IE and CSS "Compliance"


I was surprised that when I got back from vacationing in Hawaii to read a Slashdot post by Jeff Reifman about an article Paul Thurrott wrote…last year.  I read Slashdot regularly, and it didn’t make sense; not only is this old information, but it’s not even factually correct (which I’d like to think is a requirement for Slashdot posts).


Jeff did later figure out the article was from last year, and edited his post. And others have pointed out that Paul has made several posts since last year about IE7.  I’d recommend reading them, such as this one from (THIS) June reviewing IE 7 Beta 3, which seems to indicate Paul thinks we might be heading in the right direction.


As for IE’s CSS compliance, I’d love to have a honest, straightforward, unbiased statement of exactly where we (and other browsers) are – despite the fact that I know we would be behind today.  Unfortunately, I don’t think that exists; any “statistical analysis” of support is by nature biased.  How do bugs count against you?  For example, one of the big developer complaints in IE6 was that we didn’t support “background-position: fixed” on non-BODY elements – but we passed the W3C CSS1 test suite, because we supported it properly on the BODY element.  And we support the ‘float’ property, and pass the set of tests in the CSS1 test suite – but even in IE7, we have some bugs in interacting with width and height, which is one of the contributing factors to us not passing the Acid 2 test suite.  Your personal view may be that that’s as bad or worse than not supporting ‘float’ at all, but a lot of designs in use on the web today require floating.


The site that Jeff’s post referred to (developed by David Hammond) is weighted in a way that makes IE in particular look bad (not particularly surprising, given some of the other pages on the site) – for example, it dings our support of each CSS property, one at a time, simply because we support a proprietary “CSS expressions” feature; similarly, the fact that we don’t support the ‘inherit’ property value across all CSS properties is counted against us one for each property, not as a global feature as it would be implemented and used.  Most unfortunately, there are no more details on many of the problems David encountered, or test cases that my team can test against.  When I contacted David a year or so ago, he couldn’t give me any further details, so I’m not even sure how we make progress against that site.  Solid test cases we can access and bug reporting would help – which is why we have a public bug database.  I’d also like to turn comments like “IE 6 and IE 7 failed to render even basic columnar layouts in CSS” in to solid work items, but I don’t understand the problem – much of the web uses columnar layouts that work in IE.  Is this a request for CSS3 multicolumn support, or a narrow bug with the way you’re trying to do columnar layouts?


The one thing that really burns my personal toast is that we’ve been working hard to improve our standards support in IE7, and I believe it is simply wrong to think that we’ve only moved the needle 2%.  In fact, we prioritized IE7 around 3 things – security, end user experience, and standards improvements in the platform.  When I look back at the work my team has done in the platform, we have done only these things.  No proprietary features added, just standards improvements.  (Look forward to an upcoming IEBlog post from Markus Mielke listing out the CSS changes we’ve done in IE7.)  I feel that we’ve addressed the biggest problems and shortcomings from IE6 for web developers and designers, and we’re hard at work shipping IE7 and getting ready to doing it again in the next release.  As I previously stated, our goal is to make the lives of web developers better by improving standards support in IE.  I think we’ve done a lot in IE7 to do just that, and I’m looking forward to doing even more.


But now I’m feeding the trolls, so I’m just going to go play my newly-acquired ukulele and chant under my breath SomePeopleGetIt” (possibly to the tune of “Tiny Bubbles.”)


-Chris


Comments (175)

  1. Internet Explorer 7 beta 2 (which was labeled "layout complete" and was thus the logical release to begin testing) introduced some pretty nasty regressions that beta 3 then fixed. In the last day or so, I have updated some of this information that it set IE7 apart of IE6 by a total of 4% in CSS 2.1 support. This is about as much as Firefox and Opera each improved in their latest versions, so this isn’t some sort of gross underestimation, it’s simply showing just how big and complex the CSS standard is.

    I recently write a blog post (http://www.webdevout.net/tidings/) addressing the claim that you guys haven’t done enough work in IE7 and defending you guys, so it’s kind of upsetting that you make me out as someone who is willing to twist the truth to make Internet Explorer look as bad as possible. In my blog post (which at this moment is down due to my site being frontpaged on digg), I laid the blame not on Microsoft’s current actions, but on its past actions: adding all of those proprietary features and then abandoning core engine development for so long. If the Gecko developers did the same, Gecko would also be terribly behind the times and would have a heck of a time trying to catch up.

    I honestly don’t know if it’s possible to catch up, considering that developers of other browsers are likely willing to put in just as much work as you guys. But I do believe that you guys are really trying now and I have seen some good results. I’m not going to ignore where IE is lacking support ("inherit" is a very important fundamental feature) but my goal is not to make one browser look better than another. This is a resource that I use myself, and I want every bug in every browser to be noted right where I expect to look.

    The main problem is that IE does things wrong at such a fundamental level that there are bugs stacked on top of bugs. Many CSS properties have multiple bugs in them, and if a new version fixes one of those bugs, the others still warrant an "I" rating for the property value. I’m sorry, but I don’t know of a more objective way I can rate features at that low of a level.

    Back when you initially contacted me, my resource was at a much more primitive stage and the site pretty much took data from other similar resources along with some personally tested information. Since then, I have expanded the tables and tested *everything* personally, and I believe I am very specific in the descriptions I give for the bugs. If you have a question about something, please ask me. If I can’t give you a demonstration of exactly what is wrong, I will correct my data.

  2. Oh, and in regards to the "inherit" property value, you ask why I list it for every single property when it’s obviously a "global feature". Well it obviously isn’t. "inherit" is supported for the "direction" property, and as far as I can tell nothing else. So no, it doesn’t seem to be a global feature, it seems to be something you implemented on a property-by-property basis.

    The upside to how it is listed is that if you add support for "inherit" in some version of IE, that would be an instance percentage boost in my tables, and I feel rightly so. I have listed "inherit" with every property since just about the beginning of my tables and I see no reason to single it out as something different. It is, after all, listed individually with every feature in the specifications.

  3. Austin Liu says:

    As a web developer, I can testify that making sites work with IE easily consumes the bulk of my time, almost as much as coding the site itself (which is rather easy with tools such as Dreamweaver). Cut it out! Microsoft doesn’t own the web, and it should stop arrogantly acting like it does and start playing by agreed upon standards.

    IE’s shoddy security, long time since the last update, and the apparent ignoring of web standards is unacceptable; all this simply goes to vindicating the anti-trust lawsuit against Microsoft: a monopoly has no incentive to strive for excellence (unless the monopoly has an inner driving passion for it, like Apple has, though it is a "monopoly" over its own little market).

    Make web standards compliance a priority! I’m not going to ignore such a large segment of my target audience as Firefox + Safari, and Microsoft doesn’t even make IE for Mac anymore, though that market segment is rapidly growing, and has a lot of disposable income.

  4. thekak says:

    When it takes you over 5 years to release an OS, let a lone a simple browser… why not just do it right this time?

    Just make it work like FF and be comliant ffs. Don’t try and be fancy, make something that actually works. Proprietary is lame. Stop it.

  5. Dave says:

    Listen Microsoft, I’m sick and tired of having to adapt all my designs to your buggy browsers. If you can’t do it just integrate Mozilla in there and stop taking advantage of your position to make the rest of us waste our time to accomodate you.

    There I said it.

  6. Scott says:

    Wow, this is one ugly blog.  Why do you have a picture behind your menu?  It makes it really hard to read.  Is this the same mentality that went into the IE design?

    BTW, the term Albatross really applies to what IE has been for developers over last 9 years.  Was the reference intentional?

    Albatross:

    "sometimes used to mean an encumbrance, or a wearisome burden." –wikipedia

  7. Brian Van says:

    I give credit where credit is due to the IE7 team, and they seem to be doing a great job in progressing to their goals in building a better product for the user. But let’s be realistic… if you’re far from standards-compliant and up-to-date with CSS support, then you should get dinged for it.

    I know that most of the complaining results from past mistakes in IE’s overall design, but it seems that IE7 is going to keep making many of those mistakes. And for what purpose? If you believe that standards-compliance is beneficial to the user, then build the standards into the browser. If you don’t think standards-compliance will benefit anyone, then do things your own way, and if anyone complains then you’ll have your moment to explain why your way of doing things is better. But if you claim that you’re trying to be standards-compliant and you release a new browser with all the same bugs and standards-holes that IE6 has, then you should expect to be criticized among the designer community. It’s just the way it is. Regardless of the fact that the source article was a year old and was factually incorrect and not up-to-date, there’s still a general problem that exists no matter how much closer you are to the standards-compliance goal. You’re still not close enough.

    I totally sympathize with your development team for all the work they must to do try to stay competitive in this application environment. Believe me, I know that you’re attempting something very difficult and that you can’t keep everyone pleased. But as a web developer, I don’t want to keep running afoul of nasty, ugly suprises when trying to implement features that my clients EXPECT, features that they’ve seen implemented in the sloppiest, most convoluted ways on other businesses’ websites. My CSS development process is competing against table template jockeys across the world, and it’s very difficult for me to justify proper CSS practices when I’m spending hours trying to figure out how IE6 and IE5 are messing up my CSS layouts and the table template jockeys don’t have to think at all. You cannot do anything about versions 1-6 at this point, but  from this point on I’d rather clean hacks out of my old websites and require IE7 for browsing as opposed to supporting a new browser with a couple of big bugs fixed and a couple of big standards violations still there. Ending the browser-hack culture would be your biggest possible achievement as a development team, and it’s almost pointless to shoot for any softer goals if you want the support of the developer and power user communities. The disruption of eliminating these long-standing bugs and standards violations now would not nearly approach the cumulative pain of having to continue to design "workarounds" for basic things like alpha blending, float positioning and precise box-widths. (again, things that are expected without exception by my clients)

  8. BTreeHugger says:

    I would like to thank the IE7 team for their honest efforts, and hope that Microsoft gives them more power over the project next time. It’s about time you were given the chance to do something worthwhile with IE, since it has not really been worth using since IE5 in terms of scaling up with features.

    As with most developers I hope you managed to get the basics working, like HTML and CSS 2, so that the world can move on like they did from Netscape 4. It is unfair that most of my audience has to download a separate stylesheet, and that I have to pollute my work with extra markup.

    I know fixing all of these things is unreasonable, since you have to catch up in such a short timeframe, but please get your superiors to stop focusing on other proprietary technologies for a while, just long enough to bring yourselves up to at least status quo.

    Thanks, good luck, and keep up the work.. we know what you guys are forced to work with/on, and realise that you have a finite amount of hair and time to lose.

  9. Chris says:

    Our organization’s website is built using standards-based CSS. It is driving our users mad, and our IT staff crazy how long it takes IE6 to render those standards-based pages.

    I have simply told our users to use Firefox until IE7 ships. Our user base is now around 3 thousand. Please fix IE7 to handle code such as CSS correctly. If you don’t, I regrettably cannot in good conscience recommend IE7 to our users.

    Thank you.

    Chris

  10. fanboy says:

    We’re not asking for much, we’re not asking for total 100% css compliance, we’re after more css improvements. The CSS improvements in IE7 were only minor fixes from IE6. Do we have to wait for many more years for any decent changes? *sigh*

  11. Andy Mabbett says:

    This page has 67 HTML validation errors. Its CSS has 7 errors and a number of warnings about things which will affect accessibility.

    The left hand menu is unreadable.

    If you can;t get these things right, or don;t care to, how can you hope to comply with standards elsewhere?

  12. Tim Wouters says:

    There’s one thing all you yay-to-nay-switchers out there seem to forget.

    A 4% increase in standards compliance may not seem like much, but it DOES address the most common IE6 shenanigans that have been bugging us for years (see positioniseverything.net). So our everyday development experience WILL get better.

    And it shows already. Try the thrusted "build for the best, then fix for the rest" method on a new site project. Preview in Firefox, Opera, Safari AND the IE7 beta. Some pixel jogs aside, all things will look pretty decent. Now take a look with IE6 and you will see just how much IE has improved.

    Specs are nothing. Execution is everything.

  13. Gaspar says:

    I think IE team now is in the right way. IF all follow the rules and try to do exactly like they are described then would be easy to all development guys.

    CSS have some issues like vertical-align, in some ways height could be a problem too and if all follow different ways then would be more difficult. who will be wronged is all web and the users, and i think this all about users.

    Congratulations, Chris and team

    Dont forget that nobody can please all people.

    Sow go on

  14. Chris says:

    Until IE behaves like Firefox, Opera and Safari, (i.e. I am free to design within the limitations of CSS and not the limitations of IE) then you will continue to be hated by web developers.  IE has spent years and years doing *nothing* and deservedly garnering bile-inducing hatred from web developers.

    Your smug "some people get it" arrogance is exactly why IE is a piece of crap.  IE6 languished for years in its diseased state because of arrogance.  Only after you began losing marketshare was some of Microsoft’s arrogance dissipated long enough to get off its bloated ass and do something about it.

    There is one and only one test to determine if IE& is worth a damn and it is this – does our browser comply with standards?  Do pages rendered in your browser look identical to pages rendered in Safari, Firefox or Opera without any stupid hacks?  If you can say yes to that fact, then your job is done.

