Things the W3C Should Stop Doing

This is a quick thought as I’m working on something much longer (and hopefully more interesting) to be published in the next week or so. Also, I just want to re-iterate that what’s said here are my own thoughts and not those of Google unless expressly stated otherwise.

I’ve been arguing — mostly over beers — for the last year or so that the W3C needs to find a ways to re-focus on the needs of the constituencies that give it credibility; namely web developers and browser vendors. This tends to fall on frustrated W3C staffer ears as individuals might agree but the organization feels the effects of its failures not as a question about how to re-focus around a credible mission but as a short-run funding shortfall. I’m excited that the new Community Groups have the potential to help fix some of this by allowing a freer flow of ideas and a more representative expressions of demand, but there’s the lingering pay-for-play issue that I think is liable to drag the organization back to the status-quo as soon as the excitement and newness of Community Groups wear off.

Now, let me say clearly up front that I don’t think that standards bodies are charities, nor should they be. People join (in the case of the W3C, “people” == “companies”, an equivalence that works as US legal fiction but never in the real world) because they’re expressing their economic interests in agreement on thorny, market-focused issues. The W3C, as an organ, has grown out of an IETF alternative into a critical mediator in the flow of ideas and technology in the web ecosystem. When it fails, we’re starved of progress because nothing else can easily grow in its place. Bless Hixie and the WHATWG for trying, but the W3C sits on fertile spec and patent rich land. It’s tough growing in the rocky craigs, particularly without a patent policy.

So, in the spirit of reform, not revolution, I have a proposal, complete with Swiftian modesty:

At this year’s TPAC, it should be agreed that the W3C will divest itself of any and all Semantic Web, RDF, XML, Web Services, and Java related activities. SVG can be saved, but only if it re-charters to drop all XML dependencies in the next version.

Where should these specs and member organizations go to get their itches scratched? I’m not sure it matters, although it seems likely that OASIS wouldn’t mind being the go-to place for XML. And there’s always ECMA, or IETF, or ISO, or…you get the idea.

Why do such a drastic thing?

Well, it’s not that drastic, really. The SemWeb folks use the term “web” in a hopeful, forward looking way that nobody else does. Their technology is showing promise inside firewalls but their work has near zero relationship to the actual, messy web we swim in every day. The result is that their narrow constituency isn’t the one that animates whatever legitimacy the W3C might enjoy, and therefore their influence in the W3C is out of all proportion to their organizational and web-wide value. Sir Tim’s belief in their goal is noble, but to use the suffix “web” on it today is simply wishful thinking. And things that don’t have anything to do with the web probably don’t belong at the W3C. Besides, the W3C imprimatur doesn’t help these specs and groups as much as their advocates might think — standards bodies aren’t venues for creating the future after all, only for cleaning it up in response to the messy, explosive process of market-driven evolution. Anyway, being endlessly frustrated that the entire web hasn’t changed to accomodate some new model even though it has the phrase “web” in the name can’t be fun. Technologies need to find their success where they really do fit and distributed extensibility and semantics might be valuable…but it hasn’t been to the web yet. It needs to go.

As for XML, well, it won the enterprise and lost the web. That’s still success, but it’s not the web.

Rinse, repeat for RDF, Web Services, and Java.

What would that leave us with? CSS, DOM & JS APIs, HTML, a11y, i18n, and all the other stuff that has legs out on the public web. More to the point, it would have the beneficial effect of re-focusing the organization around getting HTML5 done, getting DOM APIs done that don’t suck for JavaScript on the altar of IDL “language neutrality”, etc.

Organizationally, it would leave the W3C in a much leaner, more focused place. It’s much easier to build a focused constituency and membership, and the lower cost structure would only help the organization weather hard times. I have high hopes that it can be brutally effective again…but to get there, it needs to focus, and focus is the process of ignoring things. The time has come for the W3C to grab the mantle of the web, shake off its self-doubt, and move to a place where doing good isn’t measured by numbers of specs and activities, but by impact for web developers.

42 Comments

  1. Posted September 20, 2011 at 12:22 pm | Permalink

    There’s a lot of XML on the web.

    Just do a View Source of this page for an example.

    Now there’s JSON too. But XML is still very much with us.

    And btw, there is RSS too. :-)

    Dave

  2. Posted September 20, 2011 at 12:34 pm | Permalink

    XML is not, actually, on the web in any measurable sense. This page doesn’t validate, e.g., and I don’t give a whit about it’s well-formed-ness, nor does any browser (including the one I work on). The web may deal with XML as a subset, but as a grand explanation of the web, it has failed. It’s not integral to the web and shouldn’t be allowed to distort the W3C agenda any further.

    RSS/Atom/etc. could just as easily cite an OASIS standard, FWIW, and suffer no harm.

    Regards

  3. Posted September 20, 2011 at 12:35 pm | Permalink

    I agree with all these, except the idea of Semantic Web.

    The difference of semantic web, is that it’s finally becoming relevant. Albeit in small forms but what is h-card, if not semantic web? Also see Google’s attempt at easier page munging.

    Semantic Web is beginning, and it would be short sighted to kill it off right when it’s hitting on.

  4. Posted September 20, 2011 at 12:44 pm | Permalink

    More specifically, this page is served as text/html. The fact that it has a few talismans of XHTML are meaningless in the face of that fact. This page is HTML, nothing more.

    RSS isn’t XML either. It has a superficial resemblance, but using an XML parser on the web’s RSS feeds will result in a whole lot of disappointment.

  5. Yehuda Katz
    Posted September 20, 2011 at 1:08 pm | Permalink

    It might be more accurate to say that there’s a lot of SGML on the web, and nobody’s clamoring for the W3C to be responsible for SGML.

  6. Posted September 20, 2011 at 1:09 pm | Permalink

    Justin:

    H-card is great! It’s the sort of slang vocabulary to HTML’s official dictionary that can inform the future main-line standard. Schema.org is similar in that sense. But that’s not what most SemWeb folks are after…they’re chasing ontological structures that you can reason about, map divergences with, and use as the basis for structured data mining.

