Making Sites Look Their Best in Standards Mode

IE has traditionally drawn a 2-pixel border around the content area of a site. This border, drawn as part of the page rather than IE’s frame, affects calculations of distance from the top and left of the page. It also creates a not-so-modern beveled look.

In the fourth Platform Preview, you’ll notice pages running in IE9’s Standards Mode no longer have the border. Here’s a before and after:

webpage with 2px border
webpage with no border

Pages that run in legacy document modes will still have a 2-pixel border so that any site calculations dependent on the 2 pixels remain the same as in IE8.

To make sure your site runs in IE9 Standards Mode and gets this and all the other latest features in IE9, use a strict doctype. We recommend the HTML5 doctype (<!DOCTYPE html>) since it’s simple and will put your site in Standards Mode in all current browsers.

John Hrvatin
Program Manager

Comments (62)

  1. still not says:

    Just a reminder: innerHTML is still broken in IE9 platform preview 4.

  2. Joe R says:

    Small things really do add up. It's nice to see IE is still supporting legacy doc mode for those who aren't transitioned, yet. I'm pepped for the final release! : )

  3. blah says:

    Even I can't believe how ridiculous this one is. I always considered the sunken effect an innocuous bit of ugliness. It broke positioning?!?!

  4. That's very nice to hear. I always wondered about the questionable UI decisions in IE, like this adding an arbitrary border to every page, or the click sound, or "refreshing" the address bar after the page is loaded even if I'm in the middle of editing the text there.

  5. Greg says:

    That is the biggest admittance to failure I've ever seen from browser developers. But thank goodness you've noticed.

  6. Anti-Troll says:

    Greg, please either troll elsewhere or find a clue and troll something more interesting.

    Michal, you too, unless you really haven't upgraded from IE6 yet and are still having problems with your address bar. And since you're clearly new to using Windows, you should check out the SOUNDS applet in the control panel. Windows is neat because you can customize things like sounds. As for 3d-edges, they're hardly arbitrary– they're legacy. Windows 95 introduced them, and they've been present ever since. The IE team showed how to turn them off years ago if you want to, but I guess you're not used to the idea that you can configure your software as you like it. Software is neat like that.

  7. Harry Richter says:

    @ still not

    InnerHTML is still working as designed by their inventors – Microsoft! If you compare it with the implementations of other browser makers, you'll find that these other implementations are buggy, because they dont mirror image the behaviour set by its designers. Go complain to them!


  8. hAl says:

    I have also had serieus issue with UI designs and especially the interaction between UI elements and renderer.

    I am always annoyed that many extremelly slow page-load actions in IE lock down the adress bar or make the stop button not function properly.

  9. Niels van der Rest says:

    This isn't just a design thing, but also a major usability improvement!

    One of the little annoyances of IE for me has been the inability to scroll the page using the scroll wheel when the cursor is located on that 2-pixel border. I usually have my browser maximized, move my mouse out of the way (edge of the screen) and scroll to view the rest of the page. Except in IE, where that pesky little border takes away the focus from the page.

    Glad to see that's taken care of 🙂

  10. meni says:

    Harry Richter,

    I, who blame Microsoft for the web's 10 lost years, recognize great things that came from Microsoft. XMLHttpRequest, DHTML, innerHTML, ActiveX  and much more [just kidding on activex 😉 ]. HOWEVER, Microsoft as well as any other vendor, should not let ego get in the way of adopting standards of the web. If a standards body has published a standard that's different then even its inventor, go ahead and implement it. I have said it before: had Microsoft swallowed their pride and implemented SVG over the (arguably) better engineered VML in 2001 the web would be much better NOW!

    Standards >> pride

    meni, open-{standards,source} fanboi

    UNIX and the UNIX spirit fan since 1986

  11. Omar says:

    @Harry Richter – you must be the biggest ms fanboy ever! congratulations on being blinded by proprietary-ness.  Microsoft did invent innerHTML – agreed.  It must have been the most intensely amazing discovery to think that developers of a markup "tree" would want to be able to get/set the sub-tree markup – pure genius!  We thank Microsoft for their amazing invention that we could never have dreamed up and we implemented it in every other browser.  Make HTMLElement.innerHTML a getter/setter for the sub-tree HTML markup…. DONE! that was easy.

    Oh wait… Microsoft had a pile of bugs in their implementation of innerHTML such that it would return a bunch of wacky invalid (and not what I set) HTML when I use the getter… and that random elements that I would really, really like to set the innerHTML on don't support the setter option.  To be consistent, I should really spend time to find all these bugs and make sure that my browsers' implementation is as crippled as Microsoft's.

