W3C Site Launch

Today we launched the new W3C. We’ve been working on it for a while, so I’m happy that it is seeing the light of day.

Comments are flowing in, some touching on issues we identified when we announced the beta version. Here are a few:

  1. Is the CSS invalid? The CSS does not validate with the W3C CSS validator. We mentioned this as one limitation of the site back in March. As we wrote then, “Because of known interoperability issues, we have accepted to use CSS that does not validate with the CSS validator. Over time we hope to evolve towards valid CSS.”
  2. Why do some pages (such as the graphics introduction, though there are others as well) look unfinished? They are; the generic template text is still there from the beta. We decided to launch the site even without all the content we hope to have. We think the site is a significant improvement over the old one, and so prefer to begin using it rather than wait for more content. The site will continue to evolve, and I hope much more easily. We are asking staff, Working Groups, and the community to help out and provide content. We’d love your help, and are happy to acknowledge your contributions on the pages. Let us know at site-comment@w3.org.
  3. Some of the rewritten Recommendations have formatting bugs. Unfortunately, one of our processing passes modified the markup and we didn’t realize it; we’ll be fixing those problems in place. For the moment we are only using the new templates for Recommendations (old and new). As we gain more experience and resolve formatting issues, we expect to apply the new templates to more publications. One advantage of the new approach will be that it will be easier to tell right up front when a specification has been superseded by another.

There are also a few rendering issues we are aware of and plan to fix over the next few days. Please tell us about any issues you encounter on site-comments@w3.org. Please be sure to tell us the URI of the page in question and what browser and OS you are using.

37 thoughts on “W3C Site Launch

  1. Looks great, good work on the update! The site is much nicer to use now.

    I do wonder at the class naming scheme though – seems to use presentational names a fair bit, eg: h3 class=”h4 tPadding0 bPadding0 summary” Was that legacy code or was there a particular reason for those names?

  2. Regarding the presentation names for classes: I’ve found a mix of “semantic classes” and “presentation classes” has made the css a bit easier to maintain, and the templates a bit easier to manage. I think Nicole Sullivan would be able to say a lot about the grid approach; I found it very useful to be able to dissociate the (grid) layout from other semantics.

  3. It seems you still have debug info in comments in the source page f.i :
    @h h3 class=h4 is directly coming from Nicole Sullivan’s OOCSS recommendation

  4. I can’t see any use of WAI-ARIA landmark roles. I know ARIA is still a working draft, but given that browser and assistive technology vendors are actively implementing and improving support, would this not have been a good opportunity to demonstrate the accessibility benefits?

  5. The sites nice.

    Shooting yourself in the foot… no shooting yourself in the face = YES with the invalid CSS. If everyone else can manage to get Valid CSS then SO CAN YOU!

    “Because of known interoperability issues, we have accepted to use CSS that does not validate with the CSS validator. Over time we hope to evolve towards valid CSS.” ITS THE DAY TO DAY PROBLEM OF THE AVERAGE WEBSITE DESIGNER SO LIVE WITH THE RULES YOU’VE SET AND FIX THE CODE!

    The INVALID HTML CODE in your examples within the documents MUST BE CHANGED, what the hell are you playing at? I expected with the update of your website you’d fix this, but obviously NOT


    You are supposed to be setting an example.. so for gods sake do it, no wonder people are no longer taking you serious!

    Adam King #WebStandards #Nazi

  6. Regarding the use of WAI-ARIA, one reason we’ve not started to use extensively is that it is not yet a standard. I am pretty sure that we will begin to incorporate it over time.

  7. It IS a lot nicer than the old one, no doubt about that. That said, did anyone else felt a heavy sense of irony when they read “Over time we hope to evolve towards valid CSS.”?

  8. This is really funny that the new website does not even validate to WAI-A

    Fails on:-
    12.4 Associate labels explicitly with their controls.
    13.1 Clearly identify the target of each link.
    13.2 Provide metadata to add semantic information to pages and sites.

    Nor have you clearly defined the language of the document.
    +Invalid CSS

    And you guys are supposed to be the web standards folk? I seriously facepalm, even more reason for Cowboys to ignore Web Standards now.


    What next, you rebuild the site in tables? And the invalid HTML in the examples has still not been addressed.

    Why should professional web designers use web standards when you yourself do not?

  9. Adam, regarding your comments about accessibility, we’ve worked hard to ensure that the home page satisfied WCAG 2 (AA). Naturally, if there are errors we will fix them because ensuring the site is accessible is very important.

    You’ve not provided a URI to the page or pages you are talking about, so I can’t be sure. Also, please do not assume that because there are glitches in the site that we don’t care about using the standards. Of course we know that people expect W3C to use the standards, and we work hard to do so. So I’d like to ask for you help in identifying and fixing what we may not yet have done correctly.

  10. Not bad. Don’t start slacking though. Keep working on it if you want to stay at the forefront of the global web design community. There’s so much creative talent out there these days…

  11. Over time we hope to evolve towards valid CSS

    How will that be possible when much of the invalid CSS is catering to non-evolving browsers like IE?

    Technicalities aside, it’s a very nice design!

  12. Great! from long time i was expecting for new usability implitaiton on the world king site.

    Excellent work guys!

  13. Hi,

    I think I found a spelling mistake in your

    I could be wrong, but should “Validation, lint” actually read: “Validation, links” ?

    I do find your navigation confusing. I had a difficult time finding your CSS and HTML validation pages. Some improvements needed here, I feel.

    Best wishes,

    New Zealand.

    1. Hi Robin,

      “lint” is intended, in particular in the context of “charlint,” a character normalization tool:

      As to finding the validators, they appear on the home page on the right hand side. In general they will be available above the fold, but right now we have our developer gathering prominently featured; that will go away after 6 November and the validators will again move up the page.

