molly.com

Monday 13 August 2007

Dear WHAT WG and HTML 5 WG

Henri Sivonen asks the following lucid question amongst the chaotic discussion on this site recently:

On #whatwg various people (including me) commented that they don’t understand your point. Could you please clarify what problem you see with HTML 5 and what would need to be done to address your concern? Right now, this post seems to add to the discord.

Here are some direct suggestions. Please do bear in mind my prior post was not about WHAT WG, nor was it intended to upset people so. Clearly, I landed on a raw nerve. But here are some very simple suggestions to begin with:

  1. UNIFY the HTML 5 lists and all IRC activities
  2. DETERMINE the true key players in HTML5 WG with a fair balance of representation from spec authors, implementors and real-world developers as well as theoreticians and visionaries
  3. DO NOT alter the integrity of what is already in use and in existence
  4. PLACE new features in another release, perhaps a point release
  5. RELEASE only those features that repair widely agreed upon problems
  6. CONTINUE the innovative work, either within the W3C or without it but NOT on a dual path
  7. LEARN to work with existing communities, as they must LEARN to work with you (for example, finding common ground with accessibility groups)

Finding the pathway to these issues is something I had hoped that WaSP, for example, would be involved in more helpfully. That’s part of my frustration. That I aside, I hope this provides a little more clarity to what apparently came across as vague concerns.

Filed under:   general
Posted by:   Molly | 11:20 | Comments (52)

Comments (52)

  1. The “DETERMINE the true key players…” point is something I spend a certain amount of time noodling on, but… “very simple”? I don’t think so.

  2. Dan – good point. Perhaps I should have said “very clear” instead. Let me amend my statement – these suggestions clearly aren’t “simple” in the doing and the getting done, rather, they are more clear than my broader statements prior.

  3. You make a good point. Someone needs to take the leadership role and organize all those involved, make some decisions, and act as fore for good. I think there is an overall lack of concise direction that is making the whole thing look bad.

  4. It’s not clear to me what you’re trying to solve. What’s the problem you perceive with which these suggestions will help?

    Looking at each of your suggestions in turn:

    “UNIFY the HTML 5 lists and all IRC activities” — as editor I monitor several dozen mailing lists, about a dozen IRC channels, and literally hundreds of Web logs, forums, and other sites to get feedback about the HTML5 spec. I don’t see how we could possibly unify them all into one. I don’t really see that we should, either. For example, there is discussion of the HTML5 spec in the Mozilla #developers IRC channel occasionally — if we asked them to stop and to always discuss HTML5 in the #html-wg channel then people would just laugh at us and probably ignore us (and the spec!) altogether. That would be very bad. Even with the WHATWG list and the public-html list, we can’t unify them — there are people on the WHATWG list who simply cannot join the public-html list due to the restrictions that the W3C has on who can join, and there are people on the public-html list who simply cannot join the WHATWG list because of the legal situation of the WHATWG list. Unifying those two lists (presumably by closing one of them) would merely cause us to ignore hundreds of people. (There are 312 more subscribers to the WHATWG list than the W3C list.) Also, note that unifying all the sources would result in us ignoring you, Molly, since you don’t contribute to either list.

    “DETERMINE the true key players in HTML5 WG with a fair balance of representation from spec authors, implementors and real-world developers as well as theoreticians and visionaries” — I don’t understand what you mean by “key players”. As one of the editors I’m taking everyone’s input into account. I don’t think any one person is necessarily more important than any other. In fact I’m basing my work on the quality of the evidence and arguments presented, not on who presents them.

    “DO NOT alter the integrity of what is already in use and in existence” — Ok.

    “PLACE new features in another release, perhaps a point release” — We are nearing feature freeze, so new features will soon start being placed in a future release. This has already happened with Web Forms 2; new forms suggestions are all being filed away under the banner of Web Forms 3.

    “RELEASE only those features that repair widely agreed upon problems” — This is already part of the process; we are throwing out the parts of the proposed spec that don’t fit this criteria. We’ve already thrown out a ton of stuff from the Web Forms 2 part of the spec, for instance.

    “CONTINUE the innovative work, either within the W3C or without it but NOT on a dual path” — This implies that there is some sort of dual path. There really isn’t.

    “LEARN to work with existing communities, as they must LEARN to work with you (for example, finding common ground with accessibility groups)” — Indeed, there are more communities represented and active in the HTML5 work than there have been in almost any Web standards effort to date.

    I encourage you to take a more active role in HTML5’s development. Your input on technical issues is very welcome, and would make HTML a better language.