    Oh wait… That's not how the Web advances… it advances by using Web Standards… and improving those standards by adopting new technology when it makes sense for the whole Web.  Therefore ruining the innerHTML implementation in all other browsers to match IE's broken innerHTML makes no sense at all.

    In fact, the best option would be for Microsoft to fix their implementation of innerHTML so that it works on every element that can support an HTML sub-tree.

    Getter is broken on: All HTMLElement items

    Setter is broken on: HTMLColElement, HTMLColgroupElement, HTMLFramesetElement, HTMLHeadElement, HTMLHtmlElement, HTMLSelectElement, HTMLStyleElement, HTMLTableElement, HTMLTbodyElement, HTMLTfootElement, HTMLTheadElement, HTMLTitleElement, HTMLTrElement (are there others I've missed?)

    Fixing this bug in the IE9 release would be preferred thanks.

  12. the_dees says:

    @John Hrvatin:

    Ever since people started to be annoyed by this border around the viewport, a very similar issue has been reported as well.

    If an iframe has a CSS border applied, this vieport border is still drawn around the iframe.

    Opera, Safari and Firefox also draw such a border around iframes. However, if in these browsers a CSS border is applied to the iframe element, the initial border is replaced by the CSS border.

    This should work the same in Internet Explorer.

    Please see the currently active (but not yet reproduced) Issue #577604 (borders on <iframe> cannot be styled via CSS).

    In the report, its noted that the Issue had been previously reported while IE8 was in the making.

    Also see this testcase I created:…/171-iframe-vp-border.html


    I don't think innerHTML will be completely fixed in IE9. innerHTML is a property that's closely related to the HTML parser. And IE9's HTML parser is still the same as IE6's one (except a few improvements). When the developers implement the HTML5 parsing algorithm (like Gecko and WebKit did), bugs like this should be fixed (hopefully in IEnext).

    FWIW, HTMLOptionElement, HTMLOptGroupElement are definately missing.

    And what do you mean by "Getter were broken on al HTMLElements"? If there are specific issues, please name them, so someone can test and report them (a testcase would be ideal – a working, publically available testcase, not a link to some complaint website, thanks).

  13. the_dees says:

    @John Hrvatin:

    I forgot one question:

    Is it really necessary to keep this border for quirks and IE<9 modes?

    I could imagine this will cause some quirky view when a user switches from a website in standards mode to a website in quirks mode and vice versa.

    Also, are there really methods/properties that are affected by this border (if yes, which)? Do they report the same or different values when this border is drawn?

    Because when a user resizes the browser window, the calculations that may have taken this border into account will go wrong anyway.

    So I think completely removing this border would probably best for user experience.

  14. Baz says:


    All of microsoft's technology has been open in the sense that it has always been well documented and anyone is free, and encouraged to implement them. Additionally, your web "standards" rhetoric suggest a lack of understanding of the evolution and the role of html. HTML began as an markup language that was designed to be extensible. Indeed, it's success began in 1993 when the decidedly "non standard" addition of in-page images by the Mosaic team gave it the edge over the competing. The web quickly evolved beyond the bare extensible framework with embedded media and other features that that would come to define the online experience. Netscape was just as prolific a unilateral feature creator as microsoft ever was. The first streaming video arose such as realplayer and java plugins allowed. There was no real authorative standard making body until well after netscape had begun to sputter out, at which point apple, mozilla and oper formed WHATWG in 2004 to supercede to some extent the W3C (which had approved of many of the decisions being made by microsoft like XML) and to establish some of the first standards that would begin heading in a different direction of the W3C.  The IE teams motto of "same markup" instead of mentioning standards is a reflection of the thought that standards are important only for interoperability, and that they shouldn't be used to stifle new or different ideas.

  15. André Web Developer says:

    This looks interesting:

  16. André Web Developer says:

    And some information for HTML5 Forms: site is really great if want learn more about the future of web

  17. André Web Developer says:

    And more HTML5 examples:

  18. Sandro says:

    I don't understand why Microsoft and IE wait year and want every time to make a new browser, WHY don't make constant updates! With fix, patch, new feature, etc…. this is the only way to keep the people to use IE. Keep updating it please.

  19. Mo says:

    I just noticed that the UI in Outlook Web Access does NOT have this border (in IE7 and IE8) so that means there must be some way to disable this border in older versions of IE as well.

