Original comments

1. #1 On December 16th 2007 at 05:42pm Keri Henare said:

   The lack of ‘Clearly defined timetables’ has been my major gripe about both the HTML & CSS working groups in the past.

   There needs to be clear roadmaps for the completion of the components of a standard as well as the standard as a whole.  Additionally there needs to be roadmaps for the implementation of standards by browser vendors.

2. #2 On December 16th 2007 at 05:47pm Matt Wilcox said:

   Interesting post Andy, and I agree with most of your points.

   Most especially I agree with the need for a leadership role _from the people who use the specifications_ rather than the people that build them or the people that implement them. Just as not having browser vendors input in the WG is absurd, it’s equally absurd that CSS (and HTML) Working Groups are being led by software engineers rather than designers and developers. Softwre engineers are vitally important to making practicable recommendations - but they are not the people to decide what is required from the technologies themselves.

   With any product or any sort you need to question the end users to find out what they like, what they dislike, what they really need, and to sound out how your ideas for future products might be received or used. At the moment, this simply doesn’t happen without being half-hidden behind obscure working groups which then get mired down with in-fighting. What happens at the W3C now seems to be engineers getting together to produce a Homer Simpson car that no-one wants (XHTML2 ... ), or getting together to debate endlessly on nuances that are irrelevent, with a lack of focus on _why_ CSS/HTML exists and what jobs those technologies are meant to do.

   I’m looking forward to a few good, calm, debates on this issue. At the very least it’ll be good to learn a bit more about how the W3C works in reality rather than in theory.

3. #3 On December 16th 2007 at 05:50pm Tim Van Damme said:

   First of all: Thank you, Andy, for daring to take a point, and defending it! It takes huge balls to do what you just did…

   No matter what the outcome of this little ripple is, everybody should remember that the decisions made by the group, will only have a real effect in 5 to 10 years. I think that’s the time it takes a browser to grow market-share big enough to start developing for.

   So, do it for your offspring. Give them the chance to use slick html and css, and a better internet…

4. #4 On December 16th 2007 at 05:50pm zeldman said:

   Agreed we appear to be at a critical turning point. That there’s much frustration and uncertainty, we both know. Best way forward, hard to say.

   Audits are always a good way to begin tackling a problem, so I agree with that part of your proposal—the form of it if not necessarily the details.

   Since the W3C is unlikely to commission such an audit, it might need to be done externally and presented to the W3C. This is an opportunity for The Web Standards Project to lead, in the same way we externally audited top 10 CSS failings in Opera and IE back in 1998.

   The WaSP, though, seems unlikely to lead. Perhaps, then, a consortium of volunteers from among web design agencies and freelancers who understand standards-based design.

   That seems a good place to start, and as it comes out of your rough draft proposal for change, I thank you for the proposal.

   If volunteers cohere and craft an audit, they might need to present it to the community before presenting it to the W3C (or simultaneously) for credibility.

   How the W3C reacts to such an audit would tell us much.

5. #5 On December 16th 2007 at 06:00pm Gareth Rushgrove said:

   I wholeheartedly agree regarding the need for CSS3 to be more of project - with all the process and tools that that implies.

   I’d recently written of similar concerns (http://morethanseven.net/posts/css-snapshots-css3-modules-and-an-agile-way-forward/) and would like a more software orientated approach, using tools and techniques learnt from large software successes and failures to be utilised.

   I think the idea of an audit (either internal or external) is a good starting point. Are their reasonable parallels to this community driven audit anywhere else from which lessons can be learnt?

6. #6 On December 16th 2007 at 06:02pm Daniel Glazman said:

   And how many people in the world are able to standardize CSS having in mind the constraints of a layout engine first, of browser and global web strategy then ? You are looking for a utopy, good luck in your quest.

7. #7 On December 16th 2007 at 06:15pm Roger Johansson said:

   Your proposals look suitable for the HTML Working Group as well.

8. #8 On December 16th 2007 at 06:17pm prisca said:

   Andy,

   thanks for speaking up - I felt your post ‘CSS Unworking Group’ was very well written and though I feel I do not know enough about the inner workings of the CSS Working Group to fully understand all your points - I was glad to see someone with your reputation making such a strong point and challenge the current state of affairs. The follow up comments you received and now your response here finally seem to get things moving - get people talking and hopefully start moving the development of CSS into the right direction.