    You know this.  Please don’t insult our intelligence by looking us in the face and ask "what’s wrong with IE?".  Microsoft has spent years doing ab-so-lute-ly nothing with God-awful IE 6.  Surely, you can’t expect instant goodwill from web developers just because you’re working on a new version of IE?

    Until IE is reasonably standards-compliant you can expect no quarter from us.  We’ve spent too much time designing web sites that will work for CSS and then work for IE.  We’ve destroyed budgets, pissed off our bosses, irritated clients and made development more expensive because of your little (Rosemary’s) baby.

    We web developers fully aware that when your sweaty, monkey-boy of a boss was leaping around onstage  chanting "developers! developers! developers!" he most certainly wasn’t thinking about us.

    Despite the tone of this post, I truly and earnestly do wish you the best of success with IE7 and look forward to an excellent browsing *and* developing experience with it.

    You’ll just have to pardon me for not holding my breath given your company’s past behavior.

  15. Chris Wilson says:

    @David: I’m sorry that you took my comment to read “[David] is making Internet Explorer look as bad as possible”.  That’s not what I meant.  I think any non-binary analysis of “how much” CSS (or nearly any complex standard) is supported by a particular implementation is going to be biased.  I disagree with some of the biases that show in your weighting; as I said, I think taking a discount for every property because we support the proprietary dynamic properties weights against us fairly heavily.  I’m not sure I understand your interpretation system (http://www.webdevout.net/browser_support.php?uas=IE6-IE7-FX1_5-OP8-OP9#interpret), but it would seem that just that “incomplete” for every property would tend to normalize us to 50%.  Perhaps I’m misunderstanding how the numbers are arrived at.  Additionally, as I’ve said, without this being tied directly to public test cases that you simply click through – e.g., as an individual I should be able to take a copy of your site, click through the tests myself, and get identical numerical answers – I don’t even know how to make consistent progress against this.

    As an aside, I also disagree that “inherit” is a very important fundamental feature.  We implemented it for visibility (not direction) – because it has significant implications in how you can trim out branches of the display tree (you can have visible sections inside a visibility:hidden branch, which is not true of, say, “display: none”, and we needed to make our model support it – “inherit” *WAS* the initial value for the ‘visibility’ property in the CSS level 2 Recommendation, though I see it’s been changed in CSS 2.1 (to what end I don’t know).  I also seem to recall (though this was a long time ago, and I could easily be wrong) that “inherit” was initially NOT on all properties, but added after we created the visibility property with an inherit value.  For other properties, though, it seems mostly a trick for getting back the default values of normally-inherited properties, prior to some other stylesheet in the cascade being applied (for example, getting the font size to inherit through HTML table elements).  Neat, useful, etc., but not a universal panacea or anything.

    Look, I personally need to be past laying blame, but I agree with you that where IE is now is a result of Microsoft’s past actions.  Your analysis, knowingly or not, feeds the perception that we aren’t doing anything about it, because your analysis was widely interpreted as “Microsoft only cared to move the standards-compliance needle 2% (4% now) with IE7.”  I believe that is a wildly inaccurate evaluation of how much we have cared about what web developers and consumers need, and how much we have made their lives better in IE7 (from IE6).  That’s all.  Again, I apologize if you thought I was being dismissive.

    @Austin: I’m internally search-and-replace’ing all occurrences of “IE” in your comment (other than the Mac IE one) with IE6.  Would you care to make a comment on IE7?

    @thekak: I would think the work my team has done in IE7 would prove that we tend to agree with you that proprietary is lame – as I said, we spent our resources in this product cycle making our standards support better.  David even said this in the post he mentioned.

    @Scott:  I don’t give a flying leap if you think my blog is unattractive.  I’m not here to be pretty.  Use an RSS aggregator, and make it look however you want.  I can recommend one.  🙂  And if you’d looked back at any of my previous posts, or looked in the box with the oh-so-ugly image, you’d know that I understand EXACTLY what an albatross is.

    @Fanboy, et al: You’re invited to give us more input to help prioritize the issues.  We have a public database, you can vote on the issues.  Please do.

    @Andy Mabbett:  I likewise do not care very much about the validation errors that the crappy blog software on this site uses.  I hope it will be improved; I’ve made such a request; I’m not gonna get my panties in a twist, nor am I going to spend any of my time fixing it by hand.  AFAIK, the HTML version of this blog is usable from IE, Mozilla and Opera; and the RSS version validates.  That’s good enough for me.  That’s the last time I’m saying that, rather than simply deleting such comments.

    Tim:  Thanks.  That was precisely the point I’ve been trying to get across, elegantly said.

    “Chris” (sigh): And again – I’m talking about TODAY, not the last five years.  You’re free to complain about the past; I can certainly understand why; I’m not the person to blame, in fact I have a longstanding reputation as a heretic in the company, who is in the strongest position to say "I told you so"; but my "some people get it" statement was that some people get that we are moving in the right direction (at last, you can insert privately) in IE7 and beyond.  You might wish to actually read those links, and note that none of them is an unmitigated "IE ROXXXXX" post, but acknowledgment that we are doing right.

    As for “does your browser comply with standards” – how do you test that?  How do you confirm that Opera “complies with standards”, or Mozilla, or Safari?  Do you (apparently alone of all people) never have to hack for any other browser?  I know that you hack far more for IE6 than anything else; I’m interested at this point in IE7.  I also know we’re behind in this area, so you don’t need to inform me.  I didn’t ask you “what’s wrong with IE”, because I’m not a moron.  I have asked, and will continue to ask, for help prioritizing where we need to go; I will also expect appropriate recognition that we are moving in the right direction.  Not that we’re done.

    -Chris Wilson

  16. Brett Merkey says:

    I treasure my right to excoriate IE on a daily basis but somehow I get the impression that much of the complaining comes from those with a very narrow perspective.

    I have been using CSS exclusively for positioning since 2000. That is when I demonstrated to management, using IE5.0, how much faster development would be and how much cheaper operations and maintenance would be for our browser-based applications using CSS and W3C-recommended coding practices.

    Much to my surprise, management agreed and our product dev teams since have certainly benefited by pushing CSS to the max.

    Let’s keep the pressure on software companies to improve Web software but let’s focus on actually using the technology as we find it at any stage. It is all very usable and valuable *right now* in IE — and has been for years.

    Brett Merkey

    http://web.tampabay.rr.com/bmerkey/

  17. Tim says:

    While I don’t mean to discount the imporovements the IE7 team has made to imporved standards support, it really strikes me as lazy.

    Microsoft used to stand for innovation, but lately it’s been playing catch-up in almost every regard. Instead of taking this LONG overdue update to ‘do it right’, IE7 simply fixes the largest and most dire standards bugs then finds excuses for not making other improvements.

    Microsoft has the power and money to not only meet a FireFox-level compliance, but surpass it. It’s pretty disappointing when the most powerful company in the IT business can’t even keep up with an open source project.

  18. Thomas Tallyce says:

    Many web-devs are genuinely excited to see that real progress is apparently being made to the Trident engine.

    However, literally millions and millions of person-hours must have been wasted trying to code around Trident/IE6’s problems, and you can’t expect people to simply be forgiving just like that.

    The problem is not with the individuals in your team but the *business decision* by Microsoft to let the browser stagnate.

    Given the really nasty PIE-style bugs in IE6, it seems fairly clear that a large amount of refactoring to a rather broken engine was presumably necessary to get MSHTML into a maintainable state.

    The team should be contratulated for the clearly huge task of getting IE7 out the door, and in particular for making it an automatic upgrade. Later this year a very significant percentage of web users will all be using web browsers that won’t require the sort hackery we’ve had to get used to for the last 5 years. Let’s at least be thankful for that, and hope that MS will use the base of IE7 to build on further work, much as Chris and others have said they will.

    What would really help now in silencing many of the critics here is to start to publish an expected roadmap or at least outline of the CSS/standards areas the IE team plan to work on for IE7.5/8.

  19. Klaus Hartl says:

    "I know that you hack far more for IE6 than anything else; I’m interested at this point in IE7."

    Chris, I’m sorry, do you really want to know that? I can give you an example from the real world. Have a look at the latest site I’m working on: <a href="http://plazes.com">http://plazes.com</a&gt;, i.e. at the style sheet that is included for IE only (hidden by Conditional Comments).

    As soon as IE 7 Beta 2 was told to be "layout complete" I started to test in IE 7 as well. And it appeared that – though you fixed the most infamous bugs in IE7 – I have to apply the same hacks for IE 6 and IE 7 in ca. 90% of all cases. The other 8% even worse I have to apply different hacks to IE 6 and IE 7 by using special hacks (thank god you implemented advanced css selectors in IE 7) and the rest IE 6 only.

    Most of these bugs are related to elements having layout or not, which makes me starting to think that something fundamentally is wrong with IE’s rendering engine. I really had some hopes for IE 7, but sadly enough it now seems to me that nearly nothing changed from a practical standpoint. I can see no use of advanced selectors if the rendering is still broken.

    And by the way, the simple attribute selector <code>[type=’text’]</code> makes IE 7 crash.

    And we haven’t yet talked about fixes you did for JavaScript. Ah no, there aren’t any (ok, besides of the native XHR object support – how native is it anyway?).

  20. The question really should be whether if you had worked on standards since IE6 came out, would you be where Gecko, WebKit, KHTML and Presto are today. The fact that you didn’t has been you’ve had to play catch-up, which, in the world of standards, isn’t easy. You’ve made a fair amount of progress between IE6 and IE7 for the time it’s taken.

    The real question is why do you continue to improve and use Trident, if Tasman 1.0, which came out over 2 years ago, has better standards support anyway? If IE7 isn’t integrated with the OS, what’s stopping you?

    Earlier today I was having an issue with a website I was working on in a single browser. Firefox. Safari, Opera, Konquerer and IE all rendered it as per the CSS spec. Firefox did not.

    All browsers have bugs, not just some.

    I have huge amounts of respect for anyone able to make a browser do anything anywhere near the CSS specs, as I can barely imagine how hard it must be to implement.

    @Tim: I doubt that. The Mozilla Foundation is where it is because it’s had huge numbers of developers contributing (far more than IE has had working on it, certainly) and it never fell behind the standards.

  21. @Chris Wilson: Thanks for pointing out that IE supports "inherit" on the "visibility" property. I must have missed that one. It does support it on the "direction" property as well.

    In regards to the weighting system, as I said, there isn’t really a more objective way to weigh three arbitrary bugs versus two on a given feature, so any level of incomplete support on a feature that can’t clearly be broken down into further areas of support are simply given a round 50% when it comes to the calculations. If you can suggest a better way of doing this, I’m all ears.

    I think the biggest issue with the percentages is the fact that features are not weighted based on real world importance. I have considered making some kind of public rating system that would influence the weight of a feature based on the number of votes and the average importance rating. That’s on the shelf for future work.

    In regards to the importance of the "inherit" property, I have done plenty of work on systems designed only to support modern layout engines (that is, without regard for Internet Explorer) and I find myself using the "inherit" value a lot. In a system I’m currently working on, the "inherit" value appears for the cursor, font, color, and line-height properties for various reason. In most of these cases, especially since it’s a fully JavaScript-driven web application, it’s *possible* to do without the "inherit" value and get the same result, but the value saves a lot of extra code. That’s one of the kinds of things we web developers are always looking for. CSS is a very elegant language, but IE is lacking so many of the features that makes it so. In that respect, IE7 does have excellent improvements, particularly in regard to selectors. But there’s still a ton of work to be done.

    One thing that concerns me is all of the basic CSS parsing bugs and the fact that every single basic selector, combinator, pseudo-class, and pseudo-element has at least one known bug in IE7, while Firefox and Opera have perfect (as far as I have been able to see) support for most of them. Why do the universal selector, sibling combinators, and :first-child pseudo-class act like comments and doctypes are elements in IE7? Why don’t attribute selectors properly recognize attributes in minimized form? Why does "!important!" act like "!important" rather than failing the property? Why can’t comments exist between combinators and simple selectors? To me, someone who is obviously unfamiliar with Trident’s source code, it seems that Trident has serious problems at a fundamental level and is in need of a major rewrite. I feel that if you guys were to begin working on such a rewrite, it could mean great things in the future.

  22. OK, try my site and tell me why the left column is sometimes pushed under the content ?

    IE6, FF, Safari, Opera 9 handle this site ok, but not IE7 beta 3.

    shame !

  23. Arpad Borsos says:

    IE’s support for CSS has indeed improved, but there was no work done on DOM and XHTML (as application(xhtml+xml) support.

    I recently worked a lot with DOM Events and its just a pain that IE does not support addEventListener… Ok, there are plenty of addEvent functions around but IMHO these are only workarounds. And I prefer the most standards compliant way, which would be addEventListener…

    And why doesn’t IE support application/xhtml+xml?  It doesn’t even have to treat it specialy. I don’t care for embedded SVG in IE as long as the site displays at all instead of opening a download dialog.

  24. Mark Reeder says:

    First off, let me begin by thanking you and everyone else working hard to make IE7 better. You’ve squashed a number of particularly frustrating bugs and will make my life a little easier.

