Cuil search engine
From Drudge this morning: “Ex-Google engineers debut ‘Cuil’ way to search“.
It’s here.
Aside from the impressive numbers, these two paragraphs from the linked article got my attention:
Rather than trying to mimic Google’s method of ranking the quantity and quality of links to Web sites, Patterson says Cuil’s technology drills into the actual content of a page. And Cuil’s results will be presented in a more magazine-like format instead of just a vertical stack of Web links. Cuil’s results are displayed with more photos spread horizontally across the page and include sidebars that can be clicked on to learn more about topics related to the original search request.
Finally, Cuil is hoping to attract traffic by promising not to retain information about its users’ search histories or surfing patterns – something that Google does, much to the consternation of privacy watchdogs.
Yahoo! Whoopee!
Shouldn’t that be “kewl”?
Annnnywayyy….
Smart text processing is, of course, the holy grail of all of these search engine technologies. It’ll be interesting to see if the algorithms employed by Cuil are any better. Will it be evolutionary or revolutionary?
I remember discussing this in my AI class a couple of years ago. IBM had some interesting technologies in their labs (at least, that was being talked about publicly), but no targets for when (if) they would ever see the light of day. Fuzzy logic and neural networks are definitely at the core of any really new algorithms.
Fun stuff. I hope Cuil is able to give Google a run.
Er… have you actually tried their search engine? I can’t even find my own blog name associated with my actual blog.
It shows 10 pages but gives an error after the first. Their website is incomplete, so they obviously don’t have a marketing guru handy.
And as cool and elitist as they come across, let’s see how long before they come to realize what the general population wants from a search engine.
Their elitist “ranking” formula assumes to know more than the combined population about any given subject. If they’re not going to rank by most popular, then they have become the arbiters of information.
So far, I’m underwhelmed. The woman’s got the engineering chops, but lacks the street savvy of true market-driven business.
A search for Vodkapundit this morning turned up lots of hits. Now it comes up with nothing. But also just now I did turn the safe search option off. With it off, no vp hits; but with it on, lots of hits. Cuil apparently gets shy when its taken out of ‘safe’ mode. First date jitters, maybe.
IMHO, this “launch” trumps the messed up IPhone 2 launch. They didn’t have sufficient server capacity for their launch day, and I don’t know how they go about finding web pages, but I spend a lot of time in Google looking up different programming APIs, code snippets and the like, and every search I’ve tried today I tried in cuil first, and it continues to find nothing of use.
My hope is they last long enough to modify Google’s privacy policy, but I’m doubting they will.
I’ve tried Cuil on about every other search I’ve made since it launched. Probably 40-50 times so far. I REALLY want an alternative to Google.
IMHO, it sucks. It sure is purdy, though.
Maybe their algorithms just need more searches over time to build up “intelligence”. If so, my bet is they won’t have the chance to build critical mass. Their buzz wad is already blown.
i used cuil to search for: cuil search engine, and the site was not found. which proves that their search has nothing to do with popularity (while google finds and displays cuil in top ten links) but also proves that popularity is still important in searches, and at high frequency, could come really close to absolute relevance (if a site is really popular and ,000 visit it for a specific reason, it has to also be absolutely relevant). However, cuil may work better for sites that are not popular but still contain useful/relevant information (which brings us back to the funtamental question: why would a site with useful/relevant information be unpopular? I can only think of sites that contain information but do not have the ability to attract visitors). Anyways, it will take a lot to deviate users from Google. Just in the notion that by using cuil you may be missing sites/info, you stay with Google.
Interesting to see if cuil arights itself, I hope so. It is purdy, but purdy don’t feed the bulldog.
[quote]why would a site with useful/relevant information be unpopular?[/quote]
Perhaps because it doesn’t appear in the top page or so of google’s results, and so no-one finds it. One of the problems with a popularity-based ranking system is that is self-reinforcing, especially with a near-monopoly like google.
People find pages near the top of google’s results and so they tend to link to them.
But, as google has said often, page-rank is actually only a pretty small part of how they order pages these days. There are plenty of other criteria.
its worth a shot. i’m eager to use another search tool. ms live search is actually very good though it’ll never catch on with that brand. I give props to cuil for trying to change the way search is presented. but they have a lot of work to do.