   What we need are new, creative solutions to our problems. Coming up with these solutions requires a creative as well as a technical approach. It also requires a new way of designing, developing, testing and launching of these solutions, all within a defined, transparent framework and with timescales attached. We can no longer afford to treat the design and development of new CSS features as as academic exercise.

   I agree that we need both technicians and designers to work together here - and bickering in or out of court does neither help or resolve any issues. Your proposed ideas sound very practical and much more productive than the little I know about the W3C so far, yet I might be thinking very naively here, hoping that web designers and developers would be part of the CSS development.

   Andy, thanks for putting it all into words (much better than I can) - hope you will get some good responses :)

9. #9 On December 16th 2007 at 06:20pm Adrian Rinehart-Balfe said:

   Working group? In any other industry they would have been history years ago. It isn’t working at all

10. #10 On December 16th 2007 at 06:23pm Dave S. said:

    All other things being equal, the unanswered question in my mind is who pays the Invited Experts in your proposal?

    Currently, at least some of those working on the various standards are on the payroll of large companies. They don’t have to fill their day communicating with clients and doing paid work, as a percentage of their week is set aside for WG activities. This model guarantees a certain amount of attention is devoted to the activities required to get the job done, but even with such dedicated resources you see the resulting pace of development.

    If this new working group is meant to comprise of volunteers, it cannot work. Altruism is a nice kick start, but it doesn’t have legs for the immense project that is defining a standard; momentum would die unpleasantly fast. Any changes to the current system need to come with proposed funding methods as well.

11. #11 On December 16th 2007 at 06:25pm Matt Wilcox said:

    “how many people in the world are able to standardize CSS having in mind the constraints of a layout engine first”

    With respect, that is not our problem. Designers and developers should set the goals we want for our tools - the manner in which they are achieved is not down to us, nor should it be.

    Creating specifications that adhere to the inherent limits of current technology, or the excepted understanding of ‘what you can/cant do’ is exactly the road the W3C are walking, and it heads nowhere fun for anyone.

    I understand that there are limits - but the approach that’s been going on isn’t getting us where we all want to be. Realising that means realising other strategies need to be considered. Not taken, simply considered.

    The big issue is not who’s right and who’s wrong, what works and what can’t - it’s about talking to each other to find out why we have the problems we do, and how to create a way forward that works for everyone.

12. #12 On December 16th 2007 at 06:27pm Andy Clarke said:

    @ Daniel Glazman: Daniel, your lack of inspirational thinking surprises me. As a designer, my job to interpret what my clients need and work hard to find a solution. Your job is to build a browser that interpret whatever CSS features we need to satisfy our clients, CSS that has been designed and developed by people who use it everyday.

    Web designers have had to use ever more creative thinking to get CSS to do what we need. We have not allowed technologies to limit creativity. Your response makes me wonder whether negative thinking like yours has been one of the major factors in the slow progress of CSS.

    @ Dave S: I believe that the Invited Experts and all the other key staff I outlined should work full-time on the CSS project and be either loaned out by their employers, or if they are self-employed (like you or me) paid at our normal daily rate by the W3C, sponsorship or some other creative way. Working for free is not an option.

13. #13 On December 16th 2007 at 06:43pm Matt Wilcox said:

    Andy, you said in your first half of comment 12 exactly what I was trying to say in #11 - only yours was far more clear. Well said.

14. #14 On December 16th 2007 at 07:15pm thacker said:

    Clarke—

    Thank you very much.  This needed to come from the inside.

    I hope that Berners-Lee will, also, step up to the plate and make the necessary changes to the W3C’s operational structure and that includes reliable sources of income.

    Changes within the W3C are the critical components.

    To men and women at Opera, thank you very much, also.  You helped to precipitate something, inadvertently, that needs done.

15. #15 On December 16th 2007 at 07:16pm Jon Hughes said:

    @ Andy:
    I do not profess to know much about the W3C’s inner workings, but it is my understanding, which may be wrong, that the W3C makes its money from the browser vendors “opting-in” to the groups via paid membership.