    I think a lot of the frustration that people have is two-fold:

    First, many people look at MS and see a huge company and think "If they *really* wanted to support standards, they could just devote more resources to it and get something done really fast." I know that’s not exactly how things work (especially with trying to adhere to a spec. that is unclear/unworkable in some areas), but I think a lot of people feel that way. I applaud the efforts that have been made and hope that things will continue to move in a positive direction.

    Second, just because IE7 is coming out now doesn’t mean we can leave behind everyone that doesn’t upgrade from IE6. In the short term, this means that we have to not only worry about IE7 (which introduces some new problems), but we still have to make sure that IE6 functions properly. It’s probably going to be years before we can drop IE6 support. If IE7 hit even the low mark of the intersecting support of Firefox, Opera and Safari, most web developers would be cheering you instead of being frustrated about yet another platform to support.

    There are going to be some serious growing pains when IE7 gets pushed out as a high-priority update. A lot of sites are going to break (yes, ones that work just fine in IE6, Firefox, Safari and Opera). This brings up another gripe – in order to do the necessary testing, we’ll need yet another machine (physical or virtual) to test on as there is no way to run IE6 and IE7 side by side on the same system. However, even with the pains it’s going to cause, the direction you’re moving in is the right one. Thanks for what you’re doing – keep it up.

  25. Chris … what are the plans for updating IE7 when it’s released?

    I know that the focus has been on creating the new product, but will there be updates that address the standardization issues? I remember back in the day IE4.x, etc. would receive updates on a regular basis.

    Thanks!

  26. Alfonso says:

    Good joke the one about the public bug database.

    Yes, for a  moment I (foolish) thought it might be real and that from now on we could see how the bugs are fixed, search our problems and find some confirmation that this is a known problem and maybe even find a workaround.

    So I filled a pair of bugs, they have existed from as long as I can remember and wish that this way I could learn when they would be fixed or have some insight about a solution that doesn’t create any new bug.

    But shortly after both (as lots of other bugs) were closed:

    "Field Status changed from Active to Closed

    Field Resolution changed from Not Set to Won’t Fix"

    So how the hell are we suppose to fill bugs if you just close them?

    I could understand if you changed the status from Active to Future or something like that to indicate that the problem can’t be addressed in time for the IE7 release, but marking real bugs as Won’t fix just makes me delete the "bugtracking" system from my bookmarks. Why bother reproducing a bug, creating a simplified test case, redacting everything if you are going to dismiss it right away?

    If you want us to take you seriously, then you have to start behaving seriously.

  27. Dear Mr Wilson,

    You claim that IE 7 team worked hard and fixed many CSS bugs and compliance bugs. I’d say that we all agree on these 2 opinions. But there were already many CSS bugs and compliance bugs affecting MSIE 6 to begin with and MSIE 7 is still quite far from web-designers minimum floor of acceptability considering

    a) the amount of time since CSS 1 has been released and CSS 2 has been officially released

    b) the huge incredible time Microsoft had to take necessary measures to make its browser meet Microsoft’s web-standards pretentions and claims during years. Everyone knows that Microsoft did not work on improving the browser (spec compliance, spec support and spec implementation correctness) between September 2001 and June 2004

    c) the huge cooperation and constructive/positive collaboration Microsoft got from web designers and web-standards advocates/supporters regarding reporting reproducible bugs (along with clear testcases):

    Peter-Paul Koch, Bruno Fassino, Holly Bergevin and John Gallant, Aleksandar Vacić (z-index), incutio.com, http://www.jehiah.com, etc,etc.et

    Microsoft is a major corporation which has been claiming to support W3C web standards and to look at improving W3C web standards compliance in its browsers for many years. Nevertheless, you will NOT find a single webpage or a single webpage interactive demo at MSDN which will pass an HTML validation test, which will be using a strict HTML 4.01 DTD ensuring to trigger MSIE 6 or MSIE 7 into standards compliant rendering mode. So, since 1997 and still today, MSDN (and all of its articles, interactive webpage-examples) has been ignoring W3C web standards, de facto not promoting web standards conformance. Microsoft has had and still has a double-talk, a double-face and is following a double-standard.

    More than 9 years ago, webstandards.org created this page

    http://archive.webstandards.org/css/winie/inline.html

    solely to notify Microsoft that it was not complying and conforming to web standards. Still today, in MSIE 7 beta 3, Microsoft has failed to comply. 9 years! Chances are it will take Microsoft 10 years to be able to render the page correctly!

    Microsoft has only itself to blame for its bad (laziness, complacent, non-attentive/non-listening to web designers and webstandards supporters) reputation.

    Where is the following document/when will the following document be brought back on microsoft.com website?

    "Microsoft is committed to working with the World Wide Web Consortium (W3C) to implement W3C-approved HTML standards, and has confirmed its pledge to work through W3C and other standards bodies on enhancements to HTML and other key Web technologies."

    http://www.microsoft.com/standards/intro.asp

    Gérard Talbot

  28. Arthur says:

    Hi. I’ve been tracking the development of browsers since Microsoft licensed Spyglass’ Mosaic sourcecode.

    Since you appear to be a developer:

    1. Does IE 7.0 still build on existing Mosaic code. (If you open the ABout box in IE 6.0, you’ll still find references to Mosaic/Spyglass)

    2. Or is IE 7.0 a complete engine rewrite (as in when Netscape’s/Mozilla’s Gecko).

  29. Mr Wilson, you said:

    "(…) Solid test cases we can access and bug reporting would help (…)"

    For over 18 months, Microsoft got an overwhelmingly good, positive, impressive and constructive response overall in that area from hundreds of people at wiki’s channel9-feedback  website and in many web authors’ personal sites.

    CSS1 and CSS2.1 bugs in MSIE 6 with/without reduced demo cases:

    http://channel9.msdn.com/wiki/default.aspx/Channel9.InternetExplorerSupportforCSS

    HTML 4 Support and spec violations

    HTML 4 Conformance tests results at http://www.robinlionheart.com/stds/html4/results

    W3C DRAFT HTML 4 Test Suite at http://www.w3.org/MarkUp/Test/HTML401/current/tests/index.html

    Standards bugs (dozens and dozens of clearly explained, detailed, specific bugs with links, testcases, etc.):

    http://channel9.msdn.com/wiki/default.aspx/Channel9.InternetExplorerStandardsSupport

    &lt;font&gt; and vertical-align: top in malformed markup code + IE6 SP2 = Crash!

    Everything had been nicely reported and clearly explained as soon as october 2003: it took than 30 months to Microsoft to fix this!

    MSIE 7 beta 3 is unable to pass this CSS1 testpage:

    http://www.hixie.ch/tests/adhoc/css/inheritance/border-color/001.html

    More bug collections sites at

       * MSIE 6 Crash bugs, MSIE 6 programming bugs, MSIE 6 DOM bugs at channel 9

       * Explorer exposed by Holly Bergevin and John Gallant (position is everything)

       * MSIE 6 web standards violations by Holly Bergevin and John Gallant (position is everything)

       * MSIE 6 Standards Support and related bugs at channel 9

       * 109 bugs reported regarding MSIE 6 at Peter-Paul Koch site

       * Hixie’s evil CSS testcases by Ian Hickson

       * Wrong z-index implementation bugs by Aleksandar Vacić

       * CSS3 Selectors test suite results by Robert Blaut

       * MSIE 6 bugs collection by Marc Pacheco

       * Internet Explorer 6 bugs collection A very impressive list of testcases by Bruno Fassino. Lots of layout bugs can be found in MSIE 6 in those tests.

    Specific MSIE 7 bug collection sites

       * MSIE 7 beta 2 bug collection (75 bugs) by Peter-Paul Koch

       * CSS bugs in MSIE 7 at incutio.com

       * MSIE 7 Beta 2 CSS Bugs at projectseven.com

       * The Float model problem at positioningiseverything.net

       * List item and hasLayout bugs at tanfa.co.uk

       * Internet Explorer 7 bugs collection A very impressive site of testcases done by Bruno Fassino

    Everything is listed nicely at

    http://www.gtalbot.org/BrowserBugsSection/

    Again, Microsoft can not blame others except its own long-lingering negligence, carelessness, lack of solid commitment, dormant activity during October 2001 to June 2004 period.

    Gérard

    P.S.: Your blogsite fails W3C markup validation, even with a HTML 4.0 transitional DTD.

  30. cooperpx says:

    HEY CHRIS! DON’T GET BUMMED OUT MAN!

    I’m pissed that I missed out on the last chat session. I wanted to drop in, while on vacation, to give your team a pat on the back. You guys have made lots of movement (considering the environment you work in) in a short time.

    I have this idea you went to bat to get the budget to fix up CSS, and took on a herculean effort to get her done. IE7 won’t be perfect, but it will be miles ahead of IE6 … and that’s what *you* set out to do.

    Get her stable, Get her released, and Get back in there! After IE7, we’ll get you a bunch of test suites or something to help out… Just tell us what you need.

  31. IE7b3 is an improvement over IE6 when it comes to CSS but I’m very surprised that some very crucial features were not fixed:

    :active and :focus. You fixed :hover so you should have realized that these are important as well.

    box-sizing. Please, the doctype switch is just not fine grained enough.

  32. Chris Warner says:

    @pd

    For your information, this blog is powered by CommunityServer which, although written in ASP.NET, has nothing to do with Microsoft.

    Most CommunityServer sites i’ve just had a quick look at all seem to validate okay, so maybe this is just an old version, or a simple mistake.

    Chill your beans 🙂

  33. @Chris Wilson

    I forgot to mention: although there is a possibility that I wll begin constructing test pages for many of the bugs in my database, I have already submitted a number of bug reports to the Internet Explorer feedback system with simple and straight-forward test cases. I have been disappointed with the way the feedback system has been run, specifically how so many reports are marked as "By design" or "Won’t fix" when, according to the explanations in the comments, they should have been filed as "Postponed" or simply switched from a "Bug" to "Suggestion". I’m trying to be as helpful as I can.

  34. Tino Zijdel says:

    I think that as long as IE7 still has bugs and shortcomings on a fundamental level it can be justified to state that IE7 is somewhere between 50 and 60 on a scale of 1 to 100 where 100 equals full CSS2.1 support.

    Besides inherit and the bugs that still remain in selectors you may add z-index ( http://therealcrisp.xs4all.nl/ie7beta/css_zindex.html ), layout rendering ( http://therealcrisp.xs4all.nl/ie7beta/css_relative_2.html ) and collapsing margins to the areas that still need a lot of attention and at this point will still require hacks and workarounds for developers such as myself.

  35. Jaques P Sauvant says:

    I’m not the biggest fan of IE, but when my designs work in every common browser except IE, I think I shouldn’t even bother with it.

  36. Michael Stuart says:

    I would like to see IE (and all other browsers) have the ability to update their rendering engine & rules easily (like how you check for updates in Firefox).  This way, even after you release IE7, as you build in new fixes to the engine, those tested fixes could be updated in the browser as minor version updates without people having to wait who knows how long until IE version 8.  Please don’t make us wait even a whole year after the release of IE7 for rendering improvements (which I consider just behind security patches as far as importance would go).  I wouldn’t be concerned about having X number of IE7 versions to code a site for…just code for the very latest – those people are online already…just have IE check for the latest update and let them know there’s a new update to install (like Firefox does).  I really like the direction of IE7 development so far!

  37. Chris Wilson, you said

    "(…) ‘IE 6 and IE 7 failed to render even basic columnar layouts in CSS’ in to solid work items, but I don’t understand the problem – much of the web uses columnar layouts that work in IE."

    This can not be completely true. A simple search+visit to many sites which show columnar layouts will demonstrate resistant/persistent CSS bugs in MSIE 7. You should also see/notice how much pain, sweat and frustration the people went through over the years to work around so many IE 5.x and IE 6 bugs. That is blatant+obvious to me.

    http://www.info.com.ph/~etan/w3pantheon/style/abmh.html

    http://www.info.com.ph/~etan/w3pantheon/style/modifiedsbmh.html

    ——

    Examine carefully the .css files of this (Skidoo) document

    Skidoo : 2 or 3 Column Layout

    http://webhost.bridgew.edu/etribou/layouts/skidoo/index.html

    and you’ll see at least 20 different kinds of CSS hacks aiming at IE 5+ in the Main stylesheets and Extra stylesheets.

    ——

    Here is a page where MSIE 7 still fails while Opera 9, Firefox 1.5 and NS 7.0 succeed accordingly, as expected:

    Test of Gallery Floats’ bug in IE

    http://awm.adsorption.org/prog/ex/galfloat.htm

    (the last 2 of that page still fail in IE7b3)

    ——

    Another page demonstrating IE7 bug is:

    Learn CSS Positioning in Ten Steps

    8. float columns (you have to click the tab "8")

    http://www.barelyfitz.com/screencast/html-training/css/positioning/

    ——

    It took me just 1-2 hours to find all these documents. You can find more "solid work items" on flaws/bugs in IE7 regarding columnar layouts if you care searching yourself. A good place to start is:

    CssLayouts

    http://css-discuss.incutio.com/?page=CssLayouts

    Gérard Talbot

  38. MMarcus says:

    > I recently worked a lot with DOM Events and its just

    > a pain that IE does not support addEventListener…

    > Ok, there are plenty of addEvent functions around

    > but IMHO these are only workarounds. And I prefer

    > the most standards compliant way, which would be

    > addEventListener…

    You may want to try this one:

    http://en.design-noir.de/webdev/JS/addEvent/

    Sure, it doesn’t implement addEventListener. But at least you can use events in a somewhat standards compliant way.

