So, spurred on by the revelations that ex-Hitler Speech connoisseur Tom Costello from my old college TCD has got himself a new search engine that apparently returns more targetted gay porn results than Google, I did some googling searching to see what was going on with that web of people from the early 90s, with whom I apparently share 2 or 3 degrees of separation according to stalker vanity sites such as LinkedIn, FaceBook, etc. Sarah Carey is, of course, doing some PR for Costello’s lamentably spelled Cuil, and churning out the usual Naif-in-Hy-Brasil copy that a certain class of Irish readers favour as a framing device for the rather more complex entity known as “America”. But what really tickled me was clicking on a link, and then another, and then another, to find a rather late review of TrinCon 400, a science fiction drinking contest me and some friends did in the early 90s. The review nails it – we had no idea what was going on, but we had bags of cash to spend because of TCD’s 400th anniversary, and we knew what we wanted (mainly no nerds with pointy ears). So we got most of it together, despite my late attempt to sabotage everything by pissing off Trinity College’s establishment after printing up a flyer calling them a bunch of “old knobs” just before the gig (Harry Harrison managed to piss them off with more style). Anyway, it’s nice to know someone else enjoyed it.
Anyway, regarding Cuil, it’s impressive that they grabbed a small VC investment to launch a Google competitor without bothering to incestuously link to any Web 2, FOSS, or social buzzy tech popular right now (a major sin for the Valley’s hype machine, and a major constraint for any of the Google-driven AdSense-addicted blogs considering writing about them going forward). They apparently also forgot to plan for it to scale, to devise a page ranking algorithm that does not suck, and that the bucketing of search results was being done before Google arrived, has been repeatedly reinvented by companies since then and dabbled in by Ask or even Google (see, for instance, the 90s-era HotBot and today’s Clusty), and that when best realised tends to asymptotically converge on either an ontological index or a yellow pages directory, depending on how you crowd source. Cuil is an obvious attempt to flip some Google “insider” tech and a bunch of rÃ©sumÃ©s and captive H1-Bs to one of the search giants for a quick sale.
They also pulled an Apple and claimed to have the Largest Index Ever (“120 billion!”), mere days after the Google Blog announced it had surpassed a trillion URLs. That’s some fine index work, Cuil.