    From your proposal, I see the W3C putting out more money than it would be making.

    Do you have any real-world suggestions, so the W3C has enough of a cash flow to support so many “employees”?
    (As you proposed, Web Designers such as yourself would be paid your “normal” daily rate, which will vary from designer to designer, and some (read: you) are on the higher end of the spectrum, and it will mean putting out big bucks for the W3C)

    Other than that, I like most of your ideas for improvement, and while I don’t think the whole Opera vs MS is the defining factor, I do agree the W3C needs to change the current processes.

16. #16 On December 16th 2007 at 07:20pm Lachlan Hunt said:

    One major area that this article didn’t seem to address at all is test cases.  To improve the quality of the specs and implementations, writing a thorough test suite that tests every single feature of the spec needs to be a top priority.  With a good test suite, browsers have much more chance of interoperably implementing the specs, they ship fewer bugs and ultimately make the lives of web developers easier.

    In fact, I would say this is more important than getting the specs published as recommendations, since that is dependent upon, and absolutely must follow, browser implementations.

    The problem is that writing test cases is extremely time consuming and there are relatively few people who are able to write good ones.

17. #17 On December 16th 2007 at 07:35pm thacker said:

    Jon Hughes asked,

    “Do you have any real-world suggestions, so the W3C has enough of a cash flow to support so many “employees”?

    The W3C needs to go walking into a reputable investment banking firm for these answers.  They also need to visit a few business schools for advice.  There is not one reputable investment banker, b-school or major accounting firm who wouldn’t welcome Berners-Lee with open arms and provide him with services, probable solutions and advise pro bono.

    Those answers, including restructure of operations, need to be provided by professionals and not those of the ‘geek’ variety or those without sufficient knowledge [myself included].

18. #18 On December 16th 2007 at 07:40pm Jim Jeffers said:

    I agree whole heartedly.  If the what Daniel says is true -

    “It’s a BATTLEFIELD where vendors fight for competitive advantage. And that fight is not quiet or non-violent.”

    ...then there is a clear conflict of interest within the current process. 

    I’m glad your post is bringing this discussion out into the open because many of us in the profession have no clue as to what goes on in these organizations or how they are structured.

19. #19 On December 16th 2007 at 08:01pm Wilfred Nas said:

    Andy, great article and a (imho) good proposal…

20. #20 On December 16th 2007 at 08:31pm Alan Gresley said:

    @ Andy Clarke:

    “Daniel, your lack of inspirational thinking surprises me. As a designer, my job to interpret what my clients need and work hard to find a solution. Your job is to build a browser that interpret whatever CSS features we need to satisfy our clients, CSS that has been designed and developed by people who use it everyday.”

    Andy since you are a designer, I think this is why you seem to have a lack of understanding of how browsers work and it’s make this reply to Daniel look naive.

    Why should your clients wishes be part of the criteria for what features of CSS3 the CSSWG should be working on?

    Also, if you have never made a browser, how can you say that David’s reply is lacking in “inspirational thinking?” I may be naive myself but I have read a bit on how hard it is for browser implementors to implement CSS2.1 correctly. Don’t you think you should heed the advice of a browser implementor? Have you asked the advice of Håkon Lie?

    Lastly, some of what you have suggested is good but the way this message has come across isn’t good.

21. #21 On December 16th 2007 at 08:35pm Kilian Valkhof said:

    If anything is clear, from both sides, is that change is due. Nobody wants his tools to be made on a “battlefield”, yet nobody wants insane ideas that take way too many lines of code just because a group of designers thought it would be nice.

    As is usually the case, the middle road seems the best option, but I don’t think that’s possible. Big changes come from revolution (Like Andy proposes), not evolution.

    Beside this fundamental discussion, there is also the stuff surrounding it, like cost, management, test cases.

    You’ve stirred something up, and I sincerely hope you and the rest of us will be able to ride it out.

22. #22 On December 16th 2007 at 09:07pm zeldman said:

    The amount of time the W3C requires invited experts to put in without remuneration has always been a problem and has always prevented many designers from participating.

    But an external audit is something else again.

    As a good way of demonstrating that it cares about its users and wishes to avoid becoming irrelevant to them, the W3C could put our an RFP for a usability audit. Said audit to be conducted by an agency that excels at performing such audits; is familiar with the issues and history involved; and is good at design and web standards.

