The HTML 5 circus: Why I left and rejoined the W3C HTML Working Group

On July 1 2007 I decided to leave the W3C HTML Working Group. My reason was not primarily lack of time due to my new phase of life, as some have guessed.

Sadly, I left for much more troubling reasons. I left because I was fed up with the dismissive and hostile attitude a number of individuals in the working group have towards those who do not fully share their view of how the future of HTML should be shaped.

I was sick of watching as people were being ridiculed, ignored or dismissed when they asked questions, voiced objections, or otherwise tried to contribute. It happened on the public-html mailing list, on #html-wg, the working group’s official IRC channel, and on #whatwg, WHAT WG’s IRC channel. By the way, both IRC channels have public logs.

Things had gone far enough that a non-negligible number of members (including myself) were reluctant to voice their opinion or make suggestions. It even made some people decide to leave the working group. This, of course, is a completely unacceptable and unproductive working environment.

So I left. Which was probably good news to some, since it meant one less best practice advocate to deal with.

But being of the curious kind, I couldn’t stop myself from keeping an eye on the mailing list archives. Some of the problems were brought to light by a number of other working group members and discussed. After that, things seem to have improved slightly.

It has also become very clear to me that I am far from the only person to think that the current process of editing the specification for the next version of HTML is… less than ideal.

I suppose I could just give up, keep going as if no problems existed, wait for HTML 5 to be done, and then deal with it, provided I still work in this business in the year 2020. Maybe that is what I should do. But I don’t like giving up on things I am passionate about.

So I’m giving this a second chance. The other day I rejoined the W3C HTML Working Group. This time around, however, I will be taking the opinions of people who seem to lack experience of real-world Web development and apparently are uninterested in actually improving the Web with a truckload or two of salt. Hopefully that will help me keep my temper.

Looking forward I think my energy will be best spent helping to produce documents that are useful to and readable by people who create websites. By the looks of things the actual HTML 5 specification will be extremely difficult to read, and basically created for browser vendors only.

Tutorials and best practice documents for Web professionals will be desperately needed. Suggestions on what you would like to see in such documents are welcome.

Posted on August 16, 2007 in (X)HTML, HTML 5, Rants, Web Standards

Comments

  1. Way to stick to your guns!

  2. We definitely need people who are willing to be advocates for standards and best practice. Thanks for helping make HTML 5 a bit more exciting for me to look forward to Roger!

  3. August 16, 2007 by Frank

    So many people driving HTML5 have no real world experience. Many of them are in their early 20’s with only a few months working experience at Opera Software’s testing department (Anne, Lachlan and even Ian Hickson). When you’re that young, you think you know everything. One should have some real experience before one tries to re-design Web architecture.

  4. August 16, 2007 by Alejandro Moreno

    I am not experienced enough to join and participate, so I sure am glad you have decided to stick around the HTML5 WG.

    Your “translation” efforts will be appreciated!

  5. August 16, 2007 by madr

    Truly sad to hear about the whole situation. But I’m glad to hear that you give it a second chance. According to what I’ve read from you, you are one of the really important voices when it comes to future HTML versions for web development novices like me.

    Du har mitt fötroende.

  6. August 16, 2007 by Laura

    Terrific news, Roger! Welcome back. The working group needs more people like you.

  7. Thank you. You speak for a lot of us out here who appreciate your work for best practices.

  8. I’ve also been reluctant to participate due to the recent nonsense going on the HTML WG mailing list. There is way to much noise and this makes paying attention to the good stuff a bit of a chore. I don’t feel that I’ll leave the working group but my reading has become way more selective. The chances of me adding some reviews or documentation has diminished.

  9. Having someone take the lead on the tutorial writing would be awesome. I’m glad you’ve changed your mind and are coming back!

  10. You said: “So I left. Which was probably good news to some, since it meant one less best practice advocate to deal with.”

    Which made it that much more important to go back.. :wink:

    Best wishes.. :)

  11. Ahh, you came back to it Roger! ;)

    I am glad that decided to keep monitoring its progress one way or another.

    “Suggestions on what you would like to see in such documents are welcome.”

    Hey, if I think of anything, I’ll let you know!

    I’m really hoping that the frustrations raised by you and others will initiate some positive improvements to how HTML 5 will be rolled out (what it will feature, and when).

  12. Hixie’s in his early 20’s? Blimey.

    From my vague recollections of reading stuff, I think a main aim of the HTML 5 spec is to make it much friendlier to browser vendors, who benefit from insane (to us) amounts of technical detail. It’ll make interoperable implementations much more likely (i.e. Internet Explorer and Firefox will do the same thing, first time).