    In any case, cleaving off the current SemWeb efforts wouldn’t touch Microformats in any way, and would in fact strengthen their case since they’re not playing second fiddle to any “big-s Semantic” effort that won’t ever fly in the real, public web.

    Rest easy, what I’m suggesting can only help the pragmatists with which you and and I both agree.

  7. Posted September 20, 2011 at 1:29 pm | Permalink

    I have to agree with Dave on this one, since when is XML not a part of the active web? RSS is a good example.
    Then there is the whole history lesson of XML is better than SGML; we really needed the XML guys to tell us how much HTML sucked. We fixed HTML so it doesn’t suck anymore, but as things evolve it’s good to have the lone voice of dissent reminding us why a certain level strictness is good. The HTML 5 spec (and the CSS3 spec for sure) could use a little discipline, in my opinion.
    The lack of unilateralness is what makes the web great. If the web fails against future closed-platforms because it wasn’t willing to sacrifice its ideals on the altar of progress, then that’s the world’s fault, not the web’s.

  8. Posted September 20, 2011 at 1:38 pm | Permalink

    * What is the java thing?
    * RSS not done at W3C
    * Atom not done at W3C, but IETF.

    The main issue I have with the statement is that it doesn’t explain what are the practical benefits? Put aside reputation. In which ways, it improves things.

    W3C is made of its communities. Now why cutting this or that will improve the Web and/or W3C?

  9. Posted September 20, 2011 at 1:57 pm | Permalink

    karl:

    It’ll improve the web by improving the W3C’s need to keep it’s legitimacy in tact. In a world where it can fall back on pay-for-play revenues by members chasing non-web goals, it dilutes both its effectiveness and long-term brand, creating crises like the current one. Non-web goals also impose implied and explicit coordination costs; working groups feel as though they *should* be coordinating, even if stated goals are light-years apart.

    It’s telling that the non-web specs seek out the W3C and attempt to work “web” into their names. The thing of value that’s being arbitraged is the W3C’s reputation. If you don’t have that squarely in your mind, your analysis of the motivations of the people involved is going to be faulty. The W3C, just as a purely tactical matter then, should be seeking to protect it’s reputation from abuse and devaluation, and that means hewing toward whatever things are part of the web and away from those that aren’t.

    Anyway, the W3C is not made of communities, it’s made of companies who have — best case — been forced to negotiate thanks to market pressure (usually customers screaming for a standard). Communities may have sprung up around these technologies, but their representation at the W3C is weak….not that I should have to explain that to you. It’s pretty plain to see based on how slowly the needs of the larger constituencies are met.

    Regards

  10. Posted September 20, 2011 at 2:47 pm | Permalink

    This is a huge idea.
    W3C has lost focus so many years ago we seem to forget how useful they could be if they want to.

    Some cool things are already happening lately – Device API, Web Performance WG and so on. Let’s hope such focus would result in more of these and less irrelevant, naïve specs.

  11. John Foliot
    Posted September 20, 2011 at 3:21 pm | Permalink

    Alex,

    I am concerned about how and what you perceive the W3C to be, as well as how you share that perception with others.

    “people” == “companies”: That might be how you see the W3C, but as a “member” I contribute as an individual, a person and in fact my W3C contributions are even outside of my current contract in Palo Alto. I proudly operate inside the W3C as an “Invited Expert”, which when translated to English means I show up, pick up a shovel, and dig in to get work done. The majority of the other contributors I work with inside the Accessibility domain are also individuals and not mouth-pieces for their employers. (I do agree with you that the emergent Community Groups will hopefully show many rank-and-file ‘folks’ that working inside the W3C need not be onerous, overly complex or complicated, nor saddled with multiple layers of process)

    “The Web” however is a tad bigger and more pervasive than the latest fancy CSS3 visual tricks, or turning the web browsers into the next-generation, ubiquitous run-time environment for “web-apps”, with APIs for local storage, access to the cell-phone camera and all the other high-profile, (tech)headline capturing wonders we are seeing today. Is that currently a big part of the visible web? You betcha. But the “real deal” of any ice-burg is below the surface, and not just the visible tip.

    Many industries are open to, and in fact welcome, the opportunity to gather in an organization such as the W3C (just ask Comcast: http://twitter.com/#!/johnfoliot/status/115825579399454720), and while they have their own agendas and their own requirements (both financially and technologically), they recognize that coming together in a common forum benefits all participants.

    I am currently writing this while sitting in the Third W3C TV on the Web Workshop, where representatives of Apple, Google, Microsoft, Opera and other “web-focused” companies are coming together with the Television and Entertainment industries (Comcast, Netflix, HBO), the hardware and related device industries (Samsung, Panasonic, Adobe), Regulatory entities (Japan’s National Television, etc.), and others interested in the web as a delivery and communication platform (Internet Archive). The big tent of the W3C and it’s opportunity for diverse industry to come together and enhance the web (and yes, their own bottom lines – let’s honestly admit that Google has a financial stake in the outcome of the web too) is it’s strength, and not something to be whittled away and jettisoned because it seemingly has little to do with browsers per-se.

    The W3C needs industry to support and participate in their work (and their funding), but the W3C is bigger than just 4 or 5 web browsers and a collection of engineers at Google, Apple, Mozilla and a handful of other web-focused companies. It’s more than the twitterati of web designers and developers cranking out Google+, Facebook, and “CanIUse.com”.

    What it appears you are actually proposing is to re-craft the W3C to be a more legitimate WHATWG, and I think any such proposal will fall on numerous deaf ears, simply because the W3C delivers to a more diverse set of constituents than “web browsers”.

    (And as a disclaimer – these are my thoughts, and not official W3C comments)

  12. Posted September 20, 2011 at 3:55 pm | Permalink

    John,

    If you stopped being employed by Stanford you’d no longer be able to participate in the same way, and you know it. Invited experts have no voting rights, either for W3C management or for specs, and I’m really frustrated that you seem to be conflating me pointing out that the W3C is member-driven with me attacking your work. I’m doing nothing of the sort. I want folks like you to have *more* say, not less.