  5. This is the best focussed plan of attack I have seen on this.

  6. Hey Molly, the suggestions you’ve made in this post – although ‘simple’ from one perspective, are actually ‘good sense and concise’ from the way I see it!

    I second Vanderwal’s remark! This is actually one of the more direct set of suggestions I’ve seen made about HTML 5 (and the HTML 5 WG)!

    Roger (Johansson) also made some good comments, not too long ago, about what he wanted to see happen in the HTML 5 WG too [I’m not posting links to those articles – I don’t need to].

    These seem like fair, well-considered suggestions, and completely logical. Now, wouldn’t it be nice if members of the HTML 5 WG read this article – and took the suggestions to heart (and actually did something not too dissimilar!)

    HTML 5 has the potential to do some really good stuff on the Web – it would be great to see it up and running in the near future….or, well, in this decade would be nice, y’know?!!

  7. to what apparently came across as vague concerns

    They were actually quite specific. They become even more specific when the question of “What can go wrong” is applied to each one of your HTML 5.0 suggestions as it is analyzed upstream.

    Your strategy for HTML 5.0 hit that nail square on the head.

    Maybe people should ask some hard questions of what can go wrong and take the process all the way back to ‘head waters’?

  8. I’m inclined to agree with Hixie. The perceived problems are still vague.

    I actually don’t agree with a lot of the pieces I’ve read complaining about HTML5 and the WG. Patrick H. Lauke and Roger Johansson complain about the lack of understanding and attitude towards accessibility, yet I’ve never seen that attitude on the mailing list.

    What I see is a questioning of the conventional wisdom which is being interpreted as an attack on accessibility, when that’s simply not the case.

    I wonder if some experts simply don’t like being questioned. The fact is that *you* are the acessibility experts on the WG. If you’re asked to explain the ins and outs of an issue but simply return with a hostile response (and vague threats about leaving the WG) then communication is going to break down.

  9. “What I see is a questioning of the conventional wisdom which is being interpreted as an attack on accessibility, when that’s simply not the case.”

    Isn’t this because of the odd way HTML5 is being specified? My expectation would have been that HTML4 would be taken as the basis, and anyone wishing to deviate would need to produce use-examples of where, how, and why, the HTML4 element under discussion needed changing.

    But instead of that, people are randomly deciding that part of the spec is defective, proposing changes, and then expecting experts to provide proof of why the changes shouldn’t be made.

    This is utterly backward – it results in the same fights being fought over and over again. It’s THAT which is getting on the nerves of accessibility experts. We already fought the good fight with XHTML, and with general web practices over the last six years – why are we being made to fight it again? We’re all busy people – and going over old ground because someone who doesn’t understand the spec or real world usage has proposed something incredibly silly is really disheartening.

  10. andy: “Patrick H. Lauke and Roger Johansson complain about the lack of understanding and attitude towards accessibility, yet I’ve never seen that attitude on the mailing list.”

    you’re right, the worst attitude is reserved to IRC, by the look of it.
    but, casting my mind back to what made me initially join the WG…the table headers issue: when i first got wind of the fact that the @headers attribute was going to be dropped, all i heard as rationale was hixie’s quantitative analysis of a large, but arguably non-representational sample of sites, that clearly showed how the attribute wasn’t being used, or used properly. no mention at the time of “we didn’t drop it, we just didn’t include it yet” or “we’re dropping it, but at the same time we’re actively looking for a better way to achieve the same result in a far more reliable way while making it easier for authors to go about their business”…all you heard at the time was that hey, people weren’t using the attribute enough, so out it goes.

    it’s taken a lot of to-ing and fro-ing to get some discourse actually going on the issue, and it’s still not completely resolved from what i can see. and that is just ONE attribute on ONE element. such changes, dropping attributes and elements, adding others, etc, are numerous in the new spec…and frankly a lot of us late-comers (I, for my part, ignored WHAT WG until the official “coming into the fold” of W3C) see what can only be described as arbitrary changes, done on the basis of nothing more than quantitative analysis, by people who may not even know how current assistive technology actually relies on the presence of such elements/attributes. and, as with the microformats discussions, it appears that the onus is on accessibility advocates to extensively defend and demonstrate the value of *keeping* certain things in, while the same rigour and onus of proof doesn’t seem to have been applied when the decision to *drop* things was made.

    combine that with the slightly strange view that arguing about current AT behaviour/support will be often countered with “but once the spec is in place, AT will adapt to the new behaviour” on the one hand, but that at the same time we are looking at present/past authoring behaviour (of authors often just using current authoring tools, with their own behaviour/support) to at least in part inform part of the spec and to “pave the cowpaths”.