    It could work.

    And then of course, the W3C would need to act on the audit’s recommendations.

    That’s how we do it here in the real world. It’s what we do when a corporate client has a problem communicating with its target audience. Seems to apply.

23. #23 On December 16th 2007 at 09:07pm Daniel Glazman said:

    @ Andy : “Your job is to build a browser that interpret whatever CSS features we need to satisfy our clients, CSS that has been designed and developed by people who use it everyday.”

    ah. As you know, XUL and XAML user interfaces languages are or can be styled through CSS. Ask yourself how many times the CSS resolution algorithm is called between the moment you click on a XUL or XAML-based application and the moment your window appears on the screen. Just try to reach a realistic number, and you’ll understand or at least I hope you’ll understand the browser world is not that easy. If it was THAT easy, let designers and marketing define what is needed and let’s order the geeks to code it, trust me, it would be already implemented and shipped !!!

    @ Lachlan : writing a test suite that tests - and then prooves an implementation - of all CSS is plain impossible. The number of possible interactions between all properties and property values is just impossible to test.

24. #24 On December 16th 2007 at 09:23pm Alex Mogilevsky said:

    I actually think that treating CSS development as a project is a good idea.

    As a project, it needs goals and priorities (and you’ve already mentioned timelines). Setting those will be an interesting proces… Who sets the goals, what kind of consensus or not will it take to decide on what’s in and what’s out… I share Daniel’s skepticism to a degree, but I am also honestly intersted in seeing a public process create and drive this kind of project…

25. #25 On December 16th 2007 at 11:03pm Andy Clarke said:

    @ Daniel Glazman: “As you know, XUL and XAML user interfaces languages are or can be styled through CSS. Ask yourself how many times the CSS resolution algorithm is called between the moment you click on a XUL or XAML-based application and the moment your window appears on the screen.”

    I’m sorry, what did he just say?

    Daniel, I respect your position although you’ll have guessed that I don’t agree with it. Now, if you can tell me with your hand on your heart that you think that the current methods for designing and developing the CSS ‘product’ are successful for all concerned and that nothing needs to change, then I’ll stop goading you.

26. #26 On December 17th 2007 at 12:20am Kevin Lawver said:

    Instead of calling for the mob with their pitchforks and torches, how about coming to a face to face meeting?  How about contributing to the test suite for your favorite CSS3 module?  How about helping to explain to the web design and development community what really goes on in a standards body instead of lobbing grenades - because that’s exactly what you’ve done.  This isn’t helpful.

    I don’t agree with the strategies of the browser vendors…  most of the time.  I have serious problems with how they work inside and outside of the working group, but, I have the utmost respect for the members of the CSS Working Group and wouldn’t call them out like this without discussing with them first your concerns.

    I have so much more to say, but dammit, I’m on vacation.  I wish you’d handled this better. You could be educating instead of instigating. Once people understand what goes into a specification, they can better understand where we can all help the process move faster.  Instead, you’ve chosen the easy way out.

    Yelling “fire!” in a crowded theater isn’t the same as leading a revolution.

27. #27 On December 17th 2007 at 12:21am Maciej Stachowiak said:

    I agree that the CSS standards process has issues, and could be improved. And it’s true that browser vendors may have conflicts of interest. However, I think your suggestion is off-base in a couple of ways.

    1) You assume that users of a technology are better placed to decide how it should evolve than implementors, to the point that implementors should be relegated to a token advisory role. However, that’s not the way the rest of the technology world works. Some examples:

      1.a) CPU instruction sets are designed by hardware designers, not software engineers; although they consult with compiler developers and study the needs of software engineers and end users to.

      1.b) Standards for the C and C++ language are designed by committees with heavy representation from compiler and library vendors, also taking into account the needs of users of the language.

      1.c) TCP/IP and related are designed by Working Groups with heavy representation from vendors of OS networking stacks, routers, and so forth, considering the needs of software layers above them and users.

    And so it is with CSS. Being an expert user is not necessarily enough to know how to extend it. A key part of extending a technology is knowing how it works on the inside. Furthermore, being an implementor gives you much more in the way of designer input than any small group of expert designers can provide. You hear all the time what kind of problems they run into. So, while input from users of a technology is great, input from implementors is essential for it to succeed.