    As far as I can tell it's not a CSS property that can be altered, does anyone know how this is done?

    See comparison of Outlook WA and this page:

  20. bogascorp says:

    Hurray! No more regedit edits to remove the border.

  21. bogascorp says:

    Hey have you testes the latest Firefox Beta4? It has hardware accelaration on by default but it's still slow in a lot of tests. Fishtank, flickr etc

    Give it a try.

  22. Joe Marini says:

    @Sandro Not everybody wants their browser constantly updated. There are thousands of enterprises that use IE that would throw a fit if IE was updated every few months.

  23. bogascorp says:

    @Joe 100% agree. Companies rely on IE, Microsoft can't just update it and break the web inside companies. We all rely on companies to provide us with services, if Microsoft would break the apps inside the companies we would all lose productivity.

  24. bogascorp says:

    There has to be a better way to update the browser without breaking companies apps.

    Maybe a customized IE browser for Enterprises?

  25. Sam says:

    Enterprises should have their own IE if they really have to, but it should be locked down to access Intranets only, so people can't use it to browse 'normal' websites.  If employees want to access outside, they should use a regularly updated of IE, or any other browser.  The IE lifecycle is really to slow and I believe it cripples businesses in the long run, especially those that provide services on-line, because 90% of an on-line business time is spent catering to the broken needs of IE.  Imagine all the vibrant technologies we could all use if there was no IE6, 7, 8; the presence of those browsers forbids all that.  There is no point in adopting CSS3/HTML5 as long as those browsers still exist, those that do will only leave behind those stuck on IE.  A company will loose opportunities because it wouldn't access resources beyond what their version of IE would parse, people would realise that IE is a failure, and move to another platform.

  26. John Hrvatin [MSFT] says:


    getBoundingClientRect is one example, as are other event properties like clientX and clientY.  This is called out on MSDN (…/ms536433(VS.85).aspx) and other blogs (…/javascript_events).  Nevetheless, we had a similar conversation about whether to make this change in all modes because it would be an edge case for a site to break because of this.  However, we do everything we can to keep the legacy modes the same as previous versions or else developers can't count on them.  We've learned that anytime we ask "does anyone *really* depend on this?" the answer is "yes", and anytime we ask "how many sites depend on this?" the answer is "more than you'd think".  🙂

    Thanks for the bug report on the other border issue!


    In Quirks Mode, you can style this border with CSS applied to the <body>., which runs in Quirks Mode, does this.

    John [MSFT]

  27. meni says:

    Joe Marini and bogascorp, let me repeat what i said in another post:

    Your reasoning has almost killed the web. It goes like this: Hey, IE6 is from Microsoft, they would surly not break backward compatibility, so let's design a "web app" and target IE6 only!  Then we can blame someone if our app doesnt work. In my country, there are many of these crippled web sites, some banks, some govenment 🙁 These developers should be: a) punished, b) forced to learn STANDARDS and handling browser quirks.

    a) Web development is hard, nothing can be done about it

    b) If you want this kind of stability, go make a Windows app with Silverlight/WPF. The world will thank you 😉

    c) If you want a job today, learn how to "Adjust web sites to firefox" [the lame way of saying we made a mistake back then]. A lot of openings there.

    Microsoft should stop this lunacy with frequent upgrades to IE. Otherwise we'll face this same situation in the future. Corps should make Windows apps and leave us alone. [I think they kinda like the idea of the web, but they just don't quite get it yet, even in 2010]

    In conclusion, let me finish with my "Carthago delenda est":

    IEteam, can we have a statement on WebGL?

  28. wow, got so used to the border but I welcome the change.

  29. Make IE9 relevant says:

    When will MS learn not to tie IE to OS technologies? If a particular Windows version offers feature x, use it. If it does not, don't make the other leading and dominant OS users, that is, XP users suffer by not supporting IE on that at all. Your "same markup everywhere" holds no truth as "same markup" is only being offered to Windows 7 users (and Vista but that Vista support for consumer editions will end in 2012). IE9 should really have been the last version to support XP and deliver finally on the competitive standards support which sadly IE8 doesn't have. I will make it a point not to use IE9 even on Windows 7 but use the upcoming Firefox 4.0 which also uses Direct2D and DirectWrite hardware acceleration but doesn't punish XP users.

  30. Capt. Obvious says:

    meni: "Having a clue" ain't really your strong suit, is it? Is the reason you post here that the other open-source fanboys have shunned you?