    As to the Web TV thing, let me just say this: the manifest destiny of the web is to be *the* univesal runtime. To the extent that the W3C is facilitating that through HTML and other main-line web technology, that seems completely appropriate. But that has nothing to do with SemWeb faffery, XML architecture expeditions, or Java. So I’m pretty unclear what you think I’m wrong about, unless of course you just think I’m attacking the W3C (I’m not) and you’re defending it (you’re not).

    Regards

    Update: Joe Hewitt is attacking the W3C, and not without basis: http://twitter.com/#!/joehewitt/status/116292923288592384. I’m suggesting a gradual reform and providing a rubric for it. Against both backdrops you haven’t provided an argument for the W3C’s efficacy, only stated that some organization should do what it does. You’ll have to field something better, particularly against Joe and the pitchfork-weilding mob.

  13. Posted September 20, 2011 at 4:41 pm | Permalink

    I say throw ALL of it away and start from scratch using what we’ve learned over the past two decades. Not going to happen, but one can wish.

  14. Posted September 20, 2011 at 4:58 pm | Permalink

    Hi Dave,

    Well, wishing isn’t gonna get us anywhere. See SemWeb. The only question that matters is “what would cause people to move to your new future?”. Is it tons better? Is it easy to adopt? WIll they have no choice? Etc.

    Regards

  15. Posted September 20, 2011 at 6:34 pm | Permalink

    If the semantic web stuff is valuable, it can be picked up somewhere else.

    The W3C should be looking to keep critical web standards activity from going elsewhere, not taking in all the lost puppies that nobody else will feed.

    The W3C is under attack, but not by other possible HTML-and-friends venues or Alex. It’s under attack by its own organizational structure, corporate/pay-for-play basis, and lack of domain focus.

  16. Posted September 20, 2011 at 9:23 pm | Permalink

    I said five years ago (towards the end of http://dbaron.org/log/2006-08#e20060818a ) that one of the big problems with W3C is that it can’t have both of (1) doing whatever standardization work it can get its hands on (rather than focusing on a single coherent market, such as the stuff interchanged on the public Web) and (2) enforcing architectural consistency and non-reinvention between all of its standards. I think it needs to pick at most one of the two and make it clear that it’s not doing the other; I don’t particularly care which. There has been a lot less of (2) lately, but not zero. (Yet to see if it’s possible to do WebVTT in W3C, or if it’s possible to do other widget-like things that aren’t the current W3C widgets.)

    I also think the W3C has an awful lot of staff ( http://www.w3.org/People/ ); staff are momentum — people are friends with the staff, don’t want them to lose their jobs, etc. With a lot of staff, they tend to accumulate processes that require a lot of staff, even if those processes aren’t really necessary. And with staff whose jobs depend on doing work in certain areas, it’s hard to shed areas. I think really focusing the W3C would cause it to lose a bunch of members (have a look through http://www.w3.org/Consortium/Member/List ) and thus have to shed a lot of staff (many of whom are experts in areas that it would be shedding, anyway). And it’s not entirely clear to me how that shrinking would help the pieces that aren’t shed, unless the goal is to give the W3C the moral authority to impose “mandates” on browser makers.

    And if that is the goal, I’m scared of it. The W3C process isn’t remotely suited for that sort of lopsided debate (especially when with lots of small groups who really want just one addition — you end up with a situation just like earmarks in Congress, and for similar reasons; even worse if staff sell that as an advantage of membership). After all, I think the W3C’s process is what got us into the current situation of doing lots of things that aren’t relevant to browsers, and I think it would do so again.

  17. Posted September 20, 2011 at 9:30 pm | Permalink

    Actually, on second thoughts, my “I don’t particularly care which” in the first paragraph is wrong. I do care, as I think my third paragraph makes clear. I’m scared of the W3C being the place that decides what goes in browsers, because I don’t think it’s remotely close to the right setup for being that sort of body.

  18. Posted September 20, 2011 at 10:33 pm | Permalink

    Hey David,

    I’m not sure shedding staff or the members interested in non-web tech is a bad thing. It’s the natural consequence of improving focus, although if the W3C figured out some funding model that didn’t just involve member orgs exclusively (individual memberships? ok, climbing down off my hobby horse), and if it took on more of the needs of end developers, i’m not sure so many staff would need to go — it’d simply grow to fill its (large) manifest destiny.

    The moral authority question is interesting. My interest here is helping get the organization to a place where it feels the pain/needs of both webdevs and vendors such that it helps unblock the process of consolidating gains. To date, the W3C has garnered much of the praise/credit for the modern web, with competition and “proprietary” features getting written out of the history. The result is that the W3C already has the presumed authority, even if it has none of the actual power to effect change in browsers. That was the belligerent arrogance on display with XHTML2, and it doesn’t have any track-record of success in dictating terms to implementers. No amount of focus would change that. Webdevs (and most of the rest of the world) already think the W3C calls the shots, but we know that’s not how it works.

    My mental model for a standards organization doesn’t involve it dictating anything simply because it can’t. It’s incredibly difficult to build a constituency for a feature without shipping implementations, and the W3C can’t create that pre-req without vendors leading first. What it can do is to help take early features from some implementation and bang them into shape and help provide a forum for building consensus. Demands, in this model, come from developers who are responding to the desire to have something they like in one engine in all of them. And webdevs are the ones we should be listening to.

    Perhaps you’re envisioning some Brazil-like dystopia where the W3C wields some massive propaganda arm to scare browser vendors into submission…well, so long as those vendors are at least a large part of keeping the lights on, that dynamic seems a very far distance from where we are — and equally far from where we would be under my proposal.

    Regards

  19. Posted September 21, 2011 at 1:21 am | Permalink

    Alex, you make some interesting points. As someone who thinks quite a lot about how to reform W3C (probably too much), and who knows a good bit about the organization, I like your motivation and appreciate your support, though I think you’re oversimplifying things a bit. I do agree that the W3C could benefit from more focus. I definitely think we need to be more relevant to everyday developers, and we’re working on that (for example, see W3Conf: http://w3.org/conf/ )

    As far as SVG (how gracious of you to grant that it might be saved), you will probably be pleased by the next charter:
    http://www.w3.org/Graphics/SVG/2011/charter.html

    “The SVG WG develops the following technologies:
    A markup syntax, compatible with XML and HTML5 parsing, for describing shapes, text, and painting effects, including gradients, filters, and other graphical effects, through a set of elements and attributes. This includes semantics for navigation, declarative animation, and interactivity. …”

    David, I’m confused by your concern. The only authority W3C has over browser vendors is the fact they choose to collaborate at W3C, and market (or community) pressure. Can you give me a concrete example where it’s gone otherwise?

