The Boston Globe reports on "a computer algorithm that can examine an anonymous text and determine, with accuracy rates of better than 80 percent, whether the author is male or female." They don't say enough to determine what approach they used, but i have to agree with Deborah Tannen's quote that "it's not too surprising." What did surprise me is how much of the article was devoted to the politics of gender identity: do the genders communicate differently, how are the genders different, is it okay to make these claims, blah blah blah (but then, this is the Globe).

I also wonder how the use of the British National Corpus (which is certainly the largest readily available corpus that i know of) might limit the applicability of their results: it represents the more formal end of the spectrum of written language. For instance, could you get similar results based on the statistics of blog text?

This looks like a PDF of the research in question.

Speaking of hypertextual Scripture, donna pointed me to the Postmodern Bible Commentary (so far it only has Amos). The interface ties together a navigation frame for references and Bible dictionary/glossary headings, a display frame for dictionary content, another display frame for the Scripture text, and a fourth frame for commentary on the text. On top of that, the right fonts will get you Hebrew display. I found the navigation a little confusing because you can't be sure which of the frames is going to change: often it's the frame diagonally across from where you're clicking. But an interesting approach nevertheless.

I'm also a little puzzled about his use of the term postmodern here, but maybe i'm not academic enough.