    “If you’re asked to explain the ins and outs of an issue but simply return with a hostile response (and vague threats about leaving the WG) then communication is going to break down.”

    just speaking for my own responses: don’t confuse frustration with hostility.

  11. @Matt: No people are not deciding at random. There is discussion and debate, sometimes confusion, sometimes bad responses. I also spend time on internals@php.net and let me tell you, the amount of cordiality there is way below WHAT WG. There are some complaints, but the work is progressing steadily nevertheless.

    @Everybody: There was a learning curve for me when I joined this effort, but there has been no bad attitude from the most prolific members – if one could call anyone that.

    I have weighed all arguments I’ve found concerning headers, axis, scope. I am quite confident that these attributes will be part of HTML 5. Do not give up just because someone is demanding use cases. Use cases are being demanded for everything. Aside, footer, article, canvas or input type=”date” did not come about at random.

  12. I don’t know a damn thing about the actual processes that these groups use but I am seeing a lot of things like “quantitative research” and “scientific evidence” being thrown around. Why is there even discussion to remove attributes of header or alt or whatnot even being considered? Sort of like discussions to eliminate distribution and manufacture of condoms to prevent STDs because a lot of people aren’t using them?

    It appears that the basis and intent of what the HTML 5.0 spec is to accomplish in both fundamentals and real world application has not been done. If that has been done, perhaps it is too broad or not well defined. Perhaps, too, people are just getting carried away.

    Maybe reel it back in a little and keep it simple and stupid?

  13. Well for me, just me as an author, I’ve found taking the time to sit down and read parts of the draft to be well spent. It’s given me plenty of questions to ask and I’ve learned more a little more each time. A lot more than I was learning from the blogosphere prior to inviting myself into the WG.

    I’m now quite comfortable that HTML5 is very much in the early drafting stages and there’s a lot of issues to resolve. I think a little patience is needed at this point in time, to see how that pans out. We’ve yet to see the outcomes from most of the discussions. I’ve seen plenty of frustration, most of it I’m putting down to impatience. (Time may prove it to be more than that, I hope not.)

    I’d like to challenge folks who aren’t actively involved: for every blog post you read about another blog post about HTML5, read a portion of the HTML5 spec. For every comment you make on such a blog post, make a comment on the spec (on your blog if not in the WG itself). Try it.

  14. Maybe if we’d had a bit more of this sort of critical thinking applied to the W3C over the past few years, we wouldn’t have needed the WHATWG.

    For the first time in a decade, HTML is evolving and moving forward. And this is entirely due to WHATWG, not the W3C or WaSP.

    I’m unconvinced by the “sky-is-falling” rhetoric coming from some standardistas, and some coverage of the accessibility issues has been downright misleading, disgracefully so imho.

    There is an elephant in the room most are too polite to mention: many web folks have spent the past few years stridently (somewhat unquestioningly imho) advocating XHTML/the W3C’s old approach. And they’ve done very well out of it. Understandably, they’re upset by the adoption of the WHATWG’s approach.

    Perhaps taking the techie-ness of the WHATWG and combining it with WaSP’s undoubted talent for evangelism could be a mutually beneficial approach for the future.

    The important thing is: HTML5 is moving the web forward – at last. Folks are involved, discussing and questioning where we’re going – at last. Good times.

  15. “Sort of like discussions to eliminate distribution and manufacture of condoms to prevent STDs because a lot of people aren’t using them?”

    Wow, prime candidate for worst analogy ever.

    I don’t agree with many of the proposed drops in HTML 5, but I think it has something to do with removing “bloat.”

  16. I’ve commented on this at Commenting on “Dear WHAT WG and HTML 5 WG” because it wasn’t getting through here.

    Ben Boyle, I’d like to echo your sentiments. Nothing beats joining the Group and getting your hands dirty by supplying research, tending to the wiki and taking part in the surveys, imho.

  17. Montoya–

    ::chuckling:: Hell, Montoya, nominate the damn analogy. Maybe I will end up ranking right there with the likes of Dan Quayle and Mariah Carey.

    Spending time and resources to debate such things as:

    acronym is not included because it has created lots of confusion. Authors are to use abbr for abbreviations. http://www.w3.org/html/wg/html5/diff/

    at this point, doesn’t make a lot of sense.

    All the talk of a semantically correct Web can be argued as being hypocritical in certain circumstances. The logic to add the dialog attribute to improve upon semantics and ignore semantics of the correct use of acronym versus abbr smacks of trying to “carry water in both buckets”.