  20. Posted September 21, 2011 at 3:47 am | Permalink

    Alex,

    it seems we have difficulties to understand each other. What I was asking is not to shut down your ideas or anything, but to express them. Making a broad statement as in “Let’s cut this” doesn’t solve anything. It makes an “entertaining” political statement.

    What I’m trying to understand is how you proceed with it. Contracts of people, contracts with companies, progressive closing of Working Groups.

    The regulation of work at W3C has happened because of the communitieS watchdog be companies, external groups, volunteers, or implementers directly. A good example of that is the patent policy. At the very beginning the policy was basically the same than IETF, a RAND policy. The communities and specifically the open source communities (not member of w3c) have called fools and made a strong call. Invited experts of the opensource community joined as invited expert and finally the patent policy has been modified to be RF. Some big members left because of this. As costly as it was, it was beneficial for the Web ecosystem.

    What I’m trying to say is that there is regulation. Slow regulation indeed, but I tend to think that it is a feature more than something harmful (even if personally I’m frustrated sometimes by the speed of it). The slow pace helps tame down some brutal changes that we would regret later.

    I have followed the W3C for a long time, since its creation. I have been using and doing stuff on the Web since 1991 or 1992. Each time, there was a big crisis (and there have been many) W3C listened and evolved. I’m not saying that there was no mistake. Yes there was but exactly any kind of groups and communities. If people are ready to call reality all the time. This is it.

    So what I’m really interested to know from you (reading the process document, looking at the list of WGs, looking at what is happening in the communities,) is how would you move forward the bits. It’s a lot longer to write down though, but that would certainly help people understand more « how » than « it should. »

    PS: communitieS – There are many communities participating to the W3C. Valuable work from volunteers, opensource developers, member companies, invited experts (those in different areas).

  21. Posted September 21, 2011 at 3:50 am | Permalink

    (and I should not be written when waking up. Sorry for the mistakes ;) )

  22. Posted September 21, 2011 at 3:50 am | Permalink

    (writing doh!)

  23. Posted September 21, 2011 at 11:13 am | Permalink

    Karl: am I understanding correctly that the W3C has contractual requirements to continue certain working groups or areas of activity?

    Which big members left because of RF, and which public-web activities suffered because of it?

    I would find the “slow and thorough is necessary for good specs” argument more compelling if there weren’t such a rich history of blessing specifications which have neither multiple complete conforming implementations nor usefully-thorough test suites.

    (It’s true that it’s hard to write a thorough test suite for a big spec, but I think that’s a point in favour of smaller specs rather than going forward without good test suite coverage.)

  24. John Foliot
    Posted September 21, 2011 at 11:37 am | Permalink

    Alex,

    A minor if not very important clarification: I do not work *for* Stanford, I work under contract *to* Stanford, and specifically, as part of the terms of that contract I do not speak for the University – a point that I have ensured is very clearly understood by those at the W3C with whom I interact with. The association I have with Stanford today certainly benefits me professionally, but I’ve been actively involved in aspects of W3C work long before I landed in Palo Alto, and my current involvement with W3C activities is currently outside the scope of my existing contract at Stanford.

    ***************

    You wrote:
    “When it fails, we’re starved of progress because nothing else can easily grow in its place. Bless Hixie and the WHATWG for trying, but the W3C sits on fertile spec and patent rich land. It’s tough growing in the rocky craigs, particularly without a patent policy.”

    What is unclear in your post is how jettisoning SemanticWeb/XML/RDF/etc. from W3C activities will increase “success”. How has the Semantic Web work frustrated progress in HTML5? Who says that RDF/RDFa has no place on web developer’s agendas (for example, I would be curious to hear how RDF has ‘no place’ at Seevl – http://seevl.net/). Why should the W3C shelve that work? Because the browser vendors are uninterested in supporting that technology?

    What I am really hearing is 2 things, a frustration with process, and a despair caused by a high incidence of “failure” (i.e. slow adoption, or more factually, slow validation, of emergent specifications). What I am also hearing is that you wished that the validation that the W3C brings to a specification could emerge more like how the WHATWG operates, without realizing that therein lies the very problem.

    The legitimacy that the W3C brings to its work is the very fact that it is not a Rubber Stamp organization; that it’s processes and goals is to ensure that emergent web-related technologies are examined from multiple angles and view-points. It strives to ensure that everyone who has a stake in the larger web is heard; that their concerns, dreams, and requirements are acknowledged and addressed. It specifically rejects a single arbitrator of “right or wrong” (i.e. Editor for Life) in favor of real consensus, which means actively listening to all view-points, and not chasing off those who’s ideas of what the web is, can and should be differ from yours. Doing that takes time, and requires a certain amount of process (which also includes negotiation and discussion strategies, arbitration, and yes, sometimes – but hopefully rarely – voting).

    Case in point: at this week’s TV on the Web workshop, the issue of DRM, or rather what Comcast’s Sree Kotay rightfully called secure playback (to avoid theft) surfaced. The television and movie industry want to use the web as well (and why shouldn’t they?), and they want to be actively involved in the standards process (which is why Comcast, HBO and others joined the W3C). But they also want to have *their* concerns listened to, addressed, and hopefully resolved via a consensus process. Not surprisingly then, you can understand their current frustration at the (WHATWG-based) response to the issue – “DRM is evil”. That is not a solution, that is a value judgment, and if the current browser vendors don’t want to deal with the issue, others will step in to fill that void.