  39. I’ve read through as many comments as I could before my eyes started to glaze over. Basically, all I have to say is this: A majority of the market doesn’t give you a license to play Web-god. There are standards and they exist for a reason. Web developers like me will stand for nothing short of 100% compliance with CSS1, CSS2, and CSS3. Do other browsers currently comply at such a level? No. Do they come close? Sort of. Are they striving toward this goal. Absolutely. Based on what I’ve read from Microsoft employees sofar, I can hardly say the same for IE.

    Even if 100% standards compliance is reached with IE7, what about all of the other billions of people ineligible to upgrade, stuck with IE6? Release IE7 for older operating systems. These platform restrictions are ridiculous and only harmful to us, the users and developers of the Web.

    As an example, Mozilla Firefox is written by volunteers. As is Konqueror. Standards support is leaps and bounds better in these two browsers than in M$IE, which is a browser written entirely by paid "professionals".

    Not only should you be working on perfecting IE7 (you do owe it to us, after all), but you should be patching IE6 with these improvements, too.

    "Converting as many people to Free/Libre Web browsers, one day at a time."

  40. Tino Zijdel says:

    What exactly can we (webdevelopers) expect in terms of further improvements for CSS-support between now and the final release of IE7?

    The fact that many bugreports are being put off as ‘By Design’ or ‘Won’t fix’ don’t sound promising given the fact that many are actually *bugs* and I would consider them ‘blocking’ for the release of a browser that aims to be more standards-compliant. Not being compliant with CSS2.1 at least on a level that competitive browsers are (we’re not asking you to pass Acid2, just a simular level of compliance) will mean that developers still need to use hacks and workarounds to make pages look the same in IE7 as in other browsers. The bad thing is that IE7 as it looks now will need *different* hacks and workarounds than IE5 and/or IE6, adding to our workload and to the customer’s bill.

    IE7 (beta 3) still has fundamental flaws, bugs and shortcomings on the CSS-front (and we’re not even mentioning ECMAScript or DOM-compliance here, fronts that have barely been touched in IE7 but are 5 years behind as well), and it also looks to me as if Trident is being patched up where it should actually have been rewritten from scratch.

    I really do appreciate the effort that MS is making, but shipping IE7 as it is now when it comes to CSS-support would be a big mistake; it is only a half-finished product…

  41. Jim Pallett says:

    have you looked at THIS VERY SITE in IE7 compared to FF, Opera?

    the CSS for this site, with a page defending IE7’s amazingly improved CSS functionality?

    I’m typing this in IE7 right now, the calender at the top left encroaches into the main central column and the albatros image has an odd looking grey border to the right.

    still, at least you pro-IE7 blog hosted on MSDN looks nice in OTHER browsers eh!

  42. Jim Pallett says:

    had to post in firefox in the end, IE7 just sat there when i hit submit

  43. neil craig says:

    i’d like to say that on the face of it, i am pleased with what i hear on the development of IE7. i don’t use windows myself as i grew very tired of it’s perpetual determination to remain backwards compatible to the point of awkwardness but as a web developer i have to account for IE on windows.

    i’ve not been able to test IE7 myslef as for some reason the windows updater claims my install of xp at work is not valid when i know it is…however, thats another story…

    the points i’d like to raise are as follows:

    1. i really, really hope that IE7 will properly support the dimension properties of the box model and not expand widths and heights as the browser sees fit, ignoring my declarations. the same goes for min-height and min-width, these are fundamental properties which NEED to be fully functional.

    2. i’d agree with many of the previous posters, the IE team need to make sure they keep on track in terms of not inventing too many proprietary features, IE is WAY too far behind for that right at the moment.

    3. i absolutely applaud MS’ efforts to become more standards compliant, standards are there for a reason, to ease the pain of people such as myself who develop on a favoured browser (in my case firefox due to its invaluable range of extensions). i just hope that the typical MS bravado/arrogance which i must say i see echoes of in chris’ posts here (though to a much smaller extent than in years past) don’t take over. PLEASE PLEASE PLEASE recognise the power that MS has with IE and use that power responsibly for the better of everyone…make IE7 standards compliant and THEN innovate…we’re all stuck with the legacy battles of IE6 and previous on a daily basis…the internet is now clearly the most important data medium on the planet and MS has the bulk of users on IE, be honest and don’t abuse that fact and people will thank you for it. i think you guys are on the cusp of ‘getting it’. open source software is a real inspiration, people making software for the common good…there’s no dishonour in making a profit, just do it honestly, you seem to be almost on the right track, don’t stop now…

    apologies for the essay, just something i feel pretty strongly about…and i hope you are able  to take a few moments out to hear my opinion…

    thanks and the best of luck to you.

  44. todd says:

    Microsoft is simply putting most CSS improvements into the "too hard basket". MS priorities aren’t to web developers. Even when they push IE7 to all XP users as they’ve planned through windows update, and gain a larger market share than Firefox, doesn’t make it a better browser. Pretending IE hasn’t got CSS/DOM/XHTML/JavaScript issues, reminds me of the ‘see/speak/hear no evil monkeys’.

    Microsoft has 4 options

    * Re-write IE (unlikely)

    * License gecko/other browser maybe?

    * Do nothing, stay with the current code and see no great improvements.

    * Open source trident (ha)

    Accept you have rendering problems, and start addressing them, now.

  45. Jim Pallett says:

    i just dont think MS understand (or dont care) how much faith we have lost in IE over the last few years.

    lets say you made a car which exploded every time you went over a 10cm bump.

    contractors went round frantically filling in holes in the road, and on average, everyone was safe. Occasionally, someone purposely wouldnt fill in the holes on their driveway just in protest to the car manufacturers.

    even though other cars had been known to explode. You could drive them into a wall if your tried – over a mine field – lots of things would blow it up, none of these we nearly as big an issue as this one.

    eventually, they announced a new model with lots of improvements. It has a new look, new anti-theft devices, GPS, and they were going to force update this car to everyone it was so good.

    When they eventually made an announcement about the explosion issue they said "Fixed – now it wont explode unless you drive over a bump higher than 18cm : thats a VAST improvement"

    how many road surfacers would…well…just kill themselves there and then?

  46. cj says:

    i’m another person who had high hopes for the bug reporting system but gave up after seeing so many "won’t fixes" or "by designs" on issues i cared about.

    with many fixes (yay!) in ie7, i know you think you’ve made great strides, and in many ways you have.  on the other hand, though, you "broke" some of our work-arounds for ie6 but didn’t give us any way to fix them in ie7.  one big example in my own experience is containing divs that contain floats.  a wonderful method for containing a float uses :after (which ie7 doesn’t support [bug report #54017]) and then a hack for ie6, along the following lines:

    .menu:after {

    clear: both;

    content: ".";

    display: block;

    height: 0;

    visibility: hidden;

    }

    /* bs hoop for ie */

    * html .menu {height: 0;}/* */

    there are a few others (such as "display: table/table-cell"), but containing floats is a very big deal, at least for me.

    i’ll be honest and say i’ve hated ie for years, but my pages display better than i thought they would with no ie7 hacks, so you must be on the right track.  🙂  if you’d like demo pages of the various places my pages break in ie7, let me know.

    i appreciate the ie team’s willingness to start working on the browser, and i’ll keep my fingers crossed that it’ll be enough to lessen the load of ie6 (eventually, anyway) while not adding too many extra hack-arounds!

    [ps – as always, please, please, please at least think about letting us have ie6 and ie7 on the same dang machine so we can test side-by-side without a virtual machine.  i think you guys fail to realize that most computers at work aren’t state-of-the-art and just don’t have the ram for that.]

  47. Tino Zijdel says:

    cj: the most easy and logical way to have a container-element enclose your floats is to give the containing element an overflow: auto or overflow: hidden 😉 (also works in IE6 and IE7 after having been fixed since it was broken in beta 1)

  48. Levente says:

    I think most of us is scared that after IE7 the next version will come after another 5 years. It’s good that you fixed some bugs, but that’s not enough. Don’t stop improving IE!

  49. cj says:

    @tino

    that only works well *if* you are ok with scrollbars appearing in your pages or content being hidden, neither of which my bosses would be ok with.  😉  the only scrollbar we want is the one the browser provides, and they for sure do not want content disappearing, i can guarantee that!

  50. Mr Wilson,

    Jim Pallett is absolutely right. If you carefully examine this very blog page of yours, you’ll see several differences in the way IE7 renders this page… and this meets exactly your quest about basic columnar layout which break in IE7.

    "the calender at the top left encroaches into the main central column and the albatros image has an odd looking grey border to the right."

    Jim Pallet is entirely correct here. Opera 9.01, Firefox 1.5.0.6, Amaya 9.5 and NS 7.0 all

    – do not have a grey looking border to the right, between the left column and the main (content) central column

    – do not have the calendar (right column) encroaches into the main central column

    Yesterday, I spent 2 hours searching for CSS 1 and CSS 2.1 testcases which fails in IE 7. In 2 hours, I was able to find 25! (I will upload this into a webpage at my site at

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/CWilsonMSIE7AndCSSCompliance.html#SolidTestcasesWouldHelp

    ). I want to insist that these 25 testcases come from Ian "Hixie" Hickson CSS evil tests and that I take no credit, 0 credit for the testcases.

    IMO, Microsoft needs to

    – put a development team entirely on an entirely new rendering engine and entirely new browser (interface) (IE 8). With everything on the table:

     o sufficient technical, financial, human resources, etc.

     o drop support for backward-compatibility (legacy)

     o drop doctype switching (there should be only 1 rendering mode, always standards-compliant rendering mode)

     o develop a 4 year roadmap, schedule with goals to achieve: and make it public!

     o develop your own testcases based on the actual, real webpage designs existing on the web

     o improve your Expression Web Design software to go hand-to-hand with your long-term IE 8 browser

     o develop a web-standards evangelism team during that 4 years period. Even try to join with W3C, webstandards.org, Opera, Mozilla.org on creating that team. So that, eventually, the transition will be gradual, progressive and won’t hurt everyone

     o create a 4 year plan where IE 8 would be HTML 4.01 compliant, CSS 2.1 compliant, DOM 3 compliant, UAAG 1.0 compliant, etc. Aiming at/striving for 100% compliancy.

    – release minor/major upgrades to IE 7 for the next 4 years. Just forget about satisfying all of the web authors during those 4 years. Eg IE 7.1 should fix remaining/more CSS bugs (which does not require underlying os improvements) and have as many minor improvements as possible.

    – IE 7.1 (and IE 8) should have the same feature that M. Gueury developed for Firefox 1.x: a true SGML parser (based on OpenSP), just like the W3C validator uses, and HTML Tidy.

    http://users.skynet.be/mgueury/mozilla/index.html

    http://users.skynet.be/mgueury/mozilla/preview_080.html

    Thanks to a single right-click, users and developers would know where are the errors in their webpage. With a right-click, they could fix some errors, or go to the W3C validation online.

    – Explain publicly what are your plans. A lot of us knew well in 2001-2002 that Opera had 2 projects, one being to develop Opera 7 (DOM 2 Events compliant) while also fixing, improving Opera 6.x. Same thing with Netscape/Mozilla and its NS 6. At one point, they knew they had to rewrite entirely the rendering engine. They went through a lot of criticisms when NS 6.0 beta 1, beta 2, beta 3 and NS 6.01 were released but eventually NS 7.x browser releases were excellent ones.

    Gérard Talbot

  51. Mr Wilson,

    You wanted something better than bitching and moaning and you got that during 20 months from web authors, web standards advocates with all kinds of links, testcases, ideas, suggestions, constructive feedbacks, relevant quotes, concrete examples, demo pages, etc. in wiki-channel9 webpages.  And you still got all of that in your very own personal blog. You were given explanations and useful replies, etc. on each/all of your demands/requests.

    I fully understand the mistrust of many posters, their irritations, exasperations, the intensity of their frustrations, the depth of their resentment. And I do share their perspective, feelings.

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/CWilsonMSIE7AndCSSCompliance.html

    Gérard Talbot

  52. ceasar says:

    I have one question for you all.

    If IE7 will be fully "Compliance" then every webdesigner will be a lucky guy or girl.

    But what will happen then with the old sites scripted for smaller and larger compagnies with all the far from perfect styles

    I have seen a lot of old cms systems for large compagnies only made for IE with bad scripting.

    Will they still work ?

    Otherwise a lot of compagnies in the world will have a lot of trouble getting there.

    Then these large compagnies ( like government sites ) are all paying there bills to microsoft so when IE7 will to good to be true these large compagnies will all be in trouble.

    So I think IE7 is very hard to set up.

    First it need to render the latest correct code but also it has to support the old crappy code.

    Just a thought 🙂

  53. bh6507 says:

    Absolute and fixed positioning are still broken in IE7. In IE6, we at least had a work around by using padding to fake it. That of course was taking advantage of another IE6 defect related to the box model.  Well, the box model has been "fixed" and the work around is now gone. There is no way to even fake absolute positioning in IE7.

