With amazing fanfare, the new search competitor, Cuil has emerged from stealth. Cuil (pronounced ‘cool’) is trying to change our habits and grab some of the huge search market. Arrington seems to suggest great things for them, or at least states that their index is large:
Cuil also claims to have better search results than Google and others based on how they index websites. They do not simply catalog keywords on a site and then rank the site based on its importance. They also work to understand how words are related (France – cheese – wine, for example), to return more relevant results to users. This is a semantic approach to search, but very different from Powerset’s natural language approach. Powerset uses artificial intelligence to try to understand what sentences on a website actually mean. Cuil, by comparison, simply tries to properly categorize and file a web page, even if the category name doesn’t appear on the site.
That means users search the same way they always have, but Cuil will try to return better results via refinements in a “explore by category” module to the right of results. A search for dogs, for example, will return category results for “water dogs,” “crossbreed,” “cocker spaniel,” etc. Some of these related terms do not include the term “dog.”
Cuil is another take on semantic search – knowing how words relate is all semantics – but my results were so-so at the best and terrible at the worst which was the case with Hakia, another semantic search firm.
Searching with the phrase “C++ operator order of precedence” led to zero results:
This example is always interesting, because Google comes up with an authoritative answer as number 1 (see http://www.cppreference.com/operator_precedence.html).
I tried searching for Anais Nin, the author of erotica, and got this:
Which is ok, but I guess I would have expected to see the Wikipedia entry somewhere, and perhaps the anaisnin.com website.
The semantic categories seem to hold up. She acted in a number of files, and was of Cuban and French ancestry, and worked as a model in her teens. I can find no proof of her having been a dancer. I think the semantic categorization should link to some evidence. She was born in the outskirts of Paris, so the ‘People from Eure’ – Eure is a part of France in Normandy – seems to be just wrong.
On the other hand, I have become accustomed to the Google style of search results, which may bias me. But if a new competitor comes along hoping to shake me out of the rut I have fallen into it has to offer something with a bit more than – perhaps incorrect – semantic categories.
And when I try a term that has semantically varied meanings – ‘casting’ – I don’t find Cuil giving me the sort of semantic disambiguation I might.
It provides Sculpture Techniques and Jewelry Making as two good categories. But Christian Rock Groups? It appears there is a Christian rock group called Casting Crowns, but come on. An Reality TV Series seems from left field.
It may be that Cuil and other semantic search tools will require a great investment of feedback from users before the searches can become truly reliable. But I don’t see where Cuil is getting that info, aside from the clickstream.
It’s one thing to argue that popularity isn’t everything in search results, but semantic jumps that are just wrong create more problems than they solve.