    It’s going to take process, time and active listening to resolve this issue (nobody is suggesting this will be easy), but unlike the WHATWG’s knee-jerk reaction, the W3C is taking this issue seriously. Is this ‘frustrating successes’ or slowing progress? That depends on your vantage point: failing to resolve this issue will constitute an HTML5 failure for a large number of stake-holders as well – it just so happens that none of those stake holders are web browser manufacturers (or WHATWG boosters). Hopefully the W3C discussions will produce a sane, technically viable solution that addresses the concern, and that it will do so via a consensus-based process.

    David Baron (Mozilla) writes: “I’m scared of the W3C being the place that decides what goes in browsers” – David, Mozilla will decide what goes into its browser, not the W3C. The W3C will issue its Recommendation, a recommendation that hopefully will insure ubiquitous-ness and an interoperable experience for all users, after which the Mozilla Foundation will make their own business decision on whether or not they choose to follow some, all, or none of that recommendation. Hopefully, via the consensus mechanism that is W3C process, they will have reached the point where they agree with the Recommendation, but if they don’t? Nobody will launch an army against them to “force” them to do anything; and this holds true whether we are discussing DRM/secure-delivery-of-premium-media-content, or Dash/Dart, SPDY, Do Not Track, or any other emergent technical idea.

    W3C process needs to be examined – sure – but it’s not the process that is the real problem. I think instead it is the failure of some to truly understand that the browser vendors don’t own the web, that it belongs to everyone, and to preserve the value of that precious thing everyone deserves a say in how it should work. There may be ways to tighten the process and to shorten time lines, but at the same time it also requires that those who believe that the WHATWG way of fast-tracking specifications is a better way forward, and that the W3C should “learn from that”, need also stop and reflect on active listening and accept that not everyone sees the web the way they do. The real answer is education, on both sides of the aisle, and not stripping way the richness of the diversity of voices at the W3C today. Those diverse voices are not the W3C’s weakness, it’s their strength.

  25. Posted September 21, 2011 at 1:18 pm | Permalink

    @Mike “Karl: am I understanding correctly that the W3C has contractual requirements to continue certain working groups or areas of activity?”

    huh? I wonder how what I said got translated into this. Looking for evil where there isn’t. :) What I said is that companies which paid to join the W3C don’t do it for the beauty of the Web only. They do it because they have business interests. If you follow Alex Russel’s plan: “CUT TECHNO X tomorrow”. Some of these members will be very angry. I mean I would. So the exit strategy is more complicated. It is exactly why I was using the example “Peace in the world” it is a nice feeling but it doesn’t achieve everything. And it’s why I was interesting to know what Alex Russel suggests for reorganizing the organization in a sustainable way, what are the consequences, etc.

    It is not about the technology precisely but about everything which will make this possible. The issues are there.

    BTW, I’ve said that often and I repeat it. One of the main reasons the W3C exists is because there is disagreement in the communities and there is interest of these communities to reach a consensus.

  26. Posted September 21, 2011 at 1:31 pm | Permalink

    @Mike about the slow pace.

    We have to be careful on dropping stuff. I wonder if CSS would have survived with this policy on quick results.

    About history on blessing specification. We have to be careful and reassess the specifications in their historical context. It didn’t happen in void. For example, when I joined W3C as staff in 2000 (I left in 2008), I started the QA activity. One of the goals was to advocate and evangelize for test suites. It improves a lot. Not perfect yet. Funnily enough one of the best specifications with regards to test driven development is OWL (Semantic Web), the spec was developed by creating tests and implementing them. We worked hard to convince working group to create test suites and it became part of the culture. It is not perfect yet. It takes time to change people and practices. People participating to W3C are doing it paid by their company. W3C is a place for working not for enforcement. :) It’s yet another feature. Frustrating sometimes :), but it has good consequences on the motives for working and doing things.

    What I am aiming to is that it is never that simple. It doesn’t mean that we can’t move forward. We (all of us loving the Web) can, and we have done, and we will do again. Broad statement doesn’t achieve anything. There are so broad that they become meaningless.

  27. Posted September 21, 2011 at 3:19 pm | Permalink

    Karl:

    CSS was shipping in browsers. Market-based progress is the evolution that the W3C tracks when it’s actually functional (see my more recent post on insurance). Not sure why you imagine that CSS would have been the other way around.

    And I disagree about building consensus on intent and direction. There’s shockingly little of it at the W3C AFAICT. Not sure how any organization is supposed to go forward when it can’t agree what direction that might be in.

    Regards

  28. Posted September 21, 2011 at 3:30 pm | Permalink

    John: Sorry to be blunt, but your comment makes me suspect you fundamentally misunderstand why web developers imbue the W3C with authority and value. Here’s a quick thought experiment: Assume HTML, CSS, and DOM up-and-left the W3C, would we still even be talking about it?

    All the discussion of consensus-based process, etc., is a side-light to the question of focus. You’ve said the W3C is valuable because it does more than what browsers ship. I challenge this directly. I think it’s not.

  29. John Foliot
    Posted September 22, 2011 at 2:43 pm | Permalink

    Alex,
    With due respect, but to be equally candid and frank, I believe you have a very myopic view of the multi-billion dollar industry that is the “web”.

    There are numerous players on the web today that do not fall into the category of “web developer” (the Mountain Dew and Cheetos fueled CS major sitting in their Herman Miller chair at a desk made from 2 saw horses and a panel door from Home Depot, banging out cutting edge code on their MBPs and 27” Thunderbolt monitors). Entities and organization such as the German Research Center for Artificial Intelligence (DFKI), the U.S. Environmental Protection Agency, Beijing University of Technology and more – all members of the W3C – their involvement and financial commitment to the W3C also imbues legitimacy to the work and Standards that emerge from the W3C. Why? Because there is a process and commitment to openness and equity that their respective task-masters and clients expect and demand.