    Relevant: Yeah, as the most successful software company in the world, I'm sure Microsoft has learned their lesson. I'm sure that they'll miss you among their hundreds of milliions of users. They'll probably have executive meetings about how they can win you back!

  31. Tsoh says:

    PLese please please fix the scrollbar Fitts law  problem – remove the 2 pixels that prevent you from scrolling by moving the mouse to the edge of the screen

  32. Spindel says:


    Have you tried the IE9 preview? They have removed the 2 pixels you are talking about.

  33. Blair says:

    @the_dees – the issue with the getter for innerHTML is that it does not return what you set.


    If I pass in:

    <div id="test"><input type="text" name="email" value="" maxlength="256"/></div>

    and I ask for document.getElementById('test').innerHTML I get this:

    <INPUT type=text name=email maxLength=265>

    which is NOT what I originally specified at all.  Since this mess is returned for every element there is no way to use the getter with expect-able results…. thus the getter is essentially useless in IE… thus useless cross-browser.  A classic case of IE broke the Web.

    I'll admit though that using jQuery here is much easier as it ignores the garbage IE uses and gives back usable content.  Its just sad that we need to rely on 3rd party libraries to fix IE rather than expect MSFT to step forward and fix their code.

  34. CvP says:

    nice step.

    can't wait to see IE's new UI 🙂

  35. CvP says:

    omg..just found this:…/7213…/ie9-ui.png

    cool. i wish the "tab bar" can be placed at the very top (like chrome) or under the address bar or left/right/bottom edge of the screen…as the user desires 😀

  36. Jonas says:

    The leaked IE9 UI looks very cool. But please make the placement of the tab-bar customizable.

  37. Bored says:

    And you trolls troll.

  38. J says:

    The leaked UI does indeed look good! The titlebar should not be used for UI elements (like in the screenshot) and it's really nice that no text is displayed in the titlebar. It gives the browser a really clean look and feel.

  39. J says:

    The leaked UI looks really good! I'm glad to see that you guys don't use the titlebar for UI elements like other browsers do. The titlebar is an important part of the UI and also for usability (dragging windows around). It's also very refreshing to not display any text in the titlebar, it gives IE a very clean look and feel.

  40. the_dees says:

    @John Hrvatin [MSFT]:

    Thanks fo your explanation.

    Would it help to add the 2px offset to the affected properties? Hmm, but I could imagine this would cause other incosistencies,

    But what about some magic (hypothetical) margin: -2px on the (hypothetical) browser element?

    Sigh, no easy job. I don't like inconsistent user experience, but this broder itself is bad experience as well.


    <div id="test"><input type="text" name="email" value="" maxlength="256"/></div>

    FWIW, the current IE9 output has quotes around attributes!

    <INPUT type="text" name="email" value="" maxLength="265">

    The trailing slash is of course missing because it is invalid in HTML 4.01 (it will be allowed in HTML5 though).

    As I said the parser has got some iprovements. However, I don't expect the remaining issues to be fixed before the actual HTML5 parser is implemented.

  41. David S says:

    @Make IE9 relevant

    One good thing about moving directly to Windows 7 is that you kill 2 birds with one stone. Namely IE6 and IE7. Cross browser development is a pain because of these 2 browsers. Windows 7 comes loaded with IE8 which rarely need special treatment in development. I think to allow technology to move forward MSFT must nip those 2 browsers in the bud and there is no better place to do this than at the OS level.

    It would actually be cool to have IE9 preloaded on future releases on Windows 7. By the time Enterprises upgrade their OS, IE9 would be the lowest common denominator. Development cost across the board would greatly decrease due to less cross browser time investment.

  42. Stu says:

    @David S

    Couldn't agree more.  At the company I worked for we used to integrate/slipstream IE7 into the XP setup media to have it installed by default: sadly, IE8 didn't support this officially.  Be great to have a future 7 release with it included (perhaps offer a version of by then Service Pack 2 that includes it; and one that doesn't).

    @Spindel (RE: Tsoh)

    In IE the scrollbar scroller is exactly 3 pixels away from the edge, so are you able to confirm this is definitely fixed in the current Platform Preview?  Great, if so.  It is an odd issue seeing as no other application (Notepad, WordPad, Paint, etc.) have the issue, and I recall a blog entry where the Microsoft Office team said Office applications and any windows they open always have scrollbars designed to adhere to Fitt's law.  They specifically mentioned how it used to be possible to overshoot the old Start button in Windows 95 and 98 by one pixel: so how no-one fixed the issue in IE until now is odd.

