« May 5, 2007 | Main | May 8, 2007 »

Sunday, May 6, 2007

Google/Blogger: XHTML Standards: Does Anybody Care?

Let me say at the outset that compliance with the very demanding XHTML 1.0 “Strict” document type definition can involve a lot of fussy work, as “presentation” attributes, including the ability to open a link in a different window, must be accomplished via CSS or JavaScript instead of the time-tested HTML attributes used in so many Web pages. Further, I do not set myself up as an exemplar of full compliance with these standards. While I try to validate every static Web page on my site against the relevant standards, it's very easy to overlook some circumstance in which variant content on a dynamically generated page may run afoul of the standard, and absent browsers which rigorously validate the pages they display, it is very difficult to validate all possible cases for dynamically generated pages.

That said, nobody forces you to declare your pages XHTML Strict—browsers continue to render legacy HTML and earlier, less demanding, standards just fine. So, you'd think that somebody who went to the trouble to assert strict compatibility with the XHTML standard would make an effort to actually attain that happy state.

In the case of Blogger, which is now owned by Google, you could certainly think that, but you'd be dead wrong. I recently set up a Hacker's Diet Online Development Log journal on Blogger so participants in the beta test program of that Web application will have real-time access to changes in the program and the ability to track them with an RSS feed.

Well, call me a nerd, but I always look at the HTML generated by something I use to publish content, and I was impressed that Blogger declares its pages to be: 

    <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" 
        "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
So, naturally, I immediately dispatched the page off to the W3C Markup Validation Service—“Trust, but verify”…or maybe, in this case, verify then distrust! My humble blog came back with 131 validation errors, many of which were pure bonehead blunders such as not encoding ampersands in URL query strings as XHTML text entities. To simplify the test, I created a throw-away blog with absolutely minimal content, XHTML Standards: Does Anybody Care? and subjected it to the scrutiny of the W3C Validator. The verdict?…a mere 73 errors on a page with a single posting.

Nobody compelled Google/Blogger to produce XHTML 1.0 Strict pages—they voluntarily chose to assert compatibility with that standard, while other, less demanding alternatives were available. But whatever standard you choose, you should be willing to be held to it, and in this case the blogging platform used by tens of millions of people falls flat on its face. Personally, I would be stone ashamed to ship something in this state. That Google, with what amounts to unlimited funds in our talent-constrained industry, plus the putatively smartest and certainly most smug technical staff, contents themselves with this is perhaps an indication that before expounding on issues of good and evil, one should first address the more mundane matter of competence.

Posted at 22:25 Permalink