My (current) opinions on HTML5

There's been a lot of screaming and ranting lately over the current state of HTML5—what is it, what's in it, who controls it, who will implement it, and so on. There's no shortage of good essays, and here are some places to start if you want to learn more.

However, if you want to be taken seriously as a web pundit it seems to be a requirement that you put in your own two cents, so here's mine.

If you've seen me at any web-related conferences in the last few years, you've probably seen me wearing a button like the one on the left (if you want your own, you can get them here).

I wear it because I think the state of "web standards" has been horrendously screwed up for several years; not just this year or last year, but for a good long time. I'd date my frustration as beginning with the W3C's announcement of XHTML 2, which was, imo, a solution in search of a problem. The W3C hasn't had their act together since then (did they ever?), and I don't expect that to change any time soon.

Initially, the WHATWG and their rough HTML5 spec looked promising. With representation from Apple, Mozilla, and Opera, I thought they could really accomplish something positive—if Microsoft could/would join the club. When I asked that question a few years ago, I got a very straightforward answer from MS: the WHATWG doesn't have a patent policy, and so long as that's the case, MS can't join. If you don't have the vendor with the #1 market share at the table, you're just wasting your time.

I became more hopeful when it looked like the W3C and the WHATWG might be able to work together on HTML5. But so much for that—the W3C architecture astronauts couldn't leave HTML5 untouched, and apparently the WHATWG may have drunk a little of the AA Kool-aid as well.

And now? The WHATWG have picked up their ball and gone home. They've scratched the number off of "HTML5," so it's now just HTML. No more version numbers—we're now in the "versionless" future, what ever that means. For web designers and developers (Remember them? Funny, the W3C and WHATWG don't) it means they can never again hold browser vendors to a promise to support x version of a spec. For tech writers, it means that they can never again say their book, article, tutorial, or video covers up through y version of HTML.

What's next? My crystal ball is fuzzy, but my guesses:

• Browser vendors are going to start doing whatever the hell they want. Why shouldn't they? With no way to hold them accountable, and no way to measure their compliance with a standard, it'll be just like the bad old days.
• Side note: when is the Mozilla Foundation going to realize that they're dead? Most of their income comes from Google searches by Firefox users—and now that Google's got their own browser, why keep paying? If I was a Google stockholder, I'd be interested in the answer to that question. Even non-profits need business models, and Mozilla doesn't have one.
• Book publishers are going to freak the hell out. Not right away, but when they get it, it's gonna be ugly.
• Your average web designer who just wants to know how to do something is going to go insane. Looking at the specs won't be useful, as browser makers can ignore them. Trying to find how-to's online will be a PITA, as they'll only be useful for the one moment in time that they were written.
• Life will be very good for Adobe—so long as they can update Dreamweaver often enough to keep up with the browser maker's zigging and zagging. Adobe BrowserLab will be huge.
• In the long run, a new group will arise: a web equivalent of Underwriters Laboratories. A vendor wants the seal of approval for their browser? Gotta meet this list of requirements. If you don't, fine—but after enough pain, governments and corporations will start requiring it, so it's better to get with the program early.
• Thinking about starting a group like this? My advice: don't allow browser vendors to join. Even better: don't allow W3C or WHATWG reps to join. Keep the decision making about the requirements list completely separate from the spec designers and the browser makers, or the tail will end up wagging the dog.
• That requirements list may or may not have anything to do with some version of HTML—and that's okay. Much of the interesting stuff on the web in the last decade has come from vendors, not standards writers. If it shipped, and cool stuff could be done with it, web builders pressured other vendors to implement it too. That method has been working, and that's okay.

And at the end, it all comes back to that slogan on the button—all that really matters is what actually ships and what's actually usable. There are some lovely theories in all those XHTML and HTML specs that will never see the light of day, and that makes them completely irrelevant to 99% of us.

Let's stick with figuring out how to make the stuff that does work, work for us.