      _ Ian

  14. From an artistic point of view ,the new design is nice and fresh
    and probably long overdue.

    From a usability and mission point of view I’m
    not really that happy with the new design.

    I really like to use and recommend the online validators, and found it
    of great value for them to be up front where their presence
    made them both easy to use, and easy to promote.

    I sorta get this mental image of someone standing there saying
    “we are the folks that design web authoring standards”
    while they try to inconspicuously push the validators in the closet
    with one leg while grinning sheepishly.

    So now with the standards buried , the validators buried
    I guess it’s time to fill the front page with promotions for
    Adobe, and perhaps next Microsoft’s latest browser offering,
    and self aggrandizing calender notes saying “look at how busy we are”
    instead of focusing on the tools and reference materials that
    most folks came for.

    I would recommend taking most of that chatter on the
    front page and sticking it under magazine.w3.org and
    keeping the http://www.w3.org focus on the gospel of the standards
    and the tools to allow people to comply with them.

    Sad thing that a standards site would find a
    new paint job so important to release that actually
    putting their money where their mouth is
    took a back seat.

    1. Hi Dr. Clue,

      As I’ve mentioned to others, the validators will move up the page again after next week. Our current big meeting has pushed them down temporarily. The will be above the fold 95% of the time.

      Hope that helps,

      _ Ian

  15. Well it’s a very nice job, I seek the beta version for a while now, and I’m impressed to see the final version. Looks nice, modern with a lots of little things that save time and help navigation. Nicole Sullivan and all the team have done a good job!
    (and thanks to Nicole I learnt how to use object oriented CSS for the next website version :) )

    Nicolas Chevallier, CEO of Allogarage

  16. Well it’s a very nice job, I seek the beta version for a while now, and I’m impressed to see the final version. Looks nice, modern with a lots of little things that save time and help navigation. Nicole Sullivan and all the team have done a good job!
    (and thanks to Nicole I learnt how to use object oriented CSS for the next website version :) )

    Nicolas Chevallier, CEO of Allogarage

  17. Looks nice, but I cannot find the test suites anymore (e.g. the 3.0 selector tests or the CSS 2.1 tests).

    Are these not available anymore?

    1. Robert,

      The site redesign didn’t touch the test suites, so they should still be there, probably linked from the CSS WG home page. There are new places where links to test suites would probably also be helpful. Can you tell me what path you followed looking for them?

      _ Ian

  18. I would like to thanks all the development team who made minor updates since the deployment of this new version and correct the bugs, like always in this kind of complete rewrite of a big website like W3C one.
    The CSS part is quite impressive.

    Nicolas Chevallier, CEO Allogarage

    1. Hi Cecil,

      W3C’s use of invalid CSS has surprised a number of people. I’ve mentioned in other messages that we acknowledged this as sub-optimal, that we plan to improve it over time, and that we welcome suggested fixes.

      The site is also not 100% mobileOK. Like anyone else, we had to make some decisions and tradeoffs. I am not sure that we need to apologize for acknowledging real world interop issues while also trying to follow correctly all of the relevant Web standards.

      _ Ian

  19. I am not sure that we need to apologize for acknowledging real world interop issues while also trying to follow correctly…


    With respect, your reply would only seem to me to make sense if it were the case that strictly following standards is sometimes impossible.

    Two things: (i) Your site should serve as an example of best practice – this is what is expected of you as a standards champion. (ii) I suggest that is never a need to release invalid CSS under any circumstances provided that you are masters of your own desting, since there are techniques available that allow you to hide CSS from badly behaved UAs. Modern well-behaved UAs never need to see CSS that you are ashamed of, even if you might need to feed older UAs with something odd.

    Your organisation’s failure here is in not being as a first class example to authors. That seems to me to be undeniable. Other authors can cope with badly behaved UAs, and I fail to see what makes you so different.

    It’s of course not for Ian to apologize to the community for lack of leadership, that should be down to those at the top of the organisation.

    Those arguing for the importance of web standards will now for ever be confronted with the “even the w3c failed” argument by the naysayers.

    1. Hi Cecil,

      We’ve reduced the number of CSS validity errors since the launch. We now have essentially two types of errors: those related to IE6 support, and those related to using CSS3 properties (which are accompanied by some proprietary properties during the transition to CSS3 support).

      If we decide to drop support for IE6, the first set will go away. I expect we will keep IE6 support for a short while longer. We have not wanted to use explicit browser sniffing or IE6 conditional comments. I realize that these hacks are implicit brower sniffing. But we’ve wanted to confine this to CSS rather than use other mechanisms. However, we are now discussing adding IE6 support via javascript. (I welcome suggestions on how to do this well.) I consider this pushing the problem under the rug, but it will be an improvement if the main CSS is “more valid, more of the time.”

      For the CSS3 support, I think the validator can be improved. CSS 2.1 validity requires that you only use properties and values defined in CSS 2.1: “…it must contain only at-rules, property names, and property values defined in this specification.” I think a different result from the validator would be more useful here, like “valid CSS 2.1 + some other well-formed properties.” For the moment, I’m inclined to keep the CSS3 properties we are using and tolerate the validity errors, and wait for validator improvements and/or CSS3.

    1. There are very good reasons to promote and ensure validity. I also believe validity is a tool, to be wielded with judgment among many considerations. I hope I’ve been clear from the start that we expect this particular issue to go away as we make improvements and learn from actual usage.

  20. Well done !!! I’m glad to finally see the final version (It’s been a long time since I last visited your blog !) It pleasant and easy to use! Anyway, can’t wait to see a French version of w3 ! ;)

    Marie Robin, Trouvea’s Webmarketing Manager

Comments are closed.