Claiming to be the biggest search engine on the web is a pretty confident thing to do on the day that your search engine launches, but that is exactly what the creators of Cuil have done today.
Combining the "biggest Web index" with "content-based relevance methods" whereby results are organised by way of ideas, is what Cuil hopes will differentiate it in this already crowded market.
Cuil claims 120 billion indexed Web pages, which it maintains is a stonking three times more than any other search engine. Google has just revealed that during its automatic indexing process it recently counted 1 trillion unique Web pages but it admits that it did not index all of them. There was good reason for this, as many are dynamically generated copies of the same URL and many have duplicated content. So it remains to be seen just how Cuil has come up with a figure that apparently obliterates the current search king indexing number on day one.
Certainly it talks a good fight, with a little dig at Google in the press release saying that it goes beyond the normal 'link analysis and traffic ranking' of its competitors in order analyze the context of each page and the concepts behind each query. It then organizes similar search results into groups and sorts them by category.
There is also no shortage of technical pedigree at Cuil, which again sounds odd when it has only just launched but bear with me. You see the technology that drives this new search engine was developed by a company with the husband and wife team of Tom Costello and Anna Patterson at the helm. Costello is best known for his pioneering research work into search while at Stanford University and which continued later at IBM. His wife, however, has perhaps the more interesting resume of the two: she was formerly the architect of Google's search index having led the web page ranking team. Throw in the weight of her former Google colleague Russell Power and you can see why people might just take Cuil seriously.
“Since we met at Stanford, Tom and I have shared a vision of the ideal search engine,” said Anna Patterson, President and COO of Cuil. “Our team approaches search differently. By leveraging our expertise in search architecture and relevance methods, we’ve built a more efficient yet richer search engine from the ground up. The Internet has grown and we think it’s time search did too.”
Cuil’s methods guarantee online privacy for searchers. Since the search engine ranks pages based on content instead of number of clicks, personal data collection is unnecessary, so personal search history is always private.
So what do you actually get at Cuil? here's a list of its main features:
- Biggest Internet search engine—Cuil has indexed 120 billion Web pages, 3x more than any other search engine
- Organized results—Cuil’s magazine-style layout separates results by subject and allows further search by concept or category
- Different results—Unlike other search engines, Cuil ranks results by the content on each page, not its popularity
- Complete privacy protection—Cuil does not keep any personally identifiable information on users or their search histories
Which looks good on paper, but paper makes for a bad tasting cake in my experience. Indeed, the only proof of any pudding is in the eating, so let's see how Cuil actually compares with Google head to head. These are not laboratory condition tests, nor meant to be interpreted as such, but instead just reflect an average user looking for stuff. Which is what a search engine is meant to address, in my never humble opinion.
My first search was the single keyword 'Mojave' as there has been a storm of online publicity surrounding the Microsoft Mojave Experiment which told users they were seeing a new version of the Windows OS when in fact it was just Vista and Microsoft had lied about the name in order to prove a point. Because of the amount of coverage over the weekend, and with Microsoft broadcasting a video of the experiment in action tomorrow, I thought that Cuil might be able to deliver the goods.
In fact there was no mention of the Microsoft Mojave site, news or coverage anywhere to be found on the first page of admittedly well laid out results. Hit number four at Google was a collection of news results for Mojave which covered exactly what I was looking for.
OK, let's try something a little more obvious and not open to any misinterpretation then, in order to give the new search boy a chance: how about 'iPhone 3G PAYG' which I hope will tell me something about the availability of the Jesus Phone on a pay as you go basis. I already know, having read the DaniWeb blog entry on the subject that these will not be available here in the UK until Christmas. I thought it would be interesting to see how quickly either search engine could inform me of the same thing.
Google managed it in just the fourth hit in the list, courtesy of its automatic localisation which delivers me (as a UK user) with UK relevant results first. Cuil managed to get some UK specific stuff on the first page, but nothing relating to the Christmas delays and it has to be said when compared to the vanilla list approach from Google the Cuil layout is hard to scan quickly. The extract paras for each hit are confusing as they often include header information all within a chunk of text.
Not half as confusing as what happened when I clicked to get the second page of results and was greeted by an error message informing me that 'We didn't find any results for iPhone 3G PAYG' apparently unaware that I had just been looking at a page of them. Cuil suggested it was a type, or perhaps a very rare search term (iPhone - very rare indeed - not) or just too many search terms. I suggest it might be teething troubles, and ones that lead me to think that simply claiming the biggest search index does not the best search engine make.
Finally, I hit it with one of my own blog postings. A DaniWeb news story for which I picked up the Best Information Security News Story award earlier this year and which helped me towards winning the overall Information Security Journalist of the Year title as well. 'TomTom admits Satnav device is infected with virus' got news coverage cross the web, with links to major publications and mentions at the like of Wikipedia, PC World and ZDNet for example. I used a straight cut and paste of the news story title for my search. Both Google and Cuil returned this as the first hit, which was satisfying.
However, overall there is little doubt that Google delivered upon my search requirements in a much more effective manner. Indeed, Cuil only managed to find one out of the three pieces of information I was after, at least as far as first page hits are concerned.
I will leave you to perform your own tests and come to your own conclusions. Maybe you would be kind enough to publish the results here in the form of a comment to this posting?