    As such, I can imagine the spec is hard to follow from an HTML author’s point of view. We want to know what an element means, to people, and how we can use it.

    Although we all groaned at the multiple huge WCAG v2 documents, I can’t imagine one HTML5 spec serving both browser makers and HTML authors very well.

  13. Good for you, Roger. Much respect for staying in the fight.

  14. Sadly this almost happened to me. I joined the group with great expectations but there have been lots of problems that almost made me leave. I’m still there, reading most of what goes through the mailing list, but most of the discussions are so convoluted that is almost impossible to help or add some constructive opinions. I even had a wiki edit “censored” (but never looked who did it) by no reason as I was just adding information with reference to one mail in the list archives… Let’s just hope things clear up a little and the group start to work as intended :)

  15. August 17, 2007 by Rob Burns

    In reply to pauldwaite:

    You’re right that HTML5 tries to specify browser behaviors in a much more complete way. However, I would say that may also require more specification for authors too. And we really want one document to be authoritative no matter how many tutorials are produced from that document. Also, the W3C has set the bar fairly high in making documents readable (which is why when exceptions occur they get so much attention).

    I think we need to make the HTML5 recommendation readable for both authors and implementors. I also think that by using the @class attribute, CSs and javascript and maybe even XSLT transforms) we can make that document easy to tailor to whomever is reading it or printing it. So while there will, no doubt, be lots of room for tutorials, I think we should be able to make the recommendation itself very readable for authors. Remember this is only the first draft (for the W3C anyway).

  16. Keep up the work. Everything I put onto my site gets validated, if it fails I edit it or remove it. Some people may not think much about standards and validation but I do and so your helping me. Thanks

  17. Glad to see you’re back in! Good luck!

  18. August 17, 2007 by Brian LePore

    In reply to pauldwaite:

    Has Microsoft made any public statement about their thoughts on HTML5? I know Opera, Apple, and Mozilla are backing this work and will work to implement the language ASAP, but without Microsoft’s support it seems like HTML5 will only be theoretical at best. Heck, I would be presently surprised if anything HTML5 related was supported in IE8 or even 9.

    Not that I don’t want to see HTML improved upon. Work obviously needs to be done, but it seems odd for the smaller players to make up rules and then assume that the largest player will follow them once all is said and done.

  19. August 17, 2007 by Dr Livingston

    Good for you. I have seen situations just like what you’ve described and I just cannot be bothered with that.

    I walk off and just leave them to it; You know what? The world actually goes on, without them and it’s far better off without them anyways.

  20. @Brian LePore

    a) Chris Wilson of MS is chair of the W3C HTML Working Group. That is a sign of at least some commitment. But considering how far behind they are it will take a while until they have implemented most features.

    b) Until then there are multiple efforts to make HTML 5 features work in MSIE through JavaScript. Such as this one: http://sourceforge.net/projects/wf2/

  21. How do you write a tutorial for an unstable document?

    I had volunteered for writing tutorials but withdrew my name from the offical list for various personal reasons as well as the impracticality of writing tutorials for sections whose browser implementation requirements have not been resolved. [Example: “Section”. How will UAs implement this? and, when will they implement this?] It would be different if the requirements were stable and only the spec’s verbiage needed minutiae discussions (or, arguments). Further, the Conformance Checker is far from complete. Presently, most of the errors are not truly errors but unresolved issues which are stated as being an error but ignored at this time.

    I do not see any reason for writing tutorials until next year. I’ll put my name - Officially - back on the tutorial writing list next year.

    As for the age of HTML WG participants, I don’t care. It doesn’t matter to me as long as they have social interactive skills - which most participants do not have. Of any age. What’s interesting about reading shrill, pipedream HTML WG mailing list comments is that - Seemingly - very few of those participants have ever had dealings with the real world of web development or web authoring.

    I only read a few authors messages as they come into my inbox. Everyone else, I read complete mailing list threads over the week-ends.

    As for your question about what tutorials or best practices you should write…. That’s a difficult one. How about accessible data tables with HTML5? for a start.

  22. It’s a serious problem when a good number of people on the HTML5 WG have no real world experience or their experience is very narrow. Age should not matter but age and experience often go hand-in-hand.

    As for me personally, I am not getting involved in HTML5 mainly because I don’t feel I can take pride in this spec. 10 years from now I don’t want to say to people that I helped create this tag-soup maker.

  23. I’m glad to see you’re not entirely happy with the legibility of the forthcoming document; this seems to be turning into a bigger problem as the years go by.

    Using CSS as an example: CSS 1 fit on a single (long) web page. CSS 2 was pretty verbose but still understandable. Now the proposed box model module of CSS 3 is longer than CSS 1 in its entirety. (I’ll admit some of the CSS 3 modules have lots of examples, but they are still full of incomprehensibilities.)

