You, me and the W3C (aka Reinventing HTML)

Some of you may not be aware that the W3C is recognizing that there has been an unfulfilled need for the evolution of HTML, and they are chartering a new HTML Working Group to do this (rather than the XHTML 2 track that the HTML WG has been developing for several years now).  Some more of you might also be unaware that I was asked to be the initial chair of this new working group.  I’ve gotten a bit behind over the last few weeks, but thankfully the W3C extended voting on the charter, so this post is still relevant.

Over the holidays, I noticed Daniel Glazman’s response to the charter, and to my proposed role as initial chair.  I’ve known Daniel for years, and I have a tremendous amount of respect for him.  I don’t take any of his comments personally, as I’m 100% certain they were not meant personally. 

Daniel, you have an objection to a major browser vendor taking the chair of the most visible WG in the Consortium.  I can understand your concern, but I think it’s misplaced.  I don’t think chairship gives any particular power in a W3C Working Group; I believe it carries the responsibility to get things done, and to be objective about the needs and requirements of the other WG members and the public we serve.  I’m quite certain that the members of the WG would quickly complain about my behavior if I did get out of line - the first thing I said to the Chris Lilley of the W3C when he said the new HTML WG was likely to happen and that several of the other W3C staff said they would like me to be Chair was that I wouldn’t even consider it unless the other browser guys thought I was a good choice.  In many ways, at the time (spring of last year) I felt like I was the moderate voice in the middle of the conflict between the WHAT-WG and the W3C in perspective; the WHAT-WG members wanted to revolt and do HTML outside the W3C (as they think “HTML5” is doing now), and I was the one who went to the W3C guys and said, “umm, you realize all the browser vendors are upset and think the HTML WG is on the wrong track entirely?  How about we do something about that?”  As for the press saying “Microsoft puts its hand on HTML”, well, I’d expect the other browser vendors to refute that if it became an issue.  I’ll try to restrain myself from inserting the MARQUEE tag into the standard.  ( :) for the humour-impaired.)

I should be really clear here – I don’t really want to be the chair of the HTML Working Group.  I know full well the likely value of that job in my career growth at Microsoft is minimal – I’ve had to defend that this would be a good use of my time, and I’m not done with getting approval – and I have an inkling of the time it’s going to take – probably 20% of my total work time (that’s less than the W3C estimate).  It sounds like a lot of work herding cats, and I’m well past the idealistic part of my career where that sounds fun in and of itself.  It needs to be done, though, and I don’t see another good person with an objective perspective to do it.

Daniel, if there were another chairperson totally independent from all desktop browser vendors who could be driven about the goals of evolving HTML, understanding of the requirements of large and small entities (from one-person shops to Microsoft-sized entities), could remain objective and were truly independent (including from the WHAT-WG), then frankly, I’m all for them taking the job.  The problem is that I don’t think there are many people who can claim that.  I agree I’m biased with my background, for better or worse – that’s why I’ve had to consider the chairship as a reason that I would be LESS personally involved in the design decisions of the HTML WG, not more.  I’ll likely need a strong Microsoft alternate on the WG since I can’t be as involved.  That’s a cost I have to trade against my confidence that I’d be a good person to get the W3C and the WHAT-WG more aligned and to actually make good things happen in the HTML WG.  (If Lauren Wood wants to take the job, I’d be delighted, as I still hold her as the gold standard of an objective, driving W3C Working Group chair from the DOM WG days.  I think she learned her lesson, though.  :))

Daniel, you also expressed concerns about the 4 yearly face-to-face meetings and conference calls since a lot of WHAT-WG members are individuals who cannot afford travels around the world on their own budget nor can spend a lot of money in costly phone conference calls.  I hear your concern; I’d prefer to minimize the travel and face-to-face meetings for personal reasons anyway, but I would also caution that the WHAT-WG method, where one person who is the employee of one particular company does the bulk of the editorship work and therefore the design, is not a good method either.  I’d also be happy to discuss modifying the "in good standing" rules, and in general figuring out how to allow non-W3C members to have their voice in the development of HTML, because (as I’ve said to many others over the past year) I strongly believe we need to allow independent involvement in the development of W3C standards; however, and this is a critical point – W3C Membership is a responsibility as well as a financial cost.  When you become a Member, you agree to abide by the W3C’s rules, and in particular, the Patent Policy. 

I should be clear here that I am not a lawyer, and the following should not be considered legal advice.  Consider it a layman’s simplification of the issues, but you should consult your own lawyer.

The Patent Policy is intended to provide protection for ALL W3C Members, as well as non-members, in making the specification “free” – free to be implemented.  (Note that I’m referring to Working Groups that operate under the Royalty-Free Patent Policy – like HTML and CSS.)  We want people to be able to implement the spec.  We also want people to participate in its design, if they are taking on the IPR commitment of that participation (that is, that they are effectively contributing their applicable IP).  I’m currently having a discussion with the W3C staff about how to open up the input process while maintaining the value of the Patent Policy; there is also a public discussion going on.

As an aside, I was asked personally to join the WHAT WG over a year ago.  I had a back-and-forth discussion in email with several of the members of the WHAT-WG.  I said something to the effect of “I agree with your goals, all except standardizing error-correction*”.  However, I also said “you have no patent policy, and that makes it impossible for me to join.”  The response was something along the lines of “yeah, we should get one of those.”  There is still no patent policy for the WHAT-WG that I’m aware of – and more importantly, no commitment from those making contributions that they are contributing any relevant IP to those implementing the WHAT-WG specifications, and in my personal opinion that makes the WHAT-WG irresponsible in ensuring that their specifications are actually “free”, even based on the contributions from their own members (inasmuch as anyone is a member).  That also makes me REALLY concerned about just rubber-stamping the WHAT-WG design for HTML5.  Unfortunately, the more open a design process is, the more possible it is to be affected by IPR (because there are more people giving input). 