    See: https://connect.microsoft.com/IE/feedback/ViewFeedback.aspx?FeedbackID=60560

    How about at least fixing the partial CSS2 support already in the browser?

  54. Kalsta says:

    Re. http://blogs.msdn.com/cwilso/archive/2006/08/15/701894.aspx

    Chris, I think most developers will understand your anger and frustration. You and your team know better than anyone the technical (and corporate?) hurdles you face, and when you’ve worked your ass off to overcome these hurdles it’s hard not to take ‘uninformed or unrealistic’ comments personally.

    But please understand the anger from the other side … As web developers we are never afforded the luxury of telling our clients, ‘Sorry, I know your website doesn’t work quite right in Internet Explorer, but we did the best we can. We hope to fix the most glaring issues by sometime next year.’

    Please understand how angry I feel when my kids haven’t seen me all week and I’ve worked through another night to make my CSS layout work in IE6 — hours and hours of work that I will never be paid for. Multiply this experience by thousands and thousands of web developers world-wide (I often imagine them suffering as I do!) and you see why Microsoft has such a solemn duty to put whatever resources are necessary to fix this sinking ship once and for all (or else admit defeat and start shipping Windows with a copy of Firefox.)

    This anger is not the kind that reacts to a snarky comment or two. This is the kind of anger you feel when someone threatens your livelihood and your freedom. Yes, I use the word freedom, because I confess to feeling like a prisoner held against my will by the necessity to work with IE.

    Morally at least, I do believe we have a right to demand nothing short of on-par compliance with the likes of Firefox and Safari. While Microsoft makes it’s billions and your development team takes home a guaranteed salary, for small business people like myself our productivity defines our income. Resource cutting by Microsoft in the development of IE multiples millions of times over in the pockets and lives of real people and their families the world over.

    Anyway, I could go on, but I’ve made my point as best I can in the limited time I don’t have. Time to play with the kids.

  55. Dilip Kumar says:

    I think Microsoft should acquire Opera and merge the code with its IE base.

  56. Seth says:

    I’ve been a project manager in this field since 1996 and have seen through a lot of web development – everything from banners and microsites, to large corporate brochureware sites to complex web-based applications.

    I’ve been following the comments here, at Digg, and at ArsTechnica, and I’m surprised by the active voice of the developer community that develops for other browsers first and then apparently seems to hack their code to get it to work with IE. The reason I’m surprised is that nearly every developer I have ever worked with uses IE as their primary browser so the sites they code start out working in IE and then they have to hack the code to get it to work in Firefox, then in Safari, then in Opera.

    And honestly, that’s the way it should happen – only because IE still holds a dominant position in the marketplace, despite our distaste for it, despite its apparent lack of standards compliance. I don’t use IE as my primary web browser, but when I look at the logs from clients – and here I’m referring to top global brands – the average usage for IE is still well above 90% across the sites I’ve worked on. And at the firms where I’ve worked, any developer knows that the primary browser that has to be supported has to be what the audience is using, first, and then we can satisfy our compliance and other aspirational needs.

    There have been many times when my teams have worked on developing a site on a codebase that just renders like crap in Firefox. Why? Because they use IE and I generally use Firefox; and don’t even go there with Safari because during development cycles when I come home and check out work in Safari on my Mac, the sites often look like crap. And it takes a long time and a lot of hacking to get it to work right.

    Seems to me that it’s not just Microsoft that needs to improve support to be more standards compliant, but all browsers. I once thought that the W3C should trademark their compliance standards, maybe even the phrase "web browser", and they would officially license compliance confirmations and the term "web browser" to any application that passed a standard set of tests across the various standards. At least then we could have some reasonably good assurance that the browser itself, and not *our* code, is compliant. (Although code compliance is also important.)

  57. Tino Zijdel says:

    Seth: you are clearly one of those ‘new amateurs’ that haven’t kept themselves up to date with recent developments.

    When you think that first creating something to work with sub-optimal pseudo-standards and then ‘hacking’ it to get it right in browsers that actually follow the real standards is a bit strange at the least. The real danger lies in the fact that some of the developers with that mindset don’t even bother to make their IE-optimised site accessible to other browsers.

    The question should not be "how many of my users use IE?" (a percentage that is still dropping – on the site I work for it is already below 50%) but "how do I make my site accessible to everyone?" and that is partly what webstandards is about.

    Coding a public website for a specific browser is just bad practice. Actually you are lucky that Microsoft so fiercely holds on to backwards-compatibility (which is partly responsible for the fact that IE *cannot* be made fully standards-compatible) else you would be having a hell of a job getting your sites right when IE7 hits the market…

  58. Robshocka says:

    We’ve been subjugated to IE’s mediocrity for so long we don’t care about your problems in development Wilson, we already have our own to deal with. So excuse us for not sympathizing that IE7 will still keep IE in the worst modern browser a person can use category.

    If IE had been kept on the up and up for the last 10 years you wouldn’t be having a rough time, so don’t cry to us. Cry to those who came before you that left you such a mess for you to clean up.

  59. AT says:

    It doesn’t matter what IE7 offers for improvements because it will only run on Windows.

    Not only will it run on Windows exclusivley, it will primarly be supported on Vista to push OS sales.  It will also be haphazardly offered to "genuine" copies of Win2K & XP.  Which will clearly outline just how many of Microsoft’s customers aren’t genuine customers.

    Microsoft is now in the business of driving paying customers away and eliminating their installed base of users.  I think we should call it the "hardcore Microsoft".

    I have never nay-sayed on Microsoft until this year.  I have loved and shown interest in ASP.NET and other neat Microsoftisms.  But I can’t support any of it because the costs just don’t make sense.

    This year marks the first year where I have really seen popular opinion for Microsoft wane.

    So, even if IE7 improved things – it just won’t matter.  Your browser is being built into an expensive, deformed, retrofitted pig of an operating system.

  60. I think that many of the developers above may be a little too close to the problem to have a good perspective on the situation. The point is not that IE is the only browser with bugs. Nor that these won’t be improved upon in the next versions.

    The point is the vast majority of internet users do not even know how to change their browser. My mother used to think that Internet Explorer *was* the internet (until I got her an iMac). Therefore, you really need to get your act together. Why not use a different engine, as has been suggested? I will leave out technical suggestions, as you have enough here already.

    But you really must realise, and convey to the upper echelons at Microsoft, that developers do not appreciate this kind of behaviour from a monopolist such as yourselves. If you want to improve the company image, IE is the first place you should start.

  61. If security is the key driver for IE improvements (over and above standards compliance), how about rolling a few examples of potential security threats (phishing seems easiest) that can be created by exploiting CSS bugs? That way we could campaign for the fixing of CSS bugs on the grounds of security, not just compliance.

    BTW, only 11 days to go until the 5th anniversary of IE6’s last update.

  62. Jeff says:

    Chris,

    I think we all appreciate that effort you’ve put in to putting MS on the road to standards compliance.  Its a long time coming, and although it appears it won’t be complete for some time, its nice to know you’re on the right track.

    That being said it is extremely frustrating developing products to work in IE, when stupid stuff that really ought to work just won’t.  This seems to happen every time I try doing any advanced CSS layout.

    I do hope these problems are addressed, and further improvements are made.

  63. David says:

    Microsoft’s management is not stupid. They recognize that the end user is more important then the web developer.

    It’s simple really: web developers must and will bend backwards to make sure their sites work for their audience.

    This is why Microsoft is putting most of their effort on the frontend and could care less about standards.

  64. Joe says:

    Why is it that MS can produce Visual Studio 2005 which is both an extremely high quality product AND manages near 100% standards compliance (with C++/ISO14482) but can also produce such complete rubbish as IE?

    Do you hang your head in shame when you pass the VS team?

  65. missing says:

    Hi, as a developer I am alos always faced with users that say I didn’t do anything, because the things they though were important aren’t done. Similarily this is to IE7, lots of things are done, but since there are so many things you *can* do, there are so many people that thing you didn’t do the right things.

    So the thin I usually do is to ignore the tone people tell it in, and extract the things that they say, and make up a todo list with *my* prioritites. So for IE7, we all know that there *are* bugs, and that it would be good if they are fixed.

    If I would develop IE7 a goal for me would be to be able to create pages that render in IE7 nearly the same way as in other browsers. If you detect a bug in their rendering, make it better than they do. So in the end it should be possible to have standard compliant documents, and no need for having to write for different browsers. When what feature is added, changed or fixed isn’t important here, as long is its somewhere on the todo list, and people agree that its a problem.

  66. James says:

    Alas, the fact remains that it’s to MS’s advantage to not be standard conformant, because nonconformance, combined with its monopoly position, encourages web sites that don’t work well with other browsers.

  67. Sean says:

    I have absolutely zero sympathy for Microsoft.  The acquisitions about Internet Explorer failing to comply with web standards are ultimately true.  Other browser have done it; there’s no reason that Internet Explorer can’t do the same.  The problem is that IE was left to collect dust for the past few years.  Now, with the push to get Vista out the door, the IE developers are forced to make difficult decisions to meet product deadlines.  Web developers have asked for years to make IE more standards compliant, and from a consumers view point, it appears that only recently has MS given the idea any thought.

    I urge Microsoft to take its time with releasing both Vista and IE.  As Vista is already delayed until the mid-part of 2007, I would strongly recommend using this time to deliver a product that is a lot more standards compliant than what it currently is.

  68. Roberto Lopez says:

    Web design has always been an activity of particular discouragement for me.  I am picky about the appearance of my work, and want my pages to look professional, if not at the very least neat.  Worst of all, I have always sorely loathed the inevitable task of building a sea of TABLE, TR, and TD tags.  Pages built with unending swarms of table tags become difficult to maintain in short order.

    To my luck, I re-discovered CSS design, having been previously unaware of its layout capabilities.  Once I discovered how easy it could be to separate the web design from the web content, I immiedately set out to establish myself upon its use.  I found learning materials on the web, and in the bookstore.  In particular, I bought a book called "CSS Mastery" by Andy Budd – mostly for refernece.

    To my surprise, a significant proportion of the material covers CSS techniques, and then how to do the exact same thing in IE5-6!  Almost every idea put forward is amended with a side note on how to get it to work on IE.  Almost every topic in this book sets aside time to describe a working solution for IE5-6.  <b>In fact, this book has an entire chapter dedicated to IE5-6 bugs – chapter 9: Bugs and Bug Fixing.</b>

    At the end of this chapter is the following summary titled, "Stop Picking on Internet Explorer":

    Page 183-184:

    Internet Explorer isn’t the only buggy bowser around, so you may wonder why I have been focusing my attentions on IE bugs. Don’t worry, its not another case of Microsoft bashing there are good reasons for this focus.

    First, IE has by far the biggest browser market share.  With so many copies in circulation, IE bugs tend to get found and documented pretty quickly.  When a major CSS bug gets discovered in IE, scores of developers will be on the case trying to find a fix or a workaround.  Because of this popularity, there are more well documented bugs and fixed for IE than any other browser.

    <b>

    The other major issue is the pace of development.  Browsers such as Firefox, Safari, and Opera are constantly being updated, with new builds appearing with remarkable frequency.  Almost as soon as a bug is discovered, it is fixed and a new version of the browser is released.  Because of this, any Firefox or Safari bug I talk about now will probably have been fixed by the next revision.</b>

    This pace of development is excellent, but it does have its own problems.  Rather than having two or three versions of a browser to deal with, you may have 20 or 30 [Lopez, personal note: I don’t think its that bad.].  You can never be sure if your users have the latest version, and this makes testing extremely difficult.  IE on the other hand, didn’t see a major revision for about 5 years.  As such, there has been much more time for bugs to surface and much more impetus to find a fix.

    Luckily, IE 7 promises to be a much more compliant browser.  Many of the better known IE bugs have been addressed, along with increased support for advanced CSS 2.1 selectors such as the child and attribute selectors.  As with all browsers, new bugs will surface, and IE 7 will be far from perfect.  However, the faster people can be convinced to upgrade to modern browsers such as IE 7 and Firefox, the quicker older browsers such as IE 5.0 can be retired.

    ===

    End Book Quote

    Having said all of this, I wonder if the IE team would consider changing their rate of release to match that of Firefox’s.  Instead of waiting every 5 or so years, we might get updates every 6 months (at least?).  Either way, after 5 years for IE to still not be up to par with other browsers does not make sense, especially for a company with the resources of Microsoft.  After so long, IE7 should be an achivement in standards compliance, not a case study on improper implementation.

    Lastly, I would like to state the primary concern: simple things do not work.  This is the heart of the problem.  I am sure Firefox has many issues with CSS, but they are far from basic as far as I can tell.  Even when I am doing basic CSS editing, I must come back to this book and review the necessary hack.  Very simple layouts and techniques require me to check for IE hacks, where I find basic functionality in Firefox – thats really all I want, that the basics work! But they don’t!

  69. The short and sweet version of my response:

    First, look at: http://www.webdevout.net/browser_support_summary.php

    The measure of compliance is not (and never well be!) improvement over IE6; it’s how well you match the existing standards.  The above site’s breakdown shows how far behind IE7 is still, regardless of improvements over a previous release.

