waffle

Guy Kawasaki once said that there are two kinds of products: the kind people pervert for other purposes and the kind no one uses.

HTML is a very successful product in this manner. It was started as a standard interchange format for research facilities. Then it evolved through pages with cat pictures, pages of coffee pots, pages of magenta-on-lime Comic Sans prose with animated Calvin and Hobbes horizontal rules, pages containing the words “I KISS YOU”, pages on which you could order stuff, pages on which you could do your banking, pages on which your cursor is followed by an analogue clock or a swarm of bees, pages where parts continuously reload, pages totally done in Flash and finally pages on which people weather their opinions on videos.

For the first few years, things were invented in order to save these rapid shifts of purpose. During the past decade, things just froze up. Someone at the W3C decided “screw this, XML is the future” and created XHTML, HTML4 reformulated in XML. Then work began at XHTML 2.0 where people on mailing lists plan for a “backwards incompatible” (code for “the perfect markup language”) version. All they’ve accomplished is throwing out a lot of baby and very little actual bathwater, make all elements links and outsource the form model to XForms.

XHTML 2.0 may be semantically more correct, but at what cost exactly? When do the pontiffs on the mailing list gather that the time has come for everyone to jump from HTML (or XHTML 1.0/1.1) to XHTML 2.0 in one fell swoop?

Meanwhile, a Powerpuff Girls fan has worked with a) people that actually use HTML on their own time and for clients in the commercial space; and b) browser vendors that are actually active and actually care about standards; for the past few years, and produced the most complete HTML specification ever, including a recommended specification of parsing and tokenization based on real-world use today. It also adds about three times as many (and five times as good) new things as any HTML or XHTML version since HTML 4. And it’s backwards compatible.

Four years ago – in, tellingly, a response to a XHTML 2.0 working draft dropping some of the tags that were both semantically and practically useful and that came from HTML 4 – Jeffrey Zeldman said that “the W3C should shuck off its modest, academic shell and begin selling the community on the benefits of and thinking behind the spec it is now calling XHTML 2″. The W3C has not done so. At all. And now we know: there never were any actual practical benefits of adopting XHTML 2. There were only renaming (and removal) of tags for semantics’ own sake.

How do we move forward? By codifying the recent ad-hoc extensions such as the canvas tag and the XMLHTTPRequest object, by listening closely to what’s needed to solve problems that are still out there and by making it easier for browsers to work the same way where it matters. The W3C has shown no desire at all to do this, just to resume the non-XML-based HTML track in their standards railway, headed by the same kind of committee-think that got us into this mess in the first place.

The WHATWG is moving HTML forward. The W3C is academic and increasingly irrelevant. Lack of validator or browser support notwithstanding, HTML5 is the future, and I’m going to throw in with the WHATWG.

I’ve had it with the academic nonsense. I’ve had it with formats designed to halt rendering on first error. I’ve had it with formats that aren’t clearly specified. I’ve had it with formats designed by closed committees, chaired by companies that are either completely irrelevant or not actually willing to support the standard in the end, and populated with people more concerned with semantic clarity than fitness for actual real-world use. And while out of respect for my clients and the stable platform they require I can’t do the same for the people whose web sites I otherwise maintain, I am going to do it here:

By this time next week, Waffle will be published in HTML5 and no other XHTML/HTML dialect. Goodbye, XHTML. It’s been nice knowing you.