    This is an often overlooked facet of ‘accessibility.’ It’s not just about blind people, it’s also about people who aren’t as intelligent as, for example, Messrs. Çelik and Hickson. (Which is a good number of us!)

  24. “This is an often overlooked facet of ‘accessibility.’ It’s not just about blind people, it’s also about people who aren’t as intelligent as, for example, Messrs. Çelik and Hickson. (Which is a good number of us!)”

    Yes I agree with that statement entirely Michael… and why I keep pushing the idea forward that unless the ‘normal in the trenches small developer’ can economically read, understand and implement this stuff then it will be a hard barrow to push on us. I will need information which can be utilised in “that tomorrow” in the working environment. Its not about the morality of making this stuff accessible to us but about the practicality - if they write it and we just don’t adopt it then they’ve wasted their time!

    Unfortunately there are a bunch of specification gurus who troll every chapter and verse daily and who are in fact very intelligent. They need to not lose sight of the fact some people will be just jumping into web work for the first time, learning new concepts. If I can’t understand something they will just ignore that direction entirely.

    Anyway just my 2 cents I guess. Human readability on my end is an economic necessity.

  25. You wrote: “Suggestions on what you would like to see in such documents are welcome.”

    I’m reading many forums with topic “Sitecheck” and “(X)HTML/CSS”. The problem of the visitors of these forums is how to create a liquid and elastic layout.

    Most questions are “How to realize a li-navigation horizontal/vertical?” and “My div-box is moved down, if I resize the text or window”.

    I think, a good suggestion is a HowTo about a simple liquid and elastic layout with HTML and CSS. Not only small code samples, a tutorial is needed.

  26. August 18, 2007 by mattur

    It’s a serious problem when a good number of people on the HTML5 WG have no real world experience or their experience is very narrow…

    There is also many very experienced people in the WG. I really don’t think lack of experience or technical expertise is a problem with the WG. Quite the contrary.

    Remember, the old WG was closed and the folks involved were very much separate from the RW, with a very narrow academic perspective.

    unless the ‘normal in the trenches small developer’ can economically read, understand and implement this stuff…

    There’s a great article on HTML5 by Elliotte Rusty Harold which may be of interest to folks looking for an easy to read introduction.

  27. Thank you for giving us a voice Roger. You are doing a great service for us all. Let us know how we can best support you.

  28. That is awesome news Roger! I am sure your opinions will be very valuable and I look forward to a richer, more powerful HTML 5.

  29. I came across an interesting post about moving forward with two semi-parallel specs, one intended for browser vendors which aims to maintain backwards compatibility with existing documents, and one a best practice (or “strict”) version for authors which details how modern HTML must be written. This is similar to HTML 4 and XHTML 1.0’s “strict” and “transitional” doctypes, but differs in that it encourages only browsers to implement the deprecated features. Thus it officially standardises best practice, rather than forcing third parties to write tutorials and advocate best practices themselves.

    I just thought it was a very simple but novel idea. I’m still not sure how practical it would be though.

  30. August 19, 2007 by Roger Johansson (Author comment)

    Rob:

    So while there will, no doubt, be lots of room for tutorials, I think we should be able to make the recommendation itself very readable for authors.

    I hope so too, but with the strong, or perhaps vocal is a better word, resistance to that in the group I don’t know.

    Sean:

    How do you write a tutorial for an unstable document?

    I don’t know yet. Perhaps by finding the parts that are least likely to change and start there. But I agree that it is a problem.

    As for the age of HTML WG participants, I don’t care. It doesn’t matter to me as long as they have social interactive skills - which most participants do not have. Of any age. What’s interesting about reading shrill, pipedream HTML WG mailing list comments is that - Seemingly - very few of those participants have ever had dealings with the real world of web development or web authoring.

    Agreed. Age doesn’t matter. Real world experience and social skill does, and I also think those are sorely missing judging by email and IRC comments.

    How about accessible data tables with HTML5?

    Hehe… I suppose that will have to wait until the headers and summary attributes are back. Or maybe the spec can just say “Use HTML 4.01 if you need to markup fully accessible data tables.” :-P

    cal:

    As for me personally, I am not getting involved in HTML5 mainly because I don’t feel I can take pride in this spec. 10 years from now I don’t want to say to people that I helped create this tag-soup maker.

    I understand what you mean. I’m trying to stay involved hoping that I will be able to say that I did my best to prevent HTML 5 from becoming a “HobbyText Mockup Language”, which some members of the working group seem eager to create.

