Petroglyphs

I’ve updated the code behind the pages to a more modern and flexible system and at the same time amended the design to make pages quicker to load and less fussy in appearance. An additional benefit is that the site is now more accessible to visitors with disabilities.

Michael Quinion, World Wide Words

In other words, WWW is now–almost–a valid XHTML 1.0 Transitional site. (Those darned character entity references in URLs… happens to the best of us.)

Mark Pilgrim and Ian Hickson might not be overjoyed. There’s lots of semantic information in the classes, what appears (to me, at least) to be appropriate use of the meta element, and even Dublin Core metadata; he even uses curly apostrophes. However, there are two things I think Mark and Ian would recommend: respectively, using the cite tag to mark up words and phrases quoted for themselves (particularly instead of <em class="citedform">), and using heading tags (h1, h2, etc.) in place of structures like <p class="heading"> and <p class="sectiontag" style="margin-top:0.5em">.

But you know what? (That’s you as in all y’all, of course.) It’s still a cool site, and I go back again and again. Content is king, and Michael Quinion has some 1250 pages that say he da man.

I was recently demonstrating to someone how easy it is to find information on the web, and then infer certain things about that data–the kind of data mining that anyone can do, given even a basic understanding of search engines. My example, at once effectively and poorly chosen, was the person’s name: effectively because Google Canada returned a single hit, to a page containing a list of people, from which I was able to guess the person’s approximate age, level of education, hometown, and various other bits of information; and poorly because by doing so I think I’ve made this person afraid, of others and of me. Coming up with personal details like that in seconds just to prove a point to someone unfamiliar with the web–let alone the Internet–is more disturbing than I’d considered.

I’m not a privacy zealot by any means. I still fill out forms with accurate demographic information (but bogus e-mail addresses), I use my credit card with abandon, and I’ve only skimmed the EFF‘s website; denying cookies is about as far as I go to conceal my web travels. Seeing the reaction to what I thought of as a trivial little exercise, though, makes me wonder if I’ve been a little too blasé about the whole matter. I’m fine with the information I’ve deliberately released about myself; this weblog may cross the line as to providing material from which certain deductions can be made, but I think I’m willing to stretch that far. It’s the information that I don’t know about–like the Google hit in my demo–that concerns me.

Not that I have anything untoward to hide… I revel in being the most boring person I know!

James Duncan Davidson probably just explained why I wasn’t chosen as a Quicken beta tester this year, after being invited back a couple of times. You see, this year, when they asked what OS I’d be testing on, I responded Windows 2000, under VMware on Red Hat Linux. James’ final words are mine too: So, I’ll be staying at the current version of Quicken Home and Business (which thankfully does run in [VMware]) for the time being and will be migrating away from Intuit’s products soon unless they fix this. Now, I just have to find some software that works as well for my needs.

Not that anyone cares what I think, but I had a lot of the same thoughts as Ian Hickson and Dorothea Salo when I read the first draft XHTML 2.0 specification a few months ago. Eric Meyer is broadly sympathetic with several of its detractors (and I’ve seen even more), but points out that you can use XSLT to bridge the gap between old browsers and new ones. In fact, that’s just what Sjoerd Visscher did five months ago, after showing that current browsers can already deal with a lot of XHTML2 through stylesheets and binding technologies like HTCs or XBL.

Right now XHTML2 is a science experiment, just like publishing a weblog in RSS with embedded HTML (instead of the other way around, published as a separate page). It’s changeable, and changing; it’s been in the works already for more than two years, and will be for at least two more.

Ian points out, correctly, that [the] fact that xHTML2 won’t be widely used before the end of the decade is not a problem. For some reason, the Internet populous has this strange idea that if something isn’t adopted overnight, it must be a failure. Perhaps eventually I’ll think it’s a pain in the ass with no demonstrable benefit (NB: I realize that’s a comment on XHTML 1.1, not XHTML 2). But for now I haven’t made up my mind.