    2) Browser vendors provide many of the most active contributors in web standards working groups. Removing the most active contributors with a hypothetical plan to replace them with people who have not been as active seems like a bad plan, if your goal is to speed up progress. When a project is in trouble, you don’t fire the top contributors and replace them with unknown quantities. Project management 101.

    3) Standards (or specifications if you prefer) are not software projects. Managing them the same way does not make sense. Software engineering management involves constantly making careful tradeoffs among features, quality, time to market, resources put into the project, and many other factors. Sometimes these factors are out of whack and you fix it in the next version. But the cost of a serious mistake in a standard is much higher than the cost of an implementation bug. By the time you notice it’s a problem, you have implementors and designers relying on it, and a process that is heavily stacked against making incompatible changes to existing specifications.

    4) Web standards with limited browser vendor input have historically been poor - overly complex, and with little consideration for practical issues or what web developers want.

    5) Most important point: standards that matter are standards that ship. It doesn’t matter how beatifully designed a spec is, how fast it’s done, or how much it needs the needs of web authors if no one ships an implementation. Disenfranchising the vendors who will need to implement the spec is not a good way to get their buy-in.

    Because of points 2 and 5, I’m not really worried that anyone at the W3C will take your proposal to heart. They want people to work on the specs, and they want the specs implemented. But I thought it was worth explaining why it’s critical to keep implementors a core part of the process.

28. #28 On December 17th 2007 at 01:31am Chris Blown said:

    I too, am a naive fool it seems. But I understand the battlefield paradigm, it is probably always going to part of the process no matter how well its structured. Having said that, Andy touches on a valid issue, a negative attitude from “implementors” is not constructive. Implementors are always happy to explain why X can’t be done, rather than looking at ways of getting X to work. A shift in attitude here would make a awful lot of difference on the battlefield.

29. #29 On December 17th 2007 at 01:31am Michael Montgomery said:

    Alan Gresley said:

    “Why should your clients wishes be part of the criteria for what features of CSS3 the CSSWG should be working on?”

    Breathtaking question.

    When Andy Clarke refers to “what my clients need”, he is referring to clients in general, which represent everyone who uses and builds the internet.

    So, does Mr. Gresley actually believe that the interests of clients who use, enjoy, and pay for sites, applications and everything else on the web are irrelevant? That standards should only be determined in a browser builder cage match?

    I hope I’m missing something, because the suggestion that’s “the way it’s been done” is one thing; suggesting that it’s the best way, or the only way, is ... breathtaking.

    No wonder Microformats are so exciting, and so many have even turned to parochial proprietary stuff like Flash.

30. #30 On December 17th 2007 at 02:15am Gareth Rushgrove said:

    Maciej said a few interesting things above I thought worth commenting on:

    “Being an expert user is not necessarily enough to know how to extend it. A key part of extending a technology is knowing how it works on the inside”

    I think their-in lies an opportunity. The weight of comments on this issue, and the heat of the argument, would suggest people are interested. But I think a-lot of those interested parties are those working day-in with CSS (rather than on it) and lack this critical information. As a web designer how do I find this information out? Who are the interesting bloggers in the area of standardisation? Is their are good reference site? Or a wiki? Or an active friendly mailing list of people willing to answer simple questions? The web (standards) design community is used to open discussion and debate - whether from blogs or projects like microformats.

    You could of course also argue that being an expert implementer is not necessarily enough to know what needs extending. A key part of extending a technology is knowing what is needed first in the real world.

    “But the cost of a serious mistake in a standard is much higher than the cost of an implementation bug”

    I think this depends on the software in question - what about critical military systems, satellites or space shuttles, power stations, etc? - but I agree in principle. Most web designers do not have the experience of working on this sort of software project. But then Andy is not putting himself forward as a project management guru.

    Their are obviously strong feelings on both sides of this debate. But something I would be interested to hear - from someone with a standards focus - is whether you think their is a fundamental problem at the moment with regards CSS? If not then why not? If so then what are your ideas for moving forward?  And feel free to say you don’t know yet. I think what Andy has done is put down his ideas. If you don’t like them then suggest better ones. If that’s sticking to the status quo then say why you feel will work out.

