mahiwaga

In the tradition of the paper “Go To Statement Considered Harmful” by Edsger W Dijkstra (written back when BASIC was king, and you had to use line numbers), Norman Walsh declares that “Embedded Markup Considered Harmful.”

While I am not completely up on all the W3C recommendations and the DOM and the RSS controversies, I do think that he has a point. Obviously, I haven’t screwed around with XML enough to understand why you would want to break the spec and allow non-parseable fragments to float around. The only example given—that of preserving HTML 4 idiosyncrasies—may perhaps be a strawman. After all, no reasonable modern browser will choke on XHTML (just put a space after the element name and before the closing slash-angle, like so: <br /> instead of <br> and while it is a pain in the ass, you can easily use something like HTMLTidy to clean up your legacy cruft.

I mean, I suppose the difficulty comes in when people whose feeds you want to aggregate refuse to modernize. (But seriously, WTF? You have an RSS feed but you refuse to use XHTML?) I haven’t played with RSS yet, so I suppose I really can’t say anything, but still.

Why break the spec? (Why put random GOTOs in your BASIC program creating spaghettit code?) Someone give me a clue.

Supposedly, this script can tell whether the writer is male or female. Details of the algorithm can be found on nature.com.

The results, however, are not encouraging, given that it is right only 50% of the time. I tried plugging in some of my old blog entries into the script, and it was right about “kid a” by radiohead but wrong about But Now What?. I also plugged in one of R’s blog entries (sorry! I couldn’t resist! I hope you don’t mind!) and the script got it right.

Interesting thought, though. If the script actually worked, you could use it to guess the gender of an anonymous blogger. Heh.

Ken (aka 1moredork online) experiments with the Gender Genie. (Link from littleyellowdifferent.com) And I was highly amused by Ken’s comment.