet Explorer 7 no soportar&#225; CSS2 — Update on Standards and CSS in

  474. Anonymous says:

    Let me join the crowd that asks for better CSS, DOM support and standards compliance in MSIE. I, as a web application developer, would rather spend time writing useful code than looking for MSIE workarounds.

    The button.value issue is really annoying, as it is not possible to have a value attribute that is different from the text content of the button. The current MSDN documentation is incomplete/incorrect on the subject. This is HTML4/DOM1, not even DOM2!

    Also, the JScript engine is not currently compliant with ECMA-262, in particular RegExps are not compatible with the above spec (empty string instead of undefined when a pattern does not match), which is also painful. An ECMA-compliant Error object that tells more details about where the error happened would be appreciated too.

  475. Anonymous says:

    The image flicker is not just a :hover problem, but also occurs when setting the style.cursor attribute using JavaScript.

    This site shows the problem when when you move in and out of the title bars, or the outer resize areas of the windows it displays. The cursor is set using JavaScript.

  476. Anonymous says:

    you don’t need wait. Get Firefox

  477. Anonymous says:

    That’s great. Keep the good work. The more bugs you fix, the more of CSS you implement, the better. Maybe IE will stop being the developer’s hell as it is now.

  478. Anonymous says:

    Al, you wrote:

    > The page where you downloaded it on MSDN has

    > links and information for reporting bugs.

    > Please report them there.


    > Thank you,

    > Al Billings [MSFT]

    This is not the case. Myself and several others have asked for info on where to submit a bug. I cannot find any info in the MSDN download section, release notes, installation files, or the given link where the IE team requests feedback.

    Please open the door for us to give proper feedback.



  479. Anonymous says:

    I would like to add to Jason’s quite valid point about the button not sending value=value that the button compared with the input type=submit button that msie currently does support seems oversized. For quick reference both buttons alongside in a snapshot:

  480. Anonymous says:

    It’s about time!! Great news.

  481. Anonymous says:


    In IE, it is not possible to determine the currentTarget with attachEvent. IE does not support currentTarget. The scope of the object (e.g. this keyword), is lost.

  482. Anonymous says:

    like jim explained,

    the * html {…} hack in my opinion and experience is very important, please let it in for filtering.

    the tan hack helps separating ie5win from ie6/ie5 mac.

    div#box{width:100px; border:10px solid #666;}

    * html div#box{width:120px; width:100px;}

  483. Anonymous says:

    Microsoft are trying to fix rendering engine bugsfor IE7. We are also hard at work doing the next rendering engine in Opera, and now Tim Altman wants you to post the most important bugs so please leave a comment at his blog with the most important rendering engine bugs. Not pet bugs, but the ones that are causing the most problems on a day-to-day basis.

  484. Anonymous says:

    Hooray for Chris!

  485. Anonymous says:

    "Alpha channel in PNG images"

    Halle-friggin-lujah!! If it weren’t for youngpup’s sleight I wouldn’t be using png now.

    May I ask if the enabled filter will work in background images?

  486. Anonymous says:

    For Simon Wilson: IE conditional comments are great but lack FULL usability. If there where a possibility to do eg IE5.5 AND IE6 like !IE6 (not IE6) it would be fully usable.

    # To select for IE 5.01 and 6 – cannot be done!

    # To select for IE 5.01 and 7.0 – cannot be done!

    # To select for IE 5.5 and 6 – cannot be done!

    # To select for IE 5.5 and 7.0 – cannot be done!

  487. Anonymous says:

    For Simon Wilson: IE conditional comments are great but lack FULL usability. If there where a possibility to do eg IE5.5 AND IE6 like !IE6 (not IE6) it would be fully usable.

    # To select for IE 5.01 and 6 – cannot be done!

    # To select for IE 5.01 and 7.0 – cannot be done!

    # To select for IE 5.5 and 6 – cannot be done!

    # To select for IE 5.5 and 7.0 – cannot be done!

  488. mazoo68 says:

    I spoke with Microsoft support today in regards to this error, they currently have no support for IE7… so I am hoping someone here might have an idea or could refer me to someone that could help with this error we are seeing…

    IE 7 reports an issue with ASWWAX.OCX (version 6.5 Authorware) and then IE 7 is forced to shut down… The ASWWAX.OCX shows a version of The IE 7 version is 7.0.5112.0 beta. Behind the message saying IE 7 will be shut down is a message box saying:

    Title bar: Sysfader:IEXPLORE.EXE – Application error.

    The instruction at ‘0x05f2b0b3’ referenced memory at "0x00000010", The memory could not be read.

    Click on OK to terminate the program

    Click on CANCEL to debug the program

    Please note that clicking CANCEL does not produce any usable information.