31. #31 On December 17th 2007 at 02:56am Maciej Stachowiak said:

    “But I think a-lot of those interested parties are those working day-in with CSS (rather than on it) and lack this critical information. As a web designer how do I find this information out? Who are the interesting bloggers in the area of standardisation? Is their are good reference site? Or a wiki? Or an active friendly mailing list of people willing to answer simple questions?”

    The best way to learn about CSS implementation issues is to get involved with one or more of the implementations. That’s hard with closed-source engines. But fortunately there are two very high quality open source implementations, Gecko and WebKit. By getting involved I do not necessarily mean learning C++ and hacking the internals. Some people will have the time and inclination to do that, many (especially if they’re coming from the web design world) will not.

    But a great way to get involved that teaches you a lot about implementation issues and about the spec is doing standards-driven QA. By this I mean creating test cases, spotting implementation bugs, helping to integrate test suites with regression testing frameworks, etc. Many of the top web standards experts got their start in browser engine QA, I would even say this might be a more useful background than coding on the implementation.

    For the WebKit project, we are very welcoming, will answer questions, and are enthusiastic about standards compliance, testing, and advancing web standards. I am sure the same can be said of Gecko and the Mozilla project. For more info, see <http://webkit.org> and <http://www.mozilla.org/developer>.

    I’m also sure there are other ways to gain web standards expertise, but I’m a “learn by doing” guy myself, so that’s the path I’m most familiar with.

    
    ““But the cost of a serious mistake in a standard is much higher than the cost of an implementation bug”
    I think this depends on the software in question - what about critical military systems, satellites or space shuttles, power stations, etc? - but I agree in principle.”

    You have a point, but military grade software follows much different processes than mainstream software projects, for this very reason. Milspec software systems tend to be developed even slower than what would be desirable for a web standard. I would say the IETF process hits a pretty good balance of progress, pragmatism and caution.

    
    “Their are obviously strong feelings on both sides of this debate. But something I would be interested to hear - from someone with a standards focus - is whether you think their is a fundamental problem at the moment with regards CSS? If not then why not? If so then what are your ideas for moving forward?”

    Yes, I think there are some problems. Ultimately the biggest problem is lack of capable and willing editors and test case authors. Partly that’s because most of the people with those skills work for browser vendors and often have other time commitments. But another part of it is that it’s not obvious how to get involved with the CSS working group at a casual level, because of it’s semi-closed nature. Insufficiently open process can also lead to bad decisions, design thrash, and time wasted on pointless ideas.

    Here are some moves that I think could help:

    1) Re-charter the CSS Working Group as a public working group, as called for by Ian Hickson, David Baron, and others. This will make the inner workings of the group more visible, and make it more open to outside feedback. This is ultimately up to the W3C though, and out of the hands of our blog arguments.

    2) If you believe implementation of current standards, and not just development of new ones is a key issue, help make standards-based test cases to donate to official test suites.

    3) If you’d like to gain expertise about CSS implementations (and more expertise on use of CSS than you thought possible), get involved in standards-driven QA for a browser engine. A couple of hours a week on one of the open source engine can teach you a lot, and will do a lot to move web standards forward.

32. #32 On December 17th 2007 at 04:32am Robert O'Callahan said:

    I agree with everything Maciej said.

    If you think designers need more sway in the CSS WG, fine. Get more designers involved either by joining the WG or by opening the CSS WG more so they can be involved without joining. But there’s no need to remove the browser reps. The more active participation there is (excluding trolls and idiots), the faster CSS will move.

    Maciej’s point about getting involved with open source browser engines is an especially good idea. By focusing on Microsoft and Opera you have portrayed “browser vendors” as warring profit-focused citadels, but that’s not how it is with Webkit and Mozilla.

33. #33 On December 17th 2007 at 03:44pm Ben Darlow said:

    So; timetables for the creation of specifications. Exactly how useful are these without timetables for implementation? Timetables for implementing these specs are vastly more important to me than timetables for writing down wishy-washy specs that may or may not ever be implemented. Oh, and guess what? This is the root of Opera’s complaint against Microsoft.