    Also indicated is that no browser is fully compliant.  IE’s compliance is markedly less than Firefox and Opera.

  70. RMX says:

    So if IE does follow standards, I assume that means I can code to the standards and Microsoft WILL SUPPORT OUR LARGE GOVERNMENT CUSTOMERS BY FIXING THE BUGS that cause our app to display incorrectly.

    Next time they come to me with one of these issues, I’m extremely tempted to change my former answer of

    "Yes sir, we do support both Standard Compliant browsers and MSIE’s defacto standard, so I’ll open a bug here"

    to

    "Sorry, sir, we support the internet standards and your MSIE browser vendor claims to as well, please call them."

  71. Alt Text says:

    An old post at Slashdot resurfaced, for whatever reason, and caused quite a stir about how non-compliant (with CSS standards) Microsoft’s…

  72. Tadd Giles says:

    That is one brutal comment thread.  Chris, unfortunately, I think you’re going to have to live with this type of commentary for a long time.  Any personal attacks on you or your team are certainly unwarranted, but the underlying angst is real and valid.  You’ll have ship multiple releases each with better standards compliance before this goes away.  Three beta releases is just not enough.  Web developers are still dealing every day with trying to make things work in IE.  Until that changes, the angst isn’t going to go away.

    I personally think you’re doing the right things.  You’re on the right track now.  You just need to keep executing on IE7 and subsequent releases and this will go away.    But you’re going to have to live with the fact that lots of web devs are going to be cranky, inpatient and unfair until you do.

  73. pinto says:

    I think that some of us are talking past each other here.  Chris is defending his priorities and his team and pointing the finger up the chain.  Well, to the extent that he’ll allow for any blame at all.  We for our part aren’t concerned so much about him or his team so much as Microsoft as a whole and their inability to produce a functioning product.

    Listen, most of us are developers here.  We know what it’s like to be undermanned and underbudgeted.  And we know what it’s like management’s priorities don’t dovetail with our vision of technical excellance.  We feel for you in that sense.

    On the other hand, let’s be honest.  Microsoft doesn’t care in the least about web devs.  How would we know if they did?  Well, they care about Windows devs – look at VS 2K5.  More than any other company on the planet, MSFT has the manpower and budget to make a great browser happen if they wanted to.  That they haven’t is surefire evidence that they don’t care.

    Chris, we’d feel for you a little more if you didn’t insult our intelligence so much.  "…our goal is to make the lives of web developers better?"  Come on.  Maybe that’s *your* goal, but it’s not MS’s goal.  If it was, you’d have the resources to do it.  CSS didn’t crawl out from under a rock last week – you (MS) have been ignoring it and us for years.  Your calls for constructive suggestions, public bug logging and "prioritizing" are condescending at best.  The web dev community has gone through extensive pains to document, reproduce and hack around the mess that you guys release.  Honestly, any issues with web-dev<->MS communication are on your side.  In the future, we’ll give you whatever info you need in the highly unlikely case that we haven’t already done so.  Priorities are up to you guys.  Perhaps you can spend a few minutes on Google (or Microsoft MSN Live Search Extreme Edition 2007 or whatever you guys use) and tally complaints for a weighting system?  

    Speaking of which, although I kind of see your point wrt the percentage battle, it seems kind of childish to equivocate in this arena.  There isn’t a single way of measuring standards support that doesn’t suggest that you guys are miles behind.  Different ballpark behind.  Embarrassingly, laughably, "oh my god is this really the richest software company in the world?" behind.  Fudging the numbers might fly (barely) in the Windows-Linux TCO battle, but not here.

    As for me, color me unimpressed with IE7.  It would have been a great rev if you’d released it, say 3-4 years ago.  And then updated every 1-2 years since then.  I’m sure it was hard work to get things where they are, but take my goal into account: I’ll consider you guys "there" when it’s no harder to support IE and Firefox than Safari and Firefox.  IE7 makes significant but nowhere near sufficient progress in that regard.

    Speaking of progress, I notice that you haven’t responded to any questions for a future revision roadmap.  I also notice that most public bug logging centers tend to be devoid of real discussion and instead full of "won’t fix" and "by design" write-offs.  If all IE7 is is a day-late, dollar-short effort to get "caught up" to 2002 or so then sit around until 2010, any good will you might earn will be gone fast than you can say ‘SELECT {style="z-index -1000;}’.  Of course it’s true that "all browsers have bugs."  For the last five years though, it’s been true that "all browsers but IE have folks fixing bugs."

    I personally don’t know how I’d handle it if my project were the laughing stock of the software industry (regardless of who was to blame).  I admire you for your courage, but ask you to consider who you work for and what they stand for.  You knew what the score was before you got involved.  If, as you suggest, you are the voice agitating for future standards support within IE then bravo.  I’m sure your team has done the best they could with what they were given.  Please keep it up and don’t let management think that you’re "there" just because you get a percentage boost once IE7 gets force-fed to every mom and pop with XP-SP2.

  74. Ben Myers says:

    Stop kidding yourself and support :before and :after pseudo-tags.

    When are you going to support these sort of things, 3 or 4 years?

  75. George says:

    I really appreciate that you actually took the time to respond (without cursing) to these idiots that keep criticizing your team. If you say that you’re the one that got CSS implemented in IE, then that’s amazing. I just hope that you aren’t affected by people calling you lazy or a liar. All I can say is, congratulations on IE7 (I like it, though the interface is kind of sparse — is a skinning mod an option?) and keep up the good work.

  76. Daniel says:

    I think IE would get a lot more CSS respect, if:

    1.) The button element worked correctly

    2.) The input[type=button] worked correctly

    Buttons are fundamental to GUI design.  The simple fact, that the 2 elements available for creating buttons in IE is flawed, has held the Web back, and since not fixed in IE7, continues to hold the Web back.

    Development of slick, mass producable web forms, for all browsers, is not possible, because supporting legacy IE software is required, and IE can’t even meet the design specs.

  77. I don't get it... says:

    I understand that IE is currently behind in standards support, I understand that more work needs to be done on it, and I understand that there is a big pile of history (the "albatross") that has lead to this point.

    The only thing I don’t understand is these kind of statements, "My customers all use IE, so I developed for Firefox, and it looks rubbish in IE" of which there are quite a few above.

    It probably is Microsoft’s fault that it looks rubbish in IE, but to deliver something to customers that doesn’t work on their platform is irresponsible. Regardless of how you shift the blame.

    Good to see progress with IE, and yes, more needs to be done to make developers lives easier, but keep at it.

  78. Pissed off... says:

    Buy Opera Software (MS can afford it I suppose), integrate it’s browser (withot much bugs in the process) and end this pain.

    You’ve shortened my life enough already…

  79. You know, it’s interesting how much passion this thread has generated (more on that in a moment).

    Sure IE6 was pretty average on CSS standards support and folks needed to work out some hacks to make things behave.

    IE7 is undoubtedly better, but sadly most of the people posting comments here have an agenda and even if MS was 100% compliant but still supported a few ground-breaking (or plain useful) extensions then they’d feel honour bound to mark them down.

    IE7 is however still in beta, not even Release Candidate stage so obviously there is a fair amount of work to be done.

    As a developer I work with designers who actually consider CSS to be broken because there are things that you either can’t do, or that require complex workarounds (even in Safari, Firefox and Opera) to make something really simple work – why no float:bottom I often hear as I’m asked to code a hack with javascript to dynamically position the footer in the right place. Ah, bring back tables!

    Back to the passion. I started reading Chris’ blog out of curiosity a while ago and have grown to respect the honesty and candid posting. As I read down the responses to his post I was really disappointed… here was a perfect opportunity for the community to present some valid test cases, to highlight some things that were important to designers, developers and (ultimately) end users so the team would have some feedback to know what’s important in the real world, where their efforts are appreciated and where they are wasting their time… instead the tone got nastier the further on I went

    You don’t encourage someone by smacking them in the face every time they open their mouth…

    Oh, and I am one of those developers who code for my 90%+ audience first (and yes, my stats of several million page impressions a day over the last few months agree with that figure) and then test on current releases of (PC and Mac) Firefox and Safari… if it works on those I’m happy

  80. This week I interviewed Microsoft’s Chris Wilson, the Group Program Manager for IE, to address the issue of Web standards compliance and IE7. There has been controversy about this lately, sparked by a Slashdot thread last week that claimed IE7…

  81. Jamie says:

    I am sick and tired of hearing excuses of why IE7 will not be as standard-compliant as other lesser established browsers after over 4 years of no attempt to fix the mess that is IE <= 6! I am a web developer and 70-80% of my time is spent trying to "back-port" my code to IE. It is disgraceful that the IE team didn’t make Standards-compliance their number 1 priority for IE7. I have used the IE7 betas and while an improvement, it’s still not good enough. MS have the money, the market-share and are forcing this piece of cr*p onto unsuspecting Windows users who, the majority, will not look for an alternative and MS knows it *BUT* web developers are starting to feel betrayed and lose their patience with IE so it’s only a matter of time before IE support is relegated to 2nd or 3rd priority and developers start the browser revolution by forcing users to ditch IE that will scare MS into getting off their lazy butts and doing something for IE that will benefit the development of the web rather than line their own pockets. I can’t wait.

  82. Kiko says:

    It it so hard not to curse the IE team. You had all the time and resources, why can’t you j-u-s-t do it right? What is so hard about supporting  :after and :before? IE team do you even listen?

  83. Mr Wilson,

    I have created 2 clear, simplified testcases (11 lines and 9 lines long now) regarding the way left: auto and margin collapsing are rendered in IE 7 beta 3.

    1st testcase:

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/LeftAutoMarginCollapsingBugInIE7.html

    2nd testcase:

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/LeftAutoMarginCollapsingBugInIE7_2.html

    This 2nd testcase does not even need a formal "left: auto" declaration.

    Both testcases are rendered as expected, accordingly in Firefox 1.5.0.6, Opera 9.01 and in a fair number of other browsers.

    I sincerly think this left: auto margin collapsing bug is serious enough to fix it before releasing IE 7 final.

    Gérard Talbot

  84. Rob says:

    Despite IE7 being in beta, it is ‘code complete’ and what you see is what you are going to get.  Although Chris complains about the web site slanted against IE7, ALL the browsers are rated the same way.  So if the other were as bad as IE7 then they, too, would get the same rating.  

    Face it Chris.  IE7 is lousy and you know it.  To quote YOU:

    "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…"  And you didn’t.

  85. Rob, you apparently wanted to skip right over that bit where I say _BETA_1_.  _BETA_1_.  Because the platform team mostly opted out of Beta 1, to put most of their changes in Beta 2.

    No, I do not think IE7 is lousy.  Grr.

  86. Tino Zijdel says:

    No indeed, IE7 is not lousy compared to IE6; it’s CSS-compliance is definitely improved albeit (unfortunately) buggy.

    It’s compliance is also by no means comparable to the level of other browsers, so in that light you might still call it ‘lousy’. It puts us developers in a hard spot – having to hack for IE6 to make our designs work, but now also having to hack seperately for IE7.

    Implementing only half a standard might be understandable, but shipping a product that has serious bugs in the part that has been implemented – while knowing it(!) – is unforgivable. I sincerely hope that the first release candidate will address the current issues, if it does not then I foresee a lot more mud-throwing in the near future…

  87. I have to admit that I am quite happy with IE7’s new support for CSS and making it more standards compliant.  There were quite a few things in IE6 that drove me nuts, and at least now things look practically the same in all the browsers I can test on (Windows based) when I use strict modes.   There was even a point where I actually considered that IE7 would be better for me to use than Firefox :).  There’s no way that would ever happen with IE6.  

  88. Hello again,

    3 testcases:

    1- Default border-color when not explicitly defined:

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/BorderColorBugIE7.html

    This is a failure to implement correctly CSS 1 border-color property. NS 6.2 passes this testcase.

    2- Default border-color

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/BorderColorBugIE7_2.html

    This testcase comes from an example taken as is and as given in the last 3 Working Draft of CSS 2.1. NS 6.2 passes this testcase.

    3- "background-attachment: scroll" should scroll within the containing block, not within the viewport

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/BackgroundCenterOfBodyBugInIE7.html

    This bug is a failure to implement correctly CSS 1 background-attachment property. It must be said that only recent releases of Opera and Firefox pass this testcase.

    Fair enough?

    Gérard Talbot

  89. Hello again,

    1 of the 5 malformed declarations come directly from the CSS 2.1 spec (Section 4.2, malformed declarations) which was already in the last 3 Working Draft, as it is. In other words, it was simple, very simple to spot it and fix it before.

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/ParsingErrorsMalformedDeclarations.html

    Firefox 1.5.0.6, Opera 9.01 and NS 6.2 passes all 3; Amaya 9.5 passes 2 out of 3. MSIE 7 beta 3 fails on all 3.

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/MalformedBorderShorthandDeclarations.html

    Firefox 1.5.0.6, Opera 9.01 and NS 6.2 passes both; MSIE 7 beta 3 fails on both.

    Gérard

  90. Keith says:

    Just buy Firefox and call it a day.