I don’t think W3C Membership is the right requirement for participation in a WG, because the financial cost is too high for a whole group of people with an important set of perspectives, but the responsibility of the Patent Policy needs to be a bar for giving any significant design input.  I want the new HTML WG to allow contributions from people who are not W3C Members, and in fact in some cases I want to specifically elicit those non-Members to be involved; but I believe it must be a requirement that those contributors explicitly agree to the Patent Policy, as we Members have done.  We can't go back to the 90s.

I’ll avoid a personal diatribe here – but I want you to understand that this is not a world I PERSONALLY enjoy in terms of IP restrictions.  This isn’t my choice.  I’m just trying to spend less time being sued, when all I’m doing is implementing a standard.  (See Eolas case.)

In short, I don’t think we can blindly follow the “everything is totally open” model, because I don’t think we will be able to preserve the Patent Policy and its value.  The W3C Legal staff should probably comment on this, because it’s clearly not my domain; I am not a lawyer, and despite all the lawyers except one that I’ve spent time with being really nice guys (and ladies), I don’t particularly want to increase that time spent either.  I just want the specs to be free to be implemented.  Anyone who isn’t concerned about this simply doesn’t understand the implications well enough, in my opinion.

There’s no perfect scalable system for developing such a central piece of the Web.  We need to have a lot of people involved; yes, conference calls do get difficult with thirty plus people on them.  I think the key is Responsibility – being a member (lower-case-m-member) of the WG is a responsibility, to represent users and developers.  The responsibility of the WG chair is to make sure that the members of the WG are fulfilling that responsibility; that’s the reason behind “in good standing,” though I think it is heavy-handed and therefore misguided.  In reality, in the WGs I’ve been part of the “in good standing rules” don’t necessarily translate to things actually getting done;  members in good standing can blow off their work items consistently and still be in good standing, while people who could be major contributors to the WG can’t be in good standing because they have to miss meetings.  I would like to make sure a more flexible model of involvement is supported.  Maybe that’s your underlying point?

I also agree with you, Daniel, that this WG's future work is deeply influenced by the work already in progress in the WHAT-WG.  I agree that the HTML WG must establish a strong working relationship with the WHAT-WG, and I agree strongly on one thing - ideally, the new HTML WG and the WHAT-WG should become one and only one entity.  We must evolve HTML, and I think I stated privately before that I believe I agree with the goals behind the WHAT-WG (all except enforcing common error-correction – see note in postscript).  Evolution of HTML 4.01 means that we need to always consider how “new content/applications” will work in downlevel systems – in IE6, for example, but also in a limited browser on a set-top-box or mobile device.  Really, that means “how does this degrade to HTML 4.01” and “how does serving this content work for old systems”.  It also means, of course, expanding the capabilities in HTML, particularly in the forms space.  HTML is the center of the web universe.  That’s a powerful responsibility.

I believe this Charter represents not a compromise between the W3C and the WHAT-WG, but (most of) the original goals of the WHAT-WG inside the W3C.  I think that’s a good thing.  The RIGHT thing, even.  I’m not sure what more you’re looking for - the HTML WG shouldn’t be chartered to be the WHAT-WG’s slave, as I’m sure you’d agree that would be a foolish charter.  I’m not sure if you’re looking for a mea culpa from the W3C for the HTML WG over the last few years; if so, I’d recommend just moving on.

On communication, I read “the HTML WG pages should be entirely public” and I’m not really sure what to make of it.  Everyone should be able to read the text of a joke that I make at Ian Hickson’s expense, in good fun during a WG meeting?  Or everyone reads the Microsoft-bashing that inevitably occurs?  I’d prefer to be a little more professional in how the WG presents itself.  Everyone can read the current works-in-progress? Of course.  The issues list and current resolutions are public?  Absolutely.  Anyone has the ability, without paying money or having to travel or spend an hour each week on a long-distance teleconference, to contribute to HTML?  Again yes, but with the caveat that we must solve the IP responsibility dilemma.  We (the responsible members of the WG) listen to the “customers” of the spec?  What a novel concept!  Again, I absolutely believe in this.

I believe as initial chair of the HTML WG, it would be my personal responsibility to be the glue between the W3C and the WHAT-WG, and I would hope in the eventuality of time the WHAT-WG would simply dissolve because it’s no longer necessary.  I don’t question that the W3C is not what it needs to be today to allow that to happen; in fact, that’s why I personally pushed to do something drastic about the HTML WG.  I do want to be clear though, that in my opinion HTML is not in the hands of the WHAT-WG and never has been, despite calling a spec or set of specs “HTML 5”; it belongs to the W3C.  Their stewardship of it over the last few years has been wanting, but that’s the point of the (new) HTML WG.  The right thing for us to do is to evolve HTML 4.01.  Let’s go do it.

Inside, though I’m really hoping that you have a name of someone else to take over the chairship.  :)


PS: On error correction standardization – I disagree with this goal of the WHAT WG because I don’t believe it is responsible for Microsoft to cause any unnecessary compatibility problems (e.g. the problems that would be caused if we changed error-handling behavior).  If the WHAT-WG wants to standardize on EXACTLY what IE6 does for any given error case, no matter how funky it may be, great, but I don’t think that’s what they would consider doing.