    In my view, this is just an insignificant example. There are and have been other examples, e.g. discussions to depreciate the alt attribute because it is not widely used. If it ain’t broke don’t be fixin it and keep it simple and stupid.

    The bloat resides not in the specification so much as it does in the amount of time it takes to get consensus on any specification and get the damn thing out into the field. Debating those types of issues as added to the bloat.

    If HTML 5 is that important to everyone for error and recovery handling and security issues, reduce the true bloat, get the basics done and leave the analysis of the other issues for subsequent implementation within incremental releases.

    As far as error handling and recovery, the lack of CMS and authoring tool developers involvement is just damn scary.

    There are much larger issues below the surface of any specification development. That is where the crisis resides and I believe Holzschlag is one of the few people who see it simmering. Asking Holzschlag to specify specific examples by using HTML 5 has diverted attention from those fundamental issues. However, it will improve performance and understanding for the short term and that, in itself, is a good thing.

  18. thacker: Do you have any links to the “discussions to depreciate the alt attribute because it is not widely used”? I have not seen any such discussions, and it’s trivial to demonstrate that it is widely used by authors (e.g. it’s on about half of all img elements, given this data). The only discussions are about how authors should be encouraged to use alt most effectively.

  19. Taylor–

    As a matter of fact, I don’t have any links other than references that have been made by others and that have been presented within Molly’s blog.

    My gut, however, says that I would be surprised that the alt attribute is widely used given that 95% of Web content, in general, does not validate. That said, I will research it.

    Better examples that help to demonstrate unnecessary bloat to achieve a fundamental implementation of the basic spec within a timely manner would probably be the suggestions to depreciate the accesskey element. Accessibility issues should probably be dealt with in an incremental release.

    Another example would be debate over “semantic” issues such as possible inclusion of the font attribute to accommodate the what you see is what you get authoring tools. Semantic clean-up and improvements of the spec should maybe be another incremental release.

    Development of these incremental releases could possibly be done in tandem with the primary spec with introduction of the incremental releases at six month intervals then followed by improvement and adaptive releases.

    Taylor, my primary point that I am trying to communicate [possibly doing a poor job of it] is that fundamental issues exist within both tactical and strategic operations that impact more than just standards development. It is my belief that these fundamental business issues need close scrutiny and a restructuring.

    Again, back to your question and point, if my reference to the alt attribute did not in fact actually occur, I apologize and stand corrected.

  20. Pingback: Perspektiv! | En webbplats på svenska om xhtml

  21. Pingback: Daily misery » Blog Archive » For the love of the web

  22. Pingback: Daily misery » Blog Archive » For the love of the web

  23. Pingback: Turbulence In The Web Standards Community | iface thoughts

  24. Right on Molly!

    You bring up so many awesome points, and direct attacks in this post. It’s great, and I think it would get some old creaky weels moving in both the W3C and the WHATWG.

    Darius J.

  25. “. . . you are encouraged to respond to the follow up discussion, Dear WHAT WG and HTML 5 WG.”

    Dear Molly,

    As I understand it, that “you” might include me. If not, or if what I say here is off topic, I apologize.

    I’m strictly an amateur. I’ve composed a few Web pages at Geocities on some topics that interest me. I’ve spent a lot of time studying w3c standards and accessibility guidelines, and trying to apply them to my pages. Some time ago I started adding a space and a slash to my br elements, in anticipation of upgrading to XHTML. Recently I started seriously studying CSS, and XHTML and trying to convert some of my pages. Then I learned about all this confusion and controversy, and since then I’ve been in quite a turmoil trying to decide what to do.

    At one point I decided to upgrade my pages from HTML 4.0 Transitional, to Strict, and wait for the dust to settle before I go any farther.

    Later, I thought about giving up on the big W sites and discussions, and going to the sites of libraries and universities for ideas about best practices. I did visit some of those sites, and that did a lot to help me get some perspective.

    Today I decided to try looking at the situation in the perspective of the possibilities I see for the future of the world, that I want to invest in.

    In spite of everything, I still see possibilities for the “Age of Aquarius,” that I want to invest in.

    “Then peace will guide the planets
    And love will steer the stars”

    “Harmony and understanding
    Sympathy and trust abounding
    No more falsehoods or derisions
    Golden living dreams of visions
    Mystic crystal revelation
    And the mind’s true liberation”

    In response to the violence, injustice, ugliness and cruelty I see in the world, I’m trying to learn what I can do to help practice and promote peace, justice, beauty, and kindness. I’m planning to think about the implications of that for how I would like to use the Web, and what I would like the Web to be. Maybe that will give me some ideas about what I would like to happen with standards and accessibility.