    Let’s examine a question you posed. You said, “Assume HTML, CSS, and DOM up-and-left the W3C…” – well Alex, that is more than just a thought experiment, it sort-of already happened. “HTML up and left” and resumed (continued?) work at the WHATWG in 2004. Contrary to what many people might think, I actually believe that the work that happened there, the motivation and history, as well as the outcome has been generally a very positive step forward for the web. Yet the financial backers behind the creation of the WHATWG also understood, and for the most part embraced the idea, that bringing HTML back into the W3C tent was an important step, for a few simple reasons: 1) getting Microsoft (and Internet Explorer) to work inter-operably with the new advancements of HTML5 meant having to go where Microsoft wanted to do such work (complete with Patent protections, etc.), 2) other invested parties in the end-to-end tool-chain (think IBM or Adobe) also expected that work on “the web” happened in a predictable, stable and accountable environment, where their issues and concerns would be listened to equitably, 3) non-web entities such as governments and academia have mandates that specifically disallow ‘rule-making’ using the single-arbitrator model that is WHATWG, and instead expect and in some instances demand that a more equitable process be used. Nobody *forced* Apple, Opera, Google and Mozilla to roll the WHATWG work back into the W3C, they chose to do so for purely selfish business reasons: failing to do so would cost them money.

    That re-integration hasn’t happened as painlessly as some would wish, (due in part to the stubborn obstinacy of The Editor, who believes he knows best about everything, and will argue even for minute editorial details of zero technical consequence just to be right, so as to somehow feel that he remains in control) but by and large, the W3C process of advancing HTML5 forward is being well steered by the 3 editors (who work for Apple, IBM and Microsoft between the 3). Is it taking time? Yes. Is that a bad thing? Again, it depends on your point of view. But I feel pretty confident that those who are supportive of the W3C process of consensus versus single arbitrator progress understand and accept that ensuring everyone gets to weigh in does take time. (Managing that time-line process is another discussion all-together…)

    You said: “…your comment makes me suspect you fundamentally misunderstand why web developers imbue the W3C with authority and value.”

    It is undoubtedly true that for some web developers, the speed and agility of how the WHATWG works and produces specs directly related to the browser environments feels like the WHATWG way is the right way, neigh the only way, that the web works. But not all web developers work at that level, at that speed, at that rapid rate of acceptance. I but only need to chat with my neighbor, a web developer who works at a financial services organization here in the Bay Area, to confirm that assertion: he’s extremely interested in doing some “HTML5 stuff”, but his day-to-day job output requires that he still work in XHTML 1.1. He, and his boss, are waiting to be sure that HTML5 is “done” before they start using HTML5 in their production output, because *their* clients have an expectation that the code they deliver, the service they provide, meets the “standard”, which to him (and his boss) means it has been vetted by the W3C and works in ALL browsers. No Alex, I understand very well how mainstream web developers work, and how they understand web standards. They don’t all get to work at the sexy “Google jobs”, where all the cool cutting edge stuff is going down; instead they sit in their industrial-park cubicles, their well thumbed copies of Zeldman’s Developing with Web Standards, Cederholm’s Bulletproof Web Design, and Resig’s Learning jQuery close at hand, creating web content with CS4 (because their boss *still* hasn’t approved the upgrade to CS5.5).

    You conclude: “You’ve said the W3C is valuable because it does more than what browsers ship. I challenge this directly. I think it’s not.”

    I respect your right to hold that opinion, and this *is* your blog, so you can write what you want. I’ve not really seen any proof to back that assertion, and while I’ve heard that sentiment before from others, I disagree (as apparently do others). You want to jettison RDF/RDFa because (…well, I still don’t understand the reasoning), but apparently Facebook and Drupal are both interested in RDFa, and RDFa has nothing to do with how or what browsers do, yet is important to Facebook and Drupal, and so to “the web” as well, in its own way. Personally I am thankful and appreciative of the work that Manu has steered within the RDFa WG at the W3C. Equally, I appreciate what the WHATWG has done for “browsers”, and am thankful that their work is being brought back into the big tent, but I also think that you need to look at the bigger picture. The W3C is way bigger than what browsers do – sure what browsers do is important at the W3C today – but the W3C is bigger than that, and the supporters of the W3C like it like that.

    Finally Alex, thanks for the opportunity to dialogue. Perhaps we’ll meet over a beer soon as well, and we can continue the discussion. Cheers!

  30. Posted September 22, 2011 at 4:31 pm | Permalink

    John:

    This is really frustrating. On the one hand you’re making my argument for me — your example about devs who are on the trailing edge perfectly matches my model of standards as insurance: http://infrequently.org/2011/09/standards-are-insurance/

    On the other hand, you seem to be arguing that the W3C should continue to underwrite insurance on things that aren’t valuable to the whole web yet and may never be (RDF, XML, Java, etc.). It’s unclear why you appear to think this. Is it good for the W3C to have its legitimacy spent this way?

    I argue that it dilutes the brand and makes it harder to reason about what the organization stands for. Is it good for those non-web communities that they are formed at the W3C and not in somewhere else with similar rules? Is there some specific need for them to happen at the W3C?

    Your argument seems to hinge on the idea that the set of activities currently convened at the W3C are the ones the supporters want. OF COURSE that’s true, particularly if you take “supporters” to mean “member organizations”. That’s got very little to do with what would be better for the W3C or for the developers who look to the specs it needs to continue to turn out in order to retain whatever legitimacy it has left (hint: not as much as many think).

    Anyway, you’ll note that I’m not arguing that we should move everything to the WHATWG. That wasn’t in my thought experiment…weird that you read it in. What you haven’t argued yet, and what I would love to hear from you, is what value you think SemWeb/XML activities are bringing that they couldn’t do in a different org and/or why a streamlined W3C would hurt users of W3C specs.

    Lastly, and I can’t stress this enough, the “standards as insurance” model might leave some room for a market-maker to help create a new product, but not much, and it’s the sort of thing that needs a sunset. When products fail in the market, they should be shut down or spun out. If the W3C defines its market as serving the users and developers of the public web, it’s pretty clear to me where things might be hewn to a first approximation.

    Do you think the W3C should be defining it’s market in some other way? And if so, why? And who is that good for?

    Also, yes to beer = )

  31. Posted September 23, 2011 at 6:56 am | Permalink

    There needs to be some rules so that unruly M$ isn’t the only web-page provider that works.

    W3C does a good job. If you don’t like the rules, get them changed. Nevertheless, we need rules and we will use the new rules after you change them.

    This web-page has two errors…..

