I don't know.
The longer answer:
Something I always wondered about Jimmy Wales' Wikia project was how the search engine was going to establish an index of billions of web pages. Even assuming the legitimacy of user-driven search, which is itself a stretch, the technology and capital required to overpower Google in search would likely evade the overmatched Wikia.
Cuil, on the other hand, promises the ability to index the internet's hundreds of billions of web pages better than Google. Not only do they intend to index better than Google--and even claim they already have an index three times as large as the search giant--but they intend to do it with far less financial resources.
Apparently, my major argument against Wikia doesn't seem to apply to Cuil. What does make me skeptical of the new search engine is the actual product that was lauched today. If Cuil is crawling roughly 125 billion web pages, why are there some fairly substantial pages absent from search results? One decent example that demonstrates my point (and possibly my partiality) is the absense of City Dictionary from the search results. Google has already crawled and indexed the website a number of times since it went live in April. (Google currently lists nearly 10,000 pages for City Dictionary's domain in its search results.) Google has also followed several links from other websites to City Dictionary, each time placing greater importance on the new website. As a result, City Dictionary has the #2 spot in Google for 'City Dictionary' and the #1 spot for "City Dictionary" with the quotation marks.
Q: Where does City Dictionary appear in Cuil's search results?Even when I put in 'City Dictionary local flavor,' which makes an obvious reference to the tagline 'The Dictionary with Local Flavor' that appears in the 'title' tag for the homepage, the results suddenly go from somewhat relevant to completely off the radar. (Top search results include pages from lawyer.com, as well as others looking to sell "cheep airline ticket.") Bottom line, I can't find any of City Dictionary's thousands of pages in Cuil.
A: Nowhere.
Why is this important? I don't have any delusions of grandeur for a site that is only a few months old (it's still not so bad if I do say so), but City Dicitonary's complete exclusion points to a very daunting problem for Cuil. If the new search engine already claims to have a larger index, that index still does not seem to have very timely content. City Dictionary has been in three different newspapers in the past few weeks. Google finds timely content because it is constantly crawling the most popular news sources. Since a great part of search is based on time-sensitive issues, I can't help but to question the quality of a search engine that can't follow links from newspaper websites in a timely fashion.
So, does Cuil really have an indexing advantage over Google? Well, not necessarily. Google no longer publishes the size of its index, so Cuil's claims to be three times larger than Google are suspect at best. Plus, very little is known of Cuil's indexing techniques; so, the young engine's ability to (re-)crawl billions of web pages on par with the incumbent search giant is questionable.
It's not just content; it's intent. Search is still an intent-based medium. Reducing the influence of links to search engine results o take away the ultimate popularity metric on the internet. Are web pages that have backlinks from several blogs and trusted news sources of higher quality than web pages without such links? Generally speaking, yes. Of course Google will continue to sort search results by content relevancy, but the democratic system of linking is still paramount until Cuil can come up with something more compelling (which may exist but has not yet been articulated). Cuil must have an answer for the very powerful Page Rank system. Otherwise its 125 billion collection of web pages is for naught. (By 'naught' I mean being sold to Microsoft for about $500 million.)
Will Cuil end up creating great search results from its gargantuan index? We'll have to see what happens in the years to come, but for now it's not that cool.
Posts
4 comments:
Sup man-- Thanks for the heads-up tweet; I actually have a view on this. I've been out in the Valley for the last week and a half doing some high priority work for my company. I saw Cuil (fucking stupid name) break on Slashdot this morning, then saw a bunch of tweets about it, then saw a front-page story on the SF Chronicle's web site. Later this afternoon I got a call from a Product Manager at my company asking if I had seen it and what was my opinion as a developer. THEN while playing poker tonight with a bunch of developers from different companies it came up right away in conversation. Long story short: the shit hit like a tidal wave out here. And yet still I don't really care, for precisely the reason you state.
I think the Valley is getting so worked up because they WANT someone to challenge Google, if only to see if it can be done. Myself? Even as a developer interested in that kind of crap I really just care about the search results. And if Cuil (STUPID WEB 2.0 NAME!!!) can't deliver the goods then I won't bother.
Here's another example: I've had a blog since 2002 (I didn't even call it a damn 'blog' when I started it). I just searched Cuil on my domain name/blog name/user name (all the same word) and got back no results with a link to my blog. I got results with zazzle.com (!) but not my blog. I don't know what their equivalent of page rank is but it's balls right now.
I won't be switching anytime soon-- we'll see if their hype steamrolls them into enough VC funding to improve their shitty search to the point where they can start picking up market share.
hi (and lois) from san franscisco.
Thanks for that insight into the Valley perspective. I must admit my ignorance of that culture. (I haven't even been to that part of the country.)
You make a really important point, that the hype may be a self-fulfilling prophecy if it gives them enough capital to make the necessary changes to their sub-par search results.
Yeah, I agree on the name, too. "Cuil" reminds me of the way seventh grade girls try to spell "cool" in a new way. Yeah, not too "kewl."
As for your blog, it's crazy that something that's been around since 2002 can't even be recognized by its own name. If your blog's in their index, it does nobody any good if they can't retrieve it.
Thanks for your thoughtful comments!
Thought I should follow up on the name thing. I have a tolerance for web 2.0 names only to an extent. Flickr? Ok. Digg? Ok. Notice that all these names can at least be pronounced. I ain't ever took no fancy business larnin's in the schoolhouse (I am simply a software developer) but I would think that if you have to tell me how to pronounce the name of your company and core web address, you've got severe marketing problems. How many folks heard chatter about 'cool.com' and went there only to find a domain squatter (though note that in a hilarious bit of 'internet-speed' response, cool.com has turned itself into a faux-search engine).
I never thought I'd say this, but kudos to that cyber squatter. If you have the temerity to take on Google in search, you should do your homework and make sure that you've taken care of the most important defensive domains. More importantly, you should make sure you have a tested product before you launch. (Cuil doesn't even show up when searching for 'cuil.')
I am even more convinced now that Cuil simply wanted to ride the wave of hype to justify a huge valuation when getting courted by rich suitors like Microsoft. Heck, Microsoft has already shown its willingness to buy non-functional search engines for serious bank.
Post a Comment