    Frankly, I think all this effort is misplaced. Specs are meaningless without implementations, and we have yet to see a good many old specs implemented by the incumbent browser. New technologies designed for a future web that has evolved beyond what we see and use today are all good and well, but if the W3C imploded tomorrow and there were never any more specs, the web and the development industry it supports would not disappear. Right now the biggest challenges the majority of front end developers face now can be summarised in two words: Internet Explorer.

34. #34 On December 17th 2007 at 04:14pm Gareth Rushgrove said:

    I’ve got to disagree with Ben on some of his points; Internet Explorer can be a pain in the proverbial but, over the last few years, we’ve learned to live and work around it. It might not be pretty, and it might take more time and cost more money but it’s doable when working with good design*.

    What I want now are new, improved features and functionality. I want new toys that allow me to do things I can’t do now, rather than just the ability to do things faster.

    As for timescales, recently at least, Microsoft has shown that they intend to follow. So that means two of Opera, WebKit or Mozilla getting to compatible, working and tested implementation of the modules we want - quickly. The timescales are important to set the focus; which modules come first - not just in one of those, but all of them. This needs to be based on a good understanding of what designers want and which, if any, are low hanging fruit. I want proof that CSS3 really is modular. I want a clearer picture of what is going on and what the community can do (their are an awful lot of us remember). If some of Andy’s suggestions help with that then viva la revolution.

    * I’ll agree this causes a huge problem for those learning the craft of front end web design and development.

35. #35 On December 17th 2007 at 05:09pm Ben Darlow said:

    @Gareth: Opera, Apple and Mozilla’s efforts are all in vain if the browser that the majority uses doesn’t follow suit. This was my main point – there are *already* loads of useful features in CSS that these three support that IE doesn’t. Imagine how much nicer the web would look - and how much simpler page markup could become - if all the ‘latest’ browsers supported border-images? Forget new proposals; we need *that* stuff implemented first, otherwise it will forever remain essentially theoretical.

36. #36 On December 17th 2007 at 10:36pm Lars Gunther said:

    Adding to Maciej, comment 27:1

    Are the manufacturers of wireless products not the primary players in developing 802.11? Are not the manufacturers of NICs, switches and routers the primary players in developing 802.3? Should their involvement be degraded and only network technicians be allowed to actually come up with new ideas?

    Should the guys at x.org be degraded and not allowed to work on the X11 spec, but as consultants?

    And should IBM, SUN, HP, et al not be allowed to work on The Single Unix specification (formerly known as POSIX) - except as consultants?

    And while we’re at it. Let’s cut Brendan from ES4, since he works for Mozilla.

    Andy, I have the utmost respect for your CSS skills. You are a true guru. I am not. I am a jack of all trades. (I teach PHP, front end dev, PHP, databases, CCNA, Linux, multimedia as well as religion and math - and then some.)

    But I teach a lot of standards that have been developed by manufacturers (CCNA 1 is just one standard after the next…) - and in many cases they have started out with competing ideas and visions. Coopetition is the name of the game!

    More transparency? Yes!

    More designers and other front end developers that participate in the CSS WG? Yes!

    Cutting the vendors? Won’t help anything.

37. #37 On December 19th 2007 at 09:16pm thacker said:

    Have been looking and waiting for any type of response from the W3C.  So far, nothing that I have found.  “The most important thing in communication is hearing what is not being said.”  In this case, volumes are being spoken.

    Bring on the private APIs, the private frameworks, the stand-alone RIAs.  Bring it on without participation of open standards and the W3C.  Bring on the confusion that will make the ‘browser wars’ appear to have been kindergarten antics in comparison.  Bring on the market forces that will declare what wins, what loses and by de facto what becomes a ‘standard’.  The Adobes, the Googles, the Microsofts, the hungry developers ... address the markets and let the ‘white elephants’ be.

    In ten years, with luck in five, Internet communication will, hopefully, become what it should.

38. #38 On December 20th 2007 at 12:07am Deborah said:

    Lots of discussion points have already been highlighted by others, so I won’t repeat. However, your ending reference to “It ain’t me babe” by Johnny Cash took me by surprise, since the song is by one of my favorite singers, Bob Dylan.