  43. Spindel says:


    Yes, the scrollbar scroller in IE9 platform preview is 1 pixel away from the edge.

  44. Stu says:

    Just my two cents while I'm here too and appologies for being off-topic but things I'd like to still see in future IEs:

    1   Download manager – current dialogs are fine but why not use the sidebar (currently used for Favourites, Feeds and History and add a Downloads tab).  Retain the option to use separate windows.

    2   Direct downloads – i.e. stop downloading to a file first in the IE Cache (Temporary Internet Files folder) and just save it directly to the location.  Or, save it in the cache then do a move operation instead of copy.  Saves time and the requirement of having to have twice the available disk space of the size of file you download!

    3   InPrivate Filtering – improve it by allowing entires to be manually added (ala Restricted Sites zone) and make it possible to block entire domains, not just elements via filename of specifics in a URL.

    4   Options dialog – needs reorganising.  I'm sure some options could be removed, mostly in the Advanced tab.  I mean, "Empty Temporary Internet Files folder when the browser is closed" is an option here: but then doing this is also possible by setting "Delete browsing history on exit" on the General tab and choosing "Temporary Internet Files" on the Delete Browsing History dialog.  Never been sure if either method does the task any differently.

    5   Font – make a full-Unicode font the default for websites that don't specify one.  So many Wikipedia pages I've viewed in IE and seen squares instead of characters.  At-least maybe introduce a "Unicode font:" option as well as Rich and Plain text.

    6   Discussion Bar ("Discuss") – can this feature be removed?  Does anyone know if this still does anything?  (View > Explorer Bars > Discuss).  Wasn't it was derived from a Microsoft Research project called Community Bar?  And I don't think that can be installed after IE6.

    7   Favourites – perhaps a bug and not specifically for the IE team but I have encountered the issue where a Favourite file in IE loses the ability to work — notable sympton of such a case being if you call up the properties dialog for an affected Favourite file it won't have the "Web Document" tab.  Apparently it's related to Windows Search holding the file open and causing corruption.  Perhaps best to not include the Favourites in the Index by default, if so?

  45. EricLaw [MSFT] says:


    #5> This is often caused by a site failing to specify its text encoding properly, and hence the characters are actually interpreted incorrectly. A proper CHARSET declaration will resolve this.

    #6> This is a Microsoft Office 2003-installed Explorer Bar plugin. It's harmless unless you happen to show it, but you can uninstall it or disable it if you like. It's used when you have a server with the Office Server Extensions installed, and allows you to collaboratively markup a document. It was neat back in the day but has mostly been replaced by other features.

  46. Sam says:

    PLEASE make the scrollbar flush with the edge of the screen – every other browser gets this right!!!!

  47. Harry Richter says:

    @ meni

    We agree on not agreeing on many points.

    One thing should be corrected though, for you, and for the one you took the quotation from:

    "Carthago delenda est" is plain wrong! I know it is used in this way by the unknowledable, but the correct quotation of Cato is "ceterum censeo carthaginem esse delendam", which is reported by Plutarch, abeit in ancient Greek: " Δοκεῖ δέ μοι καὶ Καρχηδόνα μὴ εἶναι" (but it seems to me, that also Karthago should not be).



  48. Stu says:

    @EricLaw [MSFT]

    For the automatic-select of a font to work does the option View > Encoding > Auto-Select have to be on?  It's off by default, if so, so this may need changing.

    * * *

    Couple more points to add to my others above:

    Tabs –  in IE7 it used to be possible to allow new tabs to open next to the current one (image you have three sites currently all in tabs in one window: 1, [2], 3.  You have two selected.  The next link you middle-click on opens to the right of 2, so you get: 1, 2, [4], 3).  In IE8 new tabs always appear only at the far end on the right.  Irritating when more tabs than fit the window are open as you have to scroll to the right and lose your place — or set the Ctrl-Tab switching to "use most recent order" under Advanced options.

    Cache Name – one I can't see changing but still: shouldn't that "Content.IE5" cache folder name finally be updated? 😉  As IE9 is only for Vista and 7 and I believe both of these support NTFS Junctions why not have "Content.IE9" and create a junction called "Content.IE5" to it for legacy applications?  Always find it amusing using IE8 but having a folder referring to IE5! :p

  49. Selcuk says:

    Any news from "Blurry – Bleeded fonts" problem as mentioned…/benefits-of-gpu-powered-html5.aspx <- Seems Preview 4 even does not address this…

  50. Selcuk says:

    Sorry for double-post. It seems problem is with Direct2D since Firefox 4 beta 4 is also having the very same problem that use that feature. Hope Windows team working on that…

  51. Selcuk says:

    BTW, because of that border, main Javascript frameworks having problems with offsets… Such as if you try to place a div on parent divs border, you get 2 pixels wrong…

  52. Adam says:

    I for one am happy that Internet Explorer 8 (as far as I've tested) supports these CSS elements when FireFox 3.6.8, Opera 10.61, Safari 5, and Chrome Canary Build 7.0, don't.