  91. Tino Zijdel says:

    Way to go Gérard 😉

    I’ve got two more to add with respect to IE’s broken float-model:

    reflow and wrapping: http://therealcrisp.xs4all.nl/ie7beta/css_float_1.html

    clearing: http://therealcrisp.xs4all.nl/ie7beta/css_float_2.html

    I have looked around a bit in connect but I’m quite disappointed because it is slow, messy and not enough proper feedback from Microsoft itself so I don’t feel very encouraged to report bugs there. Thereby I still feel that because there are still so many bugs on a fundamental level that Microsoft should be well aware of that situation – these are issues that should not be in a public beta-version but should have been recognized and dealt with by Microsofts own QA-team.

    My feeling is that Trident has just been hacked to deal with the issues mentioned on PIE, but that is just the tip of the iceberg and Microsoft should admit that the real problems are on a much more fundamental level and they just cannot solve those without rewriting Trident completely.

    I still see a lot of CSS-issues from IE6 present in IE7 and a lot still have to do with the ‘hasLayout’-property. I think ‘hasLayout’ is some way to allow for speedy rendering and in the time it was implemented it probably was a very nifty optimization and worked fine for the CSS1 visual formatting model. I don’t however see how it fits in in the visual formatting model of CSS2.1 which is quite a lot more complex.

    Again: please don’t ship a browser with a broken CSS-implementation – that will do more harm than good.

  92. Jordn Fenton says:

    I tried out IE7 for a bit yesterday to check out if it works all right and how good its supposed CSS compliance is.

    It works fine. Just not with basic CSS. IE6, for instance, was able to properly load the navbar at the top of WordPress blogs (my blog, at http://unisyc.wordpress.com, had THREE blue bars at the top!).

    Then there are other sites with CSS that works in decent browsers (Mozilla-based ones, in particular). My website (http://gestott.googlepages.com – please don’t think I’m advertising or anything – it’s just that I tried for about four months to get my CSS to work with IE – nothing worked) works with Mozilla-based browsers, but obviously I’m not good enough for Redmond!

    No, IE just hacks its way through the CSS I’ve been working on since March and makes it look like I was building the CSS during a fit.

    Please don’t say it’s CSS compliant when it isn’t.

    Another tip – don’t use IE. Use something that doesn’t just look nice.

  93. Leon Breedt says:

    I’m just going to murmur E4X (ECMA-357) under my breath in the hope someone notices 🙂

  94. Thomas says:

    Check out my page with IE6.

    Try to change the font size — see what happens.

    Try to click on imprint — see what happens.

    so you call this a standard compliant browser?

    Especially when reloading on imprint, you will see flickering effects. This browser is insecure patchwork, nothing more.

  95. Lonely user says:

    OMG! Theres someone working on IE again! Someone take a picture before the team gets disbanded again!

  96. Bobby says:

    The code for Firefox is all over the web.  They’ve done a pretty good job at adhering to standards.  It’s like you guys are taking a test with the book right there and yet still can’t do well.

  97. Hello Mr Wilson,

    I hope you still read this long thread.

    1- Float bug in IE 7 (CSS 1 Test Suite: 4.1.4)

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/FloatBugIE7.html

    2- Another float bug in IE 7 (CSS 1 Test Suite: 5.5.25b)

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/AnotherFloatBugIE7.html

    3- Background-color inherit (CSS 1 Test Suite: 7.1)

    http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/BackgroundColorInherit.html

    (To be absolutely fair here, MSIE 7 passes that test but, as mentionned in the testpage, it fails from a CSS 2 perspective within that CSS1 testsuite)

    If you visit this page

    http://www.richinstyle.com/bugs/ie5demo.html

    with MSIE 7 RC1 build 5700.6 and carefully examine/study each test, you’ll see/witness at least 50 CSS bugs afflicting rendering/layout.

    Gérard Talbot

  98. Tino Zijdel says:

    HTML4.01 : http://therealcrisp.xs4all.nl/ie7beta/html_nesting.html

    this can’t be in a 2006 browser in strict mode…

  99. Marco Weber says:

    i’ve just installed IE7…

    well, you’ve guys have solved a big problem: the PNG support.

    but i was really looking forward to see :focus, :before and :after support…

    i couldn’t test everything yet, but my first impression is, that there is a new interface with some minor changes…

    i can’t tell a lot about the security, but i can see, that you’re doing a lot to prevent maloperations caused by unexpierenced users…

    ( i.e. phishing filters, … )

    i hope, that future versions will support more CSS features…

  100. Klaus Hartl says:

    Did you know that the universal selector is a little too universal in IE7 RC1 and selects even HTML comments, which can in turn be some styling applied.

    Test here:

    http://stilbuero.de/demo/ie7_universal_selector_bug/

  101. DRead says:

    Try validating THIS page to see how important standards are to Microsoft…

         <a href="http://validator.w3.org/check?uri=referer"><img

           src="http://www.w3.org/Icons/valid-xhtml10&quot; height="31" width="88"

           alt="Valid XHTML 1.0!" /></a>

  102. Click on the "W3C Validator" link (or DRead in my last post). It should take you to the W3C validator to validate THIS page. Ouch!

  103. Albert Grau says:

    Hi everybody

    I have a problem with the onclick event in the input type=checkbox when the visibility:hidden

    Is this a bug? What is your opinion?

    I have an example:

    It works in ie6 and ff1.5, but it doesn’t work in ie7.0

    Thanks

    <html>

    <head>

    <style type="text/css">

    input.test{

              visibility: hidden;

             }

    </style>

    </head>

    <body>

      <label>1

        <input id="test"

               type="checkbox"

               id="test"

               name="nameTest"  

               value="1"

               onclick="alert(‘checked: ‘ +

               document.getElementById(‘test’).checked);"/>

      </label>

    </body>

    </html>

    ————-

    The tricky option has been move the onclick in the div

    <html>

    <head>

    <style type="text/css">

    input#test{

              visibility: hidden;

             }

    </style>

    </head>

    <script type="text/javascript">

    function simulation(cb) {

      if (cb.checked == false)

            cb.checked = true;

      else

            cb.checked = false;

      alert("checked: " + cb.checked);

    }

    </script>

    <body>

      <label onclick="simulation(document.getElementById(‘test’));")">1

        <input

               type="checkbox"

               id="test"

               name="nameTest"  

               value="1"

         />

      </label>

    </body>

    </html>

  104. Viktor "vitya404" Balogh says:

    I do not want to read the whole thing that’s written above, so forgive me if I repeat some arguements.

    I like the new features of IE7, I think it has great improvements.

    But guess what, if I write a program that processes a data (ie. html document) in a standard format (CSS), I won’t put any shiny and twinkling features (tabbed browsing, no menu, autocomplete, not even search field) in it before implementing the FULL support of the standard (eg. ‘inherit’). Even if I think the standard is maybe a heap of sh*t.

    You have three choices:

    1. to say you do not implement CSS, we should forget it, you’ll create an own standard

    2. do not make IE publicly available as it is only supporting a limited, only MS used subset of the standard.

    3. implement the FULL standard.

    I know that IT guys do not think as we, proper electronics engineers, but still, some minimal respect of standards would be welcome.

  105. Tachyon says:

    All this discussion means nothing.

    The fact is that Microsoft’s motivations will always be wrong. Therefore all their decision making will be compromised. This shows in all of their products.

    Example, their desire to lock-in users to their products causes them to make individual products that only work ‘properly’ when in cooperation with other Microsoft products. This frustrates users who would most often rather have product X be standards compliant and interoperate cleanly with their choice of product Y, from whatever vendor.

    An example relative to IE is that for years all of Microsoft’s products that could output HTML (Office, Frontpage, etc.) put out code that often only worked properly when viewed with IE.

    Modern bad business based on greed ultimately fails because it has one focus. Short term growth in profits, market share or sales. Enron types. Microsoft  has been of this type through all of it’s history.

    The right goal for a company is to make the best product it can, at the best price, and always put the customer first. Apply the golden rule to customers.

    This shifts the company to a long term profit model with slow, steady growth.

    Microsoft’s problem is it’s leaders and management all drink the Gates/Ballmer Kool-aid. They live with the idea that they know best, and everyone should do things their way, and pay them for the privilege.

    Microsoft is like communism, it only works if _everyone_ is communist.

    Microsoft is deluded into thinking that they can and will create a Microsoft only world, a utopia in which they will actually do the crap you see in their recent commercials. It would almost be worth happening to see them no longer have anyone else to blame for their inferior products. If all the world _was_ Microsoft, utopia is the last place it would be like. But then _maybe_ Gates and Ballmer would actually get it. They aren’t omniscient genius’ that should be running the world. They are actually the crazy Kim Jong Il’s who imagine they are perfect.

    The REAL Tachyon

  106. Tachyon says:

    P.S.

    The above post was made with Firefox, on Linux.

    I’ve avoided the Kool-aid.

  107. . Fino a qui le imperfezioni di Internet Explorer 7 sono davvero minime… ma ben si sa che i problemi del browser Microsoft che hanno portato molti (fra cui il sottoscritto, ma anche grandi nomi come il CERT) a sconsigliarne apertamente l’uso non r

  108. Thad says:

    You said you worked hard to make IE7 CSS compliant – I can believe that – the improvements are visible. You don’t take into account though how far you were behind with IE6 and there is still a lot of work to be done. Until IE will really be superb in keeping standards to make for IE6 that kept them randomly, you will always be criticized by many frustrated web developers (like me now) who have to work hard to make their sites look all right under this toy-browser IE6, because unfortunately many people are still using it.

  109. Börse says:

    What exactly can we (webdevelopers) expect in terms of further improvements for CSS-support between now and the final release of IE7?

    The fact that many bugreports are being put off as ‘By Design’ or ‘Won’t fix’ don’t sound promising given the fact that many are actually *bugs* and I would consider them ‘blocking’ for the release of a browser that aims to be more standards-compliant. Not being compliant with CSS2.1 at least on a level that competitive browsers are (we’re not asking you to pass Acid2, just a simular level of compliance) will mean that developers still need to use hacks and workarounds to make pages look the same in IE7 as in other browsers. The bad thing is that IE7 as it looks now will need *different* hacks and workarounds than IE5 and/or IE6, adding to our workload and to the customer’s bill.

    IE7 (beta 3) still has fundamental flaws, bugs and shortcomings on the CSS-front (and we’re not even mentioning ECMAScript or DOM-compliance here, fronts that have barely been touched in IE7 but are 5 years behind as well), and it also looks to me as if Trident is being patched up where it should actually have been rewritten from scratch.

    I really do appreciate the effort that MS is making, but shipping IE7 as it is now when it comes to CSS-support would be a big mistake; it is only a half-finished product…

  110. lexikon says:

    We’ve been subjugated to IE’s mediocrity for so long we don’t care about your problems in development Wilson, we already have our own to deal with. So excuse us for not sympathizing that IE7 will still keep IE in the worst modern browser a person can use category.

    If IE had been kept on the up and up for the last 10 years you wouldn’t be having a rough time, so don’t cry to us. Cry to those who came before you that left you such a mess for you to clean up.

  111. Mark Kempton says:

    Here’s a way to make IE6 behave like a standards compliant browser using javascript libraries that you can place in your websites:

    http://www.grumpycoder.co.uk/how-to-fix-ie6-for-w3c-standards-compliance-using-javascript-libraries/

  112. Johnny says:

    A few posts above: "Microsoft is like communism, it only works if _everyone_ is communist."

    Why don’t you call it "microsoftism", -and you are the founder of this religion? The bible ist written with frontpage! ;P

    Greetz, J.

  113. çeviri says:

    had to post in firefox in the end, IE7 just sat there when i hit submit..

  114. Roland says:

    A really interesting discussion!

  115. elvisparsley says:

    You can find an interesting article about IE here. The title is "In All Fairness … Internet Explorer Still Stinks" http://www.sitepoint.com/blogs/2007/11/22/in-all-fairness-%e2%80%a6-internet-explorer-still-stinks/

  116. Steve Yeoman says:

    Nobody encourages IE use at all, but with IE6 still accounting for 43% of browsing in NZ we have to design sites that will work in that.

    I take the view that hardly anybody switches browsers on a daily basis, let alone has three running at the same time, except developers, so we only have to design sites that function and are attractive, not necessarily sites that look exactly the same wherever they are displayed.   Any design brief will start with "Design Restrictions" and in our case the biggest one is that we have very little control over the environment we are are displaying our work in, unlike print where the designer can specify the type of paper and printer to be used.  I don’t mark a student down if their H1 heading is slightly different in IE and Firefox because the viewers will always be using the same browsers and won’t notice the difference.

    There is too much focus still on presentation of web content on computer screens.  For example, mobile devices will take over in the very near future as the most  common way of viewing web content and I haven’t seen a phone with a 21" screen yet.

    I’ve developed a subset of CSS rules which give me the maximum possible cross-browser consistency, with the primary aim of maximum accessibility, usability and search engine friendliness. I always lay out the html in this order

    <body>

    <H1>

    <main menu>

    <primary content>

    <secondary content>

    and right at the bottom of the script, the masthead and footer and any spare graphics to make the thing look OK to those who want to look at pictures and happen to be viewing this content on a computer.