    Written by Bob Dylan, “It ain’t me babe” was included in his Another Side of Bob Dylan album released in 1964. Johnny Cash asked permission to record it, and in 1965 recorded the song with his wife June Carter.

39. #39 On December 20th 2007 at 12:31am fantasai said:

    @thacker: There have been lots of responses from W3C in this discussion, albeit all unofficial personal responses and none from W3C Staff. Alex Mogilevsky represents Microsoft in the CSSWG; Kevin Lawver represents AOL; David Baron represents the Mozilla Foundation; Daniel Glazman, a long-time member, currently represents Disruptive Innovations; and I myself am a W3C Invited Expert since early 2004.

40. #40 On December 20th 2007 at 12:38am Andy Clarke said:

    @ Deborah: You got me babe. Yes Bob Dylan did write it, but Johnny Cash recorded my favorite version.

41. #41 On December 20th 2007 at 01:58am Deborah said:

    Aha! Bob Dylan’s version is my personal favorite.

42. #42 On December 20th 2007 at 02:07am Ola Muldal said:

    It’s nice to see this issue well articulated from someone involved. But it is all like 5 years too late. Do people really think i.e. Microsoft need to care, when they are rolling out SilverLight 2.0, which will be totally their own playground.

    I abandoned DHTML totally when Flash 5 came with support for XML, and never looked back.

    The key is that Adobe listens to its customers. I wont bother to evangilize what the Flashplayer platform has evolved into, but i recommend people unaware to go and check it out. Imagine if W3C had had the same goals and pace.

43. #43 On December 21st 2007 at 11:31pm anucha 'elias' said:

    Mr. Clarke,

    Thank you for sharing your thoughts and insight, which I respect, and shedding light on the frustrations I’ve been dealing with during my studies and future career.

    I’m at university in the states, so please pardon my lack of technical jargon, if you will.

    I started graduate studies last year at UB - Web/Graphic Design; Interaction Design & Information Architecture.

    Since my first class [9/07] everything has been about “Web-standards,” “CSS,” designing for future and backward compatibility.

    Now, we’ve come to find out everything we’ve learned “Isn’t the ‘Way.’ ” Microsoft has thrown one too many wrenches into the mix, from email, XHTML, HTML, and now, CSS. But, it’s not Microsoft’s fault, yeah? From our perspective, it seems HTML 5, XHTML 2 and CSS 3 working group’s can be bought, at least from what I’ve read? If you don’t believe me….Google it.

    My point: When are browser manufacturer’s, these working groups, and ‘experts’ going to do “What’s right?” Do what’s right for the people who use the web. For the designers. The developers. It’s a New Year…if you were to take the high road, maybe this could be your ‘Competitive Advantage?’ Make the Internet a better place to live, work and play. “Just Do It.”

    Enough with the nonsense -sorry Andy-and put an end to these wars, which don’t seem to be over with, which we read about too.

    Just my .2 quid’s worth.

    Again, many thanks to Andy, Simon Collison, Dan Cederholm, Dave Shea, Jeffrey Zeldman, Ian Lloyd and Rachel Andrew for inspiring me, day-in and day-out. And for giving me a creative outlet and new career!

    Cheers and Happy Holidays!

    aea

44. #44 On December 22nd 2007 at 03:06pm Sean DALY said:

    Greetings, I have read your comments concerning the Opera complaint with interest, so much so that I asked the Opera principals about it in an interview:

    http://www.groklaw.net/article.php?story=20071220115451997

    which may interest you. The day before I also interviewed Opera’s lawyer in Brussels which may be helpful for understanding the legal context relative to the complaint.

    The FSFE has written to EU Commissioner Neelie Kroes in support of this complaint. You may be aware that the FSFE was an intervener in the Microsoft-EU case which Microsoft lost on appeal in September.

    Thanks.

    Sean

45. #45 On December 22nd 2007 at 03:06pm thacker said:

    fantasia stated:

    “[...] albeit all unofficial personal responses and none from W3C Staff”

    Madam, I apologize beforehand and I do not take anything away from your contributions to browser development or from standards development.

    However, designers/developers who are involved in tactical development and implementation cannot and do not address strategic and operational issues.

    Again, volumes are being spoken by what is not being said.

Leave your comment