Kosmix Vs. Google

No Worries!

I have been hearing for some time now that we will soon be seeing many new search engines, and that Google's days are numbered. Given what I have seen and read so far, the developers at Google need not lose much sleep.

One of the new search engines around is Kosmix and I have to say that I am deeply unimpressed by the engine. It seems to give very superficial results and leans far too heavily on the sometimes unreliable Wikipedia. But the developers have some good ideas. The engine is editorialized. It makes use of about 1, 00,000 sites that the team has decided as reliable.

This means that if the subject of interest is something that they have not considered, or might be found on a site with which they are not familiar; your search will miss out. Now, the developers claim that if you want specific information then perhaps you should use Google, but if you want 'content based on the underlying concept" then you should use Kosmix. I have not figured out what that is supposed to mean, but if the subject is not covered in those 1, 00,000 sites then you will get nothing. If the engine cannot find what you want, it serves up Google results. It is very easy to enter a query that produces such a result. This site brings new meaning t the term 'Limited Area Search Engine'.

Kosmix is certainly not a semantic search engine. The idea behind such a search engine is that it will be able to understand questions typed by the user and produce appropriate results accordingly. For example, you may be interested in a controversy related to GM foods, and want to ask the question, "Why do some criticize the use of promoter gene from tobacco mosaic virus?" Type