    This ActiveX control works fine with all previous versions of IE. Any ideas? A representative of Macromedia’s Authorware forum said I should contact MS about this issue.

    Patrick Harvey

  489. Anonymous says:

    Hey Chris,

    I feel it’s weird there are so many people looking at negatives and not looking at positives. I like IE toolbars (the newtab button is awesome) and I have always liked its speed compared to firefox. But I absolutely can’t stomach the weird repositioning of the menu bar and tool bar… the toolbar is not even movable. I am pretty sure you guys will fix this before the actual release. Honestly I don’t mind if you rip the interface straight off firefox (think nice and compact :))

    The whole standards issue is debatable but hey what the hell, here people are complaining over every single css problem they can find. Being programmers they should know nothing works out of the box.. that’s why it’s your job to make it work — problem solving no ? 🙂 things don’t get done in one shot… it’s disappointing that IE is still not there but I hope it will be someday.

    First and foremost on my list would not be the css issues, but the security and interface problems that need to be addressed. Afterall, Joe couldn’t care less about the technical glitches; Joe wants a safe browser that’s fast and customizable. Then we talk about getting every damn standard straight… doesn’t mean we leave it completely, we rather build on it slowly. I agree with your strategy but I think more needed to be done with the whole CSS thing… oh well, back to firefox for me. I tried IE .. it’s bitter/sweet.. I hope it turns out more sweet than bitter. Right now Firefox is very sweet 🙂

  490. Anonymous says:

    I know its mentioned by several others. But as a web developer working with CSS layouts I also wish to see the CSS properties min-width, max-width, min-height, and max-height added to IE7.

    (For the record, it’s the lack of max-width that I also bang my head against the wall about all the time.)It takes ages for eveyone to upgrade browsers and it would be great if there was any way to support min-width in addition to width etc for the release

    Overall I am encoraged by this blog and the communication from Microsoft.

  491. Anonymous says:

    I skrivende stund s&#229; finnes det kun tre nettlesere som best&#229;r Acid2-testen, best&#229;r din favorittnettleser testen?

  492. Anonymous says:

    I pray you guys decide to suck it up and pass the Acid 2 test.

  493. Anonymous says:
  494. Anonymous says:
  495. Anonymous says:

    As &lt;a href=&quot;; target=&quot;_blank&quot;&gt;Olli mentioned&lt;/a&gt;, Chris Wilson has &lt;a href=&quot;; target=&quot;_blank&quot;&gt;expounded on important interoperability bug fixes&lt;/a&gt; in the future IE7. This list of bug fixes will remove a lot of cross-browser headaches, but ironically fixing them will also add some new …

  496. Anonymous says:

    As &lt;a href=&quot;; target=&quot;_blank&quot;&gt;Olli mentioned&lt;/a&gt;, Chris Wilson has &lt;a href=&quot;; target=&quot;_blank&quot;&gt;expounded on important interoperability bug fixes&lt;/a&gt; in the future IE7. This list of bug fixes will remove a lot of cross-browser headaches, but ironically fixing them will also add some new …

  497. Anonymous says:

    As &lt;a href=&quot;; target=&quot;_blank&quot;&gt;Olli mentioned&lt;/a&gt;, Chris Wilson has &lt;a href=&quot;; target=&quot;_blank&quot;&gt;expounded on important interoperability bug fixes&lt;/a&gt; in the future IE7. This list of bug fixes will remove a lot of cross-browser headaches, but ironically fixing them will also add some new …

  498. JAM LOG : IE7β関連というエントリーより。 IE7βのCSSやWeb…

  499. sou says:


  500. IEBlog says:

    We are currently locking down IE7 for shipping and I wanted to give an update on the CSS work that…

  501. core says:

    What does "standard compliant website" mean?

  502. More Gumbo says:

    For the SECOND week in a row, I’m heading into town for a lunch meeting at Bayou City Seafood and Pasta. This time, I’m working with two guys who developed the best stock trading course I’ve ever seen.

  503. The normal fellow would be under the impression that investing the time to prepare articles on this subject of thought is a waste of resources.

  504. cell cingular gsm phone unlocked

  505. Click here: says:

    The responsible plan of action is to do your research comprehensively.

  506. Sometimes, you’ll become surprised by the large quantity of background searches info obtainable.

  507. La SP1, noutățile vin &#238;n trei categorii: &#206;mbunătățirea calității – funcționare, securitate,

  508. Dan Atkinson says:

    Internet Explorer 8 passes the Acid2 test!

  509. Maribel says:

    zijn in vele modellen.. Les gens passent beaucoup de temps sur gta 5 money hack et collecter de l'argent..

Comments are closed.

Skip to main content