    CSS takes care of the display on any device and for print – Absolute positioning for the screen CSS (and liquid layout because who knows what size screen the bloody public will be using), hiding divs that contain only decoration sets the page up for handheld and screenreaders and print. I avoid padding and borders if possible, using margins on contained elements rather than padding on the containers, and if I have to use a bit of padding or borders I make sure that there’s some liquid space between that element and the next, to make allowances for silly old IE .

    I think if we focus on the important issues – accessibility & usability – and accept that we have no control over which user agent is used to view our content, and even if the majority is viewed on computer screens, that we have no control of the size, resolution or image quality, then issues like which browser is implementing which minor CSS rule more precisely than the other become less of an issue.

  117. Fred Boulton says:

    Why does MS <em>need</em> to do anything? When you have 500,000,000 users with 95% of them not even knowing how to change their browser and 99% of users not concerned with compatibility(not even knowing what it is)you don’t need to put any time, effort or dollars into changing something that works well for everyone who uses it. IE does work well,’cos Web developers make it work well, ‘cos we have clients breathing down our necks and 99.999% of our clients use IE and 95% of their customers use IE. If it doesn’t work in ALL versions of IE we are soon told about it. No one breathes down MS’s neck, they are their own boss and beyond listening to a few cries of "foul" from Web developers.

    I think that we, as Web developers, are kidding ourselves if we think that MS considers our mental welfare when they are building browsers. It’s not like there’s anyone telling them that their browsers <em>must</em> conform to standards. They have never bothered about browser compatibility and standards, so why are they going to start now?

    Lots of people here seem surprised about all of the incompatibility issues(and bugs) in IE7. We weren’t <em>really</em> surprised, were we? Of course not!

    We will carry on like we have since browsers were first built, designing for the standards-compatible browsers first and then tweaking to allow for all of the IE quirks.

    Guys, learn to live with it!(IE that is)

  118. Ashwin Kamath says:

    I think some of the comments are a bit harsh on IE (even though I spend most of my time juggling with CSS hacks to get it right on most browsers). The problem I believe is in the Web technology. Other technologies like graphics, animation have advanced so much but we web developers have to struggle to make rounded corners. CSS sucks, hope somebody comes out with better alternative.

  119. bilgi yarışması says:

    I think some of the comments are a bit harsh on IE (even though I spend most of my time juggling with CSS hacks to get it right on most browsers).

  120. Webdesign says:

    "I know that you hack far more for IE6 than anything else; I’m interested at this point in IE7."

    Chris, I’m sorry, do you really want to know that? I can give you an example from the real world. Have a look at the latest site I’m working on: <a href="http://plazes.com">http://plazes.com</a&gt;, i.e. at the style sheet that is included for IE only (hidden by Conditional Comments).

    As soon as IE 7 Beta 2 was told to be "layout complete" I started to test in IE 7 as well. And it appeared that – though you fixed the most infamous bugs in IE7 – I have to apply the same hacks for IE 6 and IE 7 in ca. 90% of all cases. The other 8% even worse I have to apply different hacks to IE 6 and IE 7 by using special hacks (thank god you implemented advanced css selectors in IE 7) and the rest IE 6 only.

    Most of these bugs are related to elements having layout or not, which makes me starting to think that something fundamentally is wrong with IE’s rendering engine. I really had some hopes for IE 7, but sadly enough it now seems to me that nearly nothing changed from a practical standpoint. I can see no use of advanced selectors if the rendering is still broken.

    And by the way, the simple attribute selector <code>[type=’text’]</code> makes IE 7 crash.

    And we haven’t yet talked about fixes you did for JavaScript. Ah no, there aren’t any (ok, besides of the native XHR object support – how native is it anyway?).

  121. Webhosting says:

    You know, it’s interesting how much passion this thread has generated (more on that in a moment).

    Sure IE6 was pretty average on CSS standards support and folks needed to work out some hacks to make things behave.

    IE7 is undoubtedly better, but sadly most of the people posting comments here have an agenda and even if MS was 100% compliant but still supported a few ground-breaking (or plain useful) extensions then they’d feel honour bound to mark them down.

    IE7 is however still in beta, not even Release Candidate stage so obviously there is a fair amount of work to be done.

    As a developer I work with designers who actually consider CSS to be broken because there are things that you either can’t do, or that require complex workarounds (even in Safari, Firefox and Opera) to make something really simple work – why no float:bottom I often hear as I’m asked to code a hack with javascript to dynamically position the footer in the right place. Ah, bring back tables!

    Back to the passion. I started reading Chris’ blog out of curiosity a while ago and have grown to respect the honesty and candid posting. As I read down the responses to his post I was really disappointed… here was a perfect opportunity for the community to present some valid test cases, to highlight some things that were important to designers, developers and (ultimately) end users so the team would have some feedback to know what’s important in the real world, where their efforts are appreciated and where they are wasting their time… instead the tone got nastier the further on I went

    You don’t encourage someone by smacking them in the face every time they open their mouth…

    Oh, and I am one of those developers who code for my 90%+ audience first (and yes, my stats of several million page impressions a day over the last few months agree with that figure) and then test on current releases of (PC and Mac) Firefox and Safari… if it works on those I’m happy

  122. "getting ready to doing it again in the next release"

    Is that just really bad grammar, or are you referring to the sound a spring makes when you try to bend it and it snaps back?

    I know you’ll be able to doing it. They’ve been doinging it at Microsoft for so long that I bet you can check the software archives and you’ll find an ‘auto-doinger’ that will doing it for you XD

  123. All I can say is AMEN. How can IE.x (put any version you want down) be so arrogant. FIX THE DANG BROWSER AND MAKE IT COMPLIANT!!!!!!!

    We have to fix websites to work with the rest of the world, and then Microsoft. What a shame.

    Microsoft, fix your dang browser. YES FIX IT, IT IS CLEARLY A PIECE OF JUNK UNTIL YOU COMPLY LIKE THE REST OF THE BROWSER WORLD! SHAME SHAME SHAME ON YOU!

    I won’t pardon my caps, because I refrained from using profanity.

  124. Gewinnspiel says:

    The question should not be "how many of my users use IE?" (a percentage that is still dropping – on the site I work for it is already below 50%) but "how do I make my site accessible to everyone?" and that is partly what webstandards is about.

    Coding a public website for a specific browser is just bad practice. Actually you are lucky that Microsoft so fiercely holds on to backwards-compatibility (which is partly responsible for the fact that IE *cannot* be made fully standards-compatible) else you would be having a hell of a job getting your sites right when IE7 hits the market…

  125. Börsenspiel says:

    CSS takes care of the display on any device and for print – Absolute positioning for the screen CSS (and liquid layout because who knows what size screen the bloody public will be using), hiding divs that contain only decoration sets the page up for handheld and screenreaders and print. I avoid padding and borders if possible, using margins on contained elements rather than padding on the containers, and if I have to use a bit of padding or borders I make sure that there’s some liquid space between that element and the next, to make allowances for silly old IE .

  126. rwf says:

    You know, if IE 7 truly complied with ALL of CSS 1 and ALL of HTML 4.01, it wouldn’t be so bad. But, it does not.

    Go look h<a href="http://www.gtalbot.org/BrowserBugsSection/MSIE7Bugs/CWilsonMSIE7AndCSSCompliance.html">here</a&gt; for more information. There are more bugs than that, though; go look <a href="http://csshack.wordpress.com/2008/08/26/165-bugs-in-ie7-browser/">here</a&gt;.

    Does it pass Acid2? Acid2 isn’t all that intense a test, honestly.

    Of course, the rest of the world is moving already towards CSS 3.

    I can write a basic website in a couple of hours, or less. To fix the CSS and HTML code so it works with all flavors of IE doubles the amount of time I spend on a website. I don’t have to add ANYTHING to the typical website to make it work with Opera.

    The point is, I shouldn’t have to code anything special so that my websites will work. Every browser should display them the same way.

    Now, if IE had extensions that would let it do other things – well, okay; if it can display Word docs without opening new windows, great. I don’t judge a browser’s usability based on its extensions; I judge a browser’s usability based on its standards compliance. In the same way, I judge a car’s usability based on its compliance with standards (all four wheels firmly attached, engine functions, et cetera), not on its extensions (T-top, turbo, et cetera). The extensions make it nicer, but if the wheels are broken, what good is the turbo?

    Same thing with IE. The wheels are broken. What good are the extensions?

  127. red slider says:

    I’m neither a troll nor a professional web-designer and have no commercial interest in any of this. I’m just a guy, like plenty of others, trying to build a website for his wife. What should have taken a week has been two months of hair-pulling for the reasons stated often enough in the comments here.

    It doesn’t matter if MS sat on its hands for nine years, or if it claims 3 or 4 or 10% solutions that it promises will make the next version "much better". The fact is, the company that killed Netscape (even if it was proven wrong to have done so) could resolve the entire matter with CSS and non-compliance in a week if it wished to do so.  It could even do more – it could work with the legitimate standards consortium to not only make all of its browser versions, from 5x on into the future, fully compliant (with appropriate fixes and replacements for old rendering engines), but could actually assist W3C to improve its own models. The modules to do that are free and open-source! MS wouldn’t even have to pay for them.  

    That it doesn’t tells us more than the volumes it has written about its "progress". In the last round of ‘proprietary wars’ it killed a perfectly legitimate attempt to provide an alternative in the market place. Netscape might have succeeded or failed on its own, but we’ll never know.  This time, it may very well kill its own horse.  I quit msie several years ago, before I even knew about CSS. The current issues are just more of the same as far as I can see, and it means that many many more will catch on and say bye-bye as this matter wears on.  In the end, a 50% or 75% solution won’t be enough. In the meantime, MS will only have succeeded in temporarily impeding the progress of the net and in wasting a lot of honest people’s time.  Its a shame, but that’s what my recent experience and reading this board tells me.  The writing is on the wall, if you care to read it.

    nothing personal – you guys probably have worked hard – too bad the thing you put your time into causes so many others so much misery. It could have been different, if enough of you, right there at MS, had had the courage to say "no!"

  128. balabo3_fn says:

    <a href= http://fcmdhau.angelfire.com >i dorothy dandridge</a> <a href= http://zpsylec.angelfire.com >sunset property management oregon</a> <a href= http://twoxmoi.angelfire.com >lake oswego football</a> <a href= http://mqpwzaj.angelfire.com >rsc cares</a> <a href= http://exeyxui.angelfire.com >straford career institute</a>

  129. Eric says:

    Stop development of IE. Destroy all known versions of IE from your company. Make it so people can not view the internet via IE. Destroy Trident. Install Firefox onto new OS. Make it so people can not use IE at all, no matter what version. You will find everyone loves your company a lot more. You guys got your wish. Netscape died when you monopolized your OS  into using IE. Now give back by letting IE die. Please and thank you.

  130. Chas4 says:

    This page:

    Errors found while checking this document as XHTML 1.0 Frameset!Result: 1749 Errors, 10 warning(s)

    http://validator.w3.org

    This is the worst one I have ever seen

  131. medyum says:

    As soon as IE 7 Beta 2 was told to be "layout complete" I started to test in IE 7 as well. And it appeared that – though you fixed the most infamous bugs in IE7 – I have to apply the same hacks for IE 6 and IE 7 in ca. 90% of all cases. The other 8% even worse I have to apply different hacks to IE 6 and IE 7 by using special hacks (thank god you implemented advanced css selectors in IE 7) and the rest IE 6 only.

    Most of these bugs are related to elements having layout or not, which makes me starting to think that something fundamentally is wrong with IE’s rendering engine. I really had some hopes for IE 7, but sadly enough it now seems to me that nearly nothing changed from a practical standpoint. I can see no use of advanced selectors if the rendering is still broken.

  132. Medyum says:

    The question should not be "how many of my users use IE?" (a percentage that is still dropping – on the site I work for it is already below 50%) but "how do I make my site accessible to everyone?" and that is partly what webstandards is about.

    Coding a public website for a specific browser is just bad practice. Actually you are lucky that Microsoft so fiercely holds on to backwards-compatibility (which is partly responsible for the fact that IE *cannot* be made fully standards-compatible) else you would be having a hell of a job getting your sites right when IE7 hits the market…

  133. Reuben says:

    Two and a half years down the track, and I’ve noticed IE8 has just been pushed to Microsoft Update for Vista.

    But I’m still having to support IE6 with my web developments, and have to make a special IE6.css because IE6 never handled multiple class selectors.

    I havent accepted responsibility for supporting IE8.  Feeling kinda gipped about it, and that its actually going to be like supporting two browsers.  One will be the one I will like, IE8 Strict, and the other will be IE8 Quirks.  And no doubt, I’ll still need installations of IE7 and IE6, because IE8 Quirks wont work like either of those.

  134. ceviri says:

    had to post in firefox in the end, IE7 just sat there when i hit submit..

  135. medyum says:

    CSS takes care of the display on any device and for print – Absolute positioning for the screen CSS (and liquid layout because who knows what size screen the bloody public will be using), hiding divs that contain only decoration sets the page up for handheld and screenreaders and print. I avoid padding and borders if possible, using margins on contained elements rather than padding on the containers, and if I have to use a bit of padding or borders I make sure that there's some liquid space between that element and the next, to make allowances for silly old IE …. http://www.medyumoguz.net