    Validation Output: 2 Errors

    Error Line 571, Column 16: the name and VI delimiter can be omitted from an attribute specification only if SHORTTAG YES is specified

    “VI delimiter” is a technical term for the equal sign. This error message means that the name of an attribute and the equal sign cannot be omitted when specifying an attribute. A common cause for this error message is the use of “Attribute Minimization” in document types where it is not allowed, in XHTML for instance.

    How to fix: For attributes such as compact, checked or selected, do not write e.g <option selected … but rather <option selected="selected" …
    Error Line 571, Column 16: required attribute "type" not specified

    The attribute given above is required for an element that you’ve used, but you have omitted it. For instance, in most HTML and XHTML document types the “type” attribute is required on the “script” element and the “alt” attribute is required for the “img” element.

    Typical values for type are type=”text/css” for and type=”text/javascript” for .

  32. Posted September 23, 2011 at 8:44 am | Permalink

    John: I can’t tell you how very, very much I don’t care about those “errors”. Did they prevent you from viewing the site? Did anything about it *not* work? Ok, then lets do what browsers do for the benefit of everyone: pick up and move on, not worrying about things that have no relationship to “better” or “worse” on a “good for users” or “good for developers” scale.

  33. Eddy Vluggen
    Posted September 23, 2011 at 10:13 am | Permalink

    The W3C is looking for something to justify it’s existence. It should simply “stop doing”.

  34. John Foliot
    Posted September 23, 2011 at 1:37 pm | Permalink

    Hey Alex,

    (T’wasn’t me bustin’ yer chops on validation errors BTW)

    A few comments for you.

    Why wouldn’t an organization or business entity want to get their “web” insurance(s) from the same location? What value is it to them to see that split apart? In the physical world all the insurance companies I see are bundling their “Life/Home/Auto” plans because it makes good business sense for both those companies, as well as their clients. Knowing that the W3C is standardizing both HTML and RDF/RDFa (or XML) gives those companies interested in those technologies a level of assurance of interoperability to their business needs. You suggest that XML and RDF are “write-offs” w.r.t. the web, yet I have a slew of engineers at our libraries system who might disagree with you: they get back OCRed text from the Google Book Project in ‘uncorrected’ XML format, where they then add additional RDF data so that they can index and then deliver (via the web) content to the students. Having inter-operable standards from one organization that ties those technologies together is a benefit to them.

    I don’t see involvement in a diverse set of related activities and Standards creation as any more incongruous or diluting than the reCAPTCHA project being part of Google (if one is to assume that Google’s core mission is indexing the world’s content – i.e. Search). For that matter, why build a browser if your core mission is search? Perhaps because seemingly unrelated efforts can sometimes stitch together in unusual ways? Because being involved in diverse but related activities is actually beneficial to a bigger picture view of the eco-system?

    If I am to understand you, you are stating a preference for narrow and deep, rather than long and broad, but is that what “the web” wants? A quick check of the emergent Community Groups at the W3C would suggest otherwise: http://www.w3.org/community/groups/, and so it would seem that even amongst the communities there is not a general agreement with your strategy. It might make “web developers” feel more listened to or focused upon to be bigger fish in a smaller pond, but again I would suggest that this is something of a limited vantage point.

    Perhaps instead of splitting off other activities, what you are desiring is that more time/money/effort be directed towards certain web-specific activities such as HTML, CSS and JavaScript? That would be interesting and worthy goal, one “the community” would need to indicate a preference and desire for that, which would also include the requirement for some financial backing: maybe what the W3C needs to do is to create a new level of financial support for individual members: perhaps $50.00/year for each individual (or something like that)? Re-examining how to interact with the larger community of individual developers who have fueled much of the web’s growth is, I think, an important step.

    Alex, I think you and I both agree that “something” needs to be done, and I suspect that our desired outcomes are not that dis-similar either. Strategies on how to get there however, we are far apart on. While you and I could go back and forth a lot more on this, I think you would also agree that getting the dialogue out there, and getting more voices to the table, will help everyone better understand the wide range of desires and goals, and then, through discussion we can all reach a consensus on what we all really need and want.

    Funny enough, that’s how the W3C has always worked. (OK, forgive them the XHTML2 thing already… )

  35. Pawel Dyda
    Posted September 24, 2011 at 1:10 pm | Permalink

    Hey,

    Actually, I must totally disagree with you. I think it is exactly the other way round: W3C should care less for browser vendors and more about *end users*. Currently, it is you – The Browser Vendors, who vote to drop really important features (for end users) out of HTML5 specification. The one I have in mind is *proper* support for Internationalization.

    It seems that nobody from Google went to Multilingual Web Workshop in Limerick (finished on Thursday). Or maybe I just didn’t notice? I wish you were there, probably it would give you more insight on why the heck W3C engages in all these initiatives.

    The real reason is: web is for users (the people), not for browser vendors. By voting us (standard users) out you only force us to create workarounds, which is to say the least costly. I understand that Google does not care much about things like i18n (many bugs in your apps are quite a proof ) but others do, and that means a lot of energy spent on fixing something that should really be done in the proper way in the HTML standard.

    Of course my lengthy tirade regards didn’t touch things like semantic web, etc. These are part of Web 3.0 concept (I believe Google actually invest in this area, isn’t it) and I can easily understand why World Wide *Web* Consortium expresses its interest in such areas.

    OK, I am not even the member, so maybe I should just shut up :>

  36. Posted September 26, 2011 at 6:23 am | Permalink

    Why do you care?

    Seriously, why do you care?

    If you don’t like XML, don’t work with XML. If you don’t care for RDF, don’t work with RDF.

    What you’re saying is that if you and Google and the other browser makers can’t find value in something at the W3C, it should be removed from the W3C. That the W3C should focus exclusively on what you five companies deem of “value”.

    Is this shallow world view what you really want of the web?