    Matthias: Thanks for mentioning examples. However, I think those are more related to CSS than HTML. That’s another problem - teaching designers and developers how to use HTML and CSS together. Hmm.

    mattur:

    There is also many very experienced people in the WG. I really don’t think lack of experience or technical expertise is a problem with the WG.

    Judging by how the real-world needs of web professionals are ignored by some of the most vocal participants I think it is a huge problem.

    Dylan: I think that is a very good idea. “Authors” should never be exposed to the tag soup parsing that browsers will need to implement. The problem is that best practice is considered as merely “code aesthetics” by some working group members, which clearly shows their lack of real-world web development experience.

  31. August 19, 2007 by mattur

    …the real-world needs of web professionals are ignored by some of the most vocal participants…

    Love your blog Roger, but imho this is pure FUD. Can you give an example of RW needs being ignored by the group?

    “HobbyText Mockup Language”

    It’s depressing the way many standardistas feel the need to deliberately misrepresent and disparage HTML5.

    I believe as more folks read up on HTML5 they will come to realise that much of the criticism of HTML5 has been, what I’ll politely term, somewhat misleading.

  32. August 19, 2007 by Roger Johansson (Author comment)

    mattur:

    Love your blog Roger, but imho this is pure FUD. Can you give an example of RW needs being ignored by the group?

    Fair enough, it may be an overstatement. But still, that is how I feel. You’re asking if the group is ignoring real world needs, and the HTML WG is not yet doing that since it hasn’t really decided on anything. But potential problems include not enforcing stricter markup in the specification because it allegedly makes things harder for hobbyists, and looking at “statistics” to decide how the language should be constructed instead of talking to web professionals.

    I believe as more folks read up on HTML5 they will come to realise that much of the criticism of HTML5 has been, what I’ll politely term, somewhat misleading.

    If any of the criticism is incorrect or misleading it isn’t intentional (speaking for myself), and has to a large extent been caused by the WHAT WG’s problems to communicate clearly.

    I hope I’m proven wrong and that HTML 5 helps improving the web for end users and web professionals alike.

  33. August 20, 2007 by janey

    Heyas!

    I actually joined the html wg based on your suggestion that anyone who thinks accessibility is important should, from a few months back. I’ve contributed absolutely nothing either way because I dislike the way things are being done, but somehow I’m still reading these emails from the list and occasionally wandering into IRC, and I’m not sure why I still do it.

    Because as much as I do want to throw in my two cents because all these things matter to me (even though I don’t do much work relevant to html etc., accessibility (or more commonly, lack thereof) is one thing I deal with quite often), I don’t feel I need to be flamed (whether or not you want to call it that, I think it’s an accurate word) for whatever I say on the list or in IRC.

    It’s currently not as bad as it used to be like even a month ago, but still…there are ways to have a civil and constructive discussion of a proposed change. The list is not representative of any of them. Maybe in the future I’ll have the time, the patience, the thick skin and the attitude to do so. Until then, I’ll be a perpetual lurker. I know a lot of what has happened was blown out of proportion, but some of it never should have happened. And I’m sure my two cents on the situations have been expounded to death by many in the various insanely long threads about them.

  34. August 20, 2007 by mattur

    If any of the criticism is incorrect or misleading it isn’t intentional (speaking for myself), and has to a large extent been caused by the WHAT WG’s problems to communicate clearly.

    Yes, I agree about the communication problems. This imho is an area where WaSP could help.

    The WHATWG are “arch-techies” and as such are not, shall we say, particularly good at presenting technical info in a non-technical way (which is a skill in itself). This is quite common in techie circles where RTFM is almost an instinctive response. But it doesn’t help, especially when info is widely dispersed and not really summarised anywhere.

    WaSP by contrast, are particularly good at communicating and reaching all types of people as the past few years have shown. I hope WaSP can help alleviate this issue, while still providing the web professionals’ POV and constructive criticism.

  35. August 21, 2007 by Frank

    For those people who are enchanted with HTML5, this example of what the WHAT WG thinks is open and inclusive should be an indication of the direction that HTML5 is taking the Web.

    Ian Hickson wrote an application to list the outstanding issues with HTML5. Here is the link:

    http://www.whatwg.org/issues/

    Guess what happens when you open the link with IE 7 (I assume the same happens with IE 6)? You get this message:

    “Your Web browser does not support DOM3 Core, which is required by this application.”

    So much for open and inclusive!

    Roger, glad you are going to re-join the HTML WG. They need more grown-ups.

  36. Roger, voices like yours will be heard and we are glad at your decision. This is really good news for the developer community as a whole!

Comments are disabled for this post (read why), but if you have spotted an error or have additional info that you think should be in this post, feel free to contact me.