    List-Style-Type: None, UL{Margin: whatever), and Display:Inline;

    Keep up the good work IE team!  Hope to see IE9 soon.

    PS.  Its been how long since IE8 came out, and its competition still can't process better looking fonts then IE8.

    (Direct X in IE, ROCKS)

  53. wendy says:

    @Adam – I'm not sure what rock you've been living under but Firefox 3.6, Safari 5, Chrome and Opera all kick IE8 to the curb with a vengeance.  IE is the only browser that doesn't render the margins on lists correctly and is known to cause invalid numbering on Ordered Lists because the numbers get truncated off the left edge of the screen/block.

    ClearType in IE was a failure.  Blurry fonts that look bold when they are not is not a good thing.

    I have no idea what CSS you claim IE8 supports that all the modern browsers do not but I'd love to see your test cases.  I have several dozen JS and CSS tests that IE9 prev. 4 still fail on that all other browsers pass (minus 1 JS bug in Opera)

    IE9 has massive speed improvements – which if coupled with a completely re-vamped UI will actually make it a contender against Chrome and Firefox however to claim that IE8 is ahead of current Firefox/Chrome browsers is absolutely beyond ludicrous.

    @John Hrvatin – speaking of browser chrome… the thumbscrew on any block element with overflow:auto; set is not correctly rendering the top / bottom lines of the button in IE9 prev 4 on Vista (maybe 7 too) when the thumb gets smaller than about 14px;

    @the_dees – woot! you are correct at least now the attributes are quoted! – the tagname is still wonky and the trailing slash is still committed regardless if it was set but thankfully this is seeing some fixes.

  54. Civility, Please says:

    "Wendy"– it's really not polite to insult other posters and retort with claims that lack evidence.

    <<IE9 prev. 4 still fail >>

    What are the bug #s on CONNECT so that others can validate your claims?

    <<ClearType in IE was a failure.>>

    ClearType in IE is a success. See, anyone can play!

  55. hAl says:


    You can find several examples of the css that IE8 supports and others don't in the 7000+ test submitted by MS to W3C during IE8 development.

  56. Evan says:

    @Civility, Please – @Adam made claims that IE8 supported CSS elements that all the other browser didn't. There were no tests provided there and @wendy called him on it.

    Did wendy's reply include tests? no but bashing her because she didn't provide them isn't right either.

    As for CONNECT #'s you seem to be under the mistaken impression that that is the place where developers/users will file all bugs.  It most certainly isn't. I know I used to use it but when there was no respect given to the developers and the reports they filed I soon bailed – never to return. (rule #1 of building a community is don't tick off your community and play games with the free QA/Tech Support they provide you)

    I'll bite my tongue on Cleartype because although I think the concept is great I think the implementation of anti-aliased fonts in Windows is not as good as other operating systems.

    however back on CSS support – I'm happy that IE is really pumping out the support here but i'm in the camp of developers that could care less about fine-tuning font jazz as I'd rather have a functional JS DOM that works where all the 5-10yr old bugs in the core of IE are fixed so that we aren't doing hacks forever just to make the most basic pages work in IE.

  57. the_dees says:


    Could you please provide test cases for the issues you mentioned? (If they still apply to IE8).

    > IE is the only browser that doesn't render the margins on lists correctly and is known to cause invalid numbering on Ordered Lists because the numbers get truncated off the left edge of the screen/block.

    > I have several dozen JS and CSS tests that IE9 prev. 4 still fail on that all other browsers pass (minus 1 JS bug in Opera)

    Note: I'd also like to see the Opera bug; Opera unfortunately is often an IE copycat /sigh

    > the thumbscrew on any block element with overflow:auto; set is not correctly rendering the top / bottom lines of the button in IE9 prev 4 on Vista (maybe 7 too) when the thumb gets smaller than about 14px;

    Words and descriptions are often not very helpful, because the web is a place where people often do not speak their native language. Test cases are the definite helper to understand a problem in the spec/iomplementation.


Skip to main content