    Meanwhile, here are some thoughts that came to my mind, thinking about the issues being discussed here.

    —-

    Contents:
    1. How I’ve been using the Web.
    2. Some of my concerns in composing Web pages.
    3. Ideas about what might happen with standards and accessibility guidelines.

    —-

    1. How I’ve been using the Web.
    – I’ve spent a lot of time in Internet discussions, practicing and promoting some ideas about community life, and working with abused and marginalized people.
    – I’ve used the Web a lot as a resource for information and ideas.
    – I post blogs and photos for family and friends, and read their blogs and look at their photos.
    – As I said above, I’ve put up some Web pages at Geocities about some topics that interest me, including international cooperation, my national heritage, and abused and marginalized people.

    2. Some of my interests and concerns in composing Web pages.
    – I’d like them to be accessible to everyone, including people with all kinds of disabilities, in all kinds of circumstances, on all economic levels, and everywhere in the world.
    – I’d like them to be easy and fun to use, for everyone.
    – I’ve tried several times to find html editors that will help me create pages that conform to the standards and guidelines, and every time I’ve given up and gone back to WordPad.

    3. Ideas about what might happen with standards and accessibility guidelines.
    Dear Molly, I have a lot more sympathy for your distress than I see in Jeffrey Zeldman’s “What Crisis?” post, but I do like his summary of the possibilities. “Happily, there are signs that the W3C wants to shed some of its hermeticism and do a better job of listening. Unhappily, there are also signs to the contrary.” “Two groups, one inside the W3C and the other outside it, are working on HTML 5. Their motivations and methods differ. Their work may come together at some point, or it may not.” “One day, people from nice homes may forsake XHTML for HTML 5, making us wonder what that XHTML pony ride was all about anyway. Or not.” “Sensible new standards may yet emerge from the W3C, or from elsewhere, or they may not come at all.”

    The possibilities are endless. I imagine there will be divisions that will impede progress for many years to come. Divisiveness is impeding progress in every part of society, and this is just one more example. As I said, I was in quite a turmoil for a while, and almost lost my faith in the whole process. I’m feeling better now. Whatever comes out of all this, I’ll find a way to work with it. After I think some more about how I’d like to use the Web, and what I’d like it to be, I might have some good ideas for the working groups. Or not. They might use them. Or not. I have my interests and priorities, they have theirs. We can still be friends! If something comes out of it that serves my purposes, all the better.

    I agree with you about trying to get people to talk to each other, and take everyone’s interests and concerns seriously.

    Jim

  26. You make a good point. Someone needs to take the leadership role and organize all those involved, make some decisions, and act as fore for good. I think there is an overall lack of concise direction that is making the whole thing look bad.

  27. You make a good point. Someone needs to take the leadership role and organize all those involved, make some decisions, and act as fore for good. I think there is an overall lack of concise direction that is making the whole thing look bad.

  28. This is really hilarious. Nevermind that CSS was a doomed inception in the first place; maybe people who know what the fuc* their doing, programming language wise, and maybe, uh, rebuild it? harder, better, faster, stronger, even? Perhaps call it something else, something along the lines of PSL (http://www.cs.uwm.edu/~multimedia/papers/jucs/jucs.html)? I don’t know; go nuts!

    Now HTML5 is being thrown around along with CSS3, and pretty soon I’ll be reading Web3.0 on Double XL; this is just madness. I fully agree that better time would be spent working with what we have now (basically what Matt said in much more eloquent & succinct words, both here: http://mattwilcox.net/archive/entry/id/948/, & here: http://mattwilcox.net/archive/entry/id/849/ ).

    I wonder all of this will be silly enough for another group of people to get some real progress done. I’m sorry, I don’t have a decade to wait until HTML5 is magically used everywhere, it’s breakfast, I gotta eat.

    At this point, nothing short of Lewis Black coming in & cleaning house with both the w3c & whatwg would give me any hope :\ .

  29. You make a good point.

  30. Later, I thought about giving up on the big W sites and discussions, and going to the sites of libraries and universities for ideas about best practices. I did visit some of those sites, and that did a lot to help me get some perspective

  31. really thanks bro…

  32. I really think that these working groups need a different set of priorities – instead of toiling for years to produce a (possibly) irrelevant spec, it would be much better for all if they could commit to producing an updated spec every year, or possibly every two or three years at most.
    This would make it easier for everyone else, not least the evangelists: imagine being able to say that IE could only support CSSv1999, when everything else supports at least CSSv2007

Newer Comments →

Upcoming Travels