  37. Posted September 26, 2011 at 11:14 am | Permalink

    Shelly:

    I get that you’re not responding to the substantive point I buried in the longer post and proposal — my apologies for not calling the logic out more clearly. The core of the argument is this: I assert that W3C currently under-serves HTML, CSS, and DOM and in ways large and small. First, cultural affinity to and explicit coordination functions (see: TAG) with the other efforts at the W3C impose coordination and mis-direction costs that cause participants to slow down the necessarily deliberative process, nearly to the breaking point. To the extent that these efforts don’t have any natural affinity, I’m arguing that it’s easier to help keep standards from “leaking in” to others when there’s an organizational boundary. We see the ill effects of this already with DOM/JS, but I’m suggesting that it can be used to improve the rates of progress of specs for which co-location is as much (or more) bug than feature.

    Your response seems to suggest those costs don’t exist or that they don’t matter. Is that your position?

  38. Posted September 26, 2011 at 11:59 am | Permalink

    Alex, you’re asserting that the W3C underserves CSS, the DOM, and HTML? Frankly, the W3C has become the dog to the HTML5 tail.

    The primarily problem at the W3C right at this moment is all energy keeps getting diverted in an attempt to contain a small handful of people from the browser companies who continually do everything in their power to undermine the organization.

    Even your post: you want specification work that supposedly isn’t useful or of interest to Google split off “elsewhere”. This is another drag on the attention of all people involved or interests in specification development at the W3C. Yet another time we have to stop, sort through the concerns from the FUD, and spend time determining whether this is a problem for all people on the web, or just a couple of companies that develop browsers.

    Of all the problems facing the W3C now, the fact that SVG is XML, and that RDF doesn’t fit into the great schema.org plan, doesn’t rate that high in my book.

  39. Posted September 26, 2011 at 4:16 pm | Permalink

    Hi Shelly,

    I’m trying to understand this: what’s the W3C’s value without HTML? Given that you think HTML5 is wagging the dog inappropriately, do you support the W3C divesting itself of HTML, CSS, DOM, etc.?

    From my perspective, I can’t say I particularly care which way the divestiture goes, so if you think those activities are harming the W3C, do you support them taking their patent policies, and perpetually sub-licensing the specs and IP rights to a different standards body with similar rules?

  40. Posted September 26, 2011 at 4:39 pm | Permalink

    Why would the W3C “divest” itself of any of the specs?

    Here’s the reality: if you’re interested in working on HTML5, you join the HTML WG. The same is said of the other efforts.

    It’s not like there’s finite resources involved. The W3C isn’t paying salaries for the people working in the groups.

    “Divesting” itself of specifications, with all the problems this will cause, won’t make things “better” at the W3C.

    What will make things better at the W3C is not having to fight against companies undermining its role.

    I don’t see the problems you see. I guess I’ll have to join with John: best of luck to you in getting support for your ideas.

  41. Posted September 26, 2011 at 4:47 pm | Permalink

    I’m confused on a couple of points:

    • Does the W3C pay staff to liase and help coordinate with working groups or not?
    • Why does “making this better at the W3C” matter? I thought the goal was to make things better for web developers?
    • Why is the W3C’s role important as more than a middle-man role in writing insurance? See: http://infrequently.org/2011/09/standards-are-insurance/
    • Do you think the W3C’s role *isn’t* writing insurance? And if not, what is it?
  42. Posted November 27, 2011 at 8:29 pm | Permalink

    If W3C would split up in several sub-groups, it would only be harder for web developers to bitch at the correct party that is messing things up. ;)

    But jokes aside…

    Truth is: It’s not the W3C failing here, it’s the browser vendors.

    People will simply start ignoring proposed standards if “they don’t work”… meaning: “when browser vendors fail to support them correctly”. I don’t have to reach back far into my memory to remember the XHTML 2.0 fiasco and what gave reason to create what is now called HTML5 (you know, that smart HTML2 excuse of “it’s new but looks as messy and familiar like it’s 1996″).

    There was and is no reason to abandon any standard or technology… unless you are a browser vendor and want to trim down your investments into human resources.

    Dropping XML and other (already proposed) standards or even splitting up the W3C to focus on smaller pieces of the cake doesn’t solve anything… browser vendors hiring more people to make their browsers support the proposed standards will – but then they would have to invest hard cash. Killing standards by not supporting them is much cheaper. And that’s where capitalism bites it’s own tail while trying to suck the life out of the internet.

    When you write “SVG can be saved, but only if it re-charters to drop all XML dependencies in the next version.” I have to work hard not to laugh, as it shows the lack of understanding what SVG actually is and where it came from. But hey, I understand the problems of a browser vendor working with SVG in a sluggish HTML5 document. Tsss… it’s as if browser vendors said: let’s kick out XHTML and forgot that it’s more than just a webpage.

    Looking into the future, I could as well say that plain text files saved in little TXT files is probably the most future-proof technology, so why bother having a W3C at all. When browser vendors finally managed to kill all standards, we could go back to setting up our own BBS systems again and abuse the internet (which is nothing more than a network) as a phone-line like back in 1993. But wait… where would Google be by then? Oh right, they would have killed their own internet and your efforts for the Crome Browser would void themselves.

    My two cents: it’s easy to press the delete button on stuff, but “easy” isn’t always the best way to do things.

    And before I go, I would like to remind you of your own writings, which contradict this rather “one-sided” article: [quot] “My goal is to make the web suck less and to the extent that I can keep politics and economics from creeping in, that’s what this blog is about.” [/quot] Question: why am I not the only one seeing you are letting too much “politics” and “economics” creeping into this article? Maybe it’s time to remember the roots… or it’s time to change that “about” text. Contradictions never manage to convince.

4 Trackbacks

  1. [...] Things the W3C Should Stop Doing (Alex Russell) [...]

  2. By What A Breakup Would Look Like | Infrequently Noted on September 21, 2011 at 11:12 am

    [...] Infrequently Noted Skip to content About Me « Things the W3C Should Stop Doing [...]

  3. [...] Russell (of Google, a browser vendor) wrote a considered blog post Things the W3C Should Stop Doing in which he suggested the W3C needs to find a ways to re-focus on the needs of the constituencies [...]

  4. By My 2 ¢ on the recent W3C debacle | Steal My Code on September 27, 2011 at 3:25 am

    [...] relevant posts in this saga are: Alex Russell’s Things the W3C Should Stop Doing Joe Hewitt’s Web Technologies Need an Owner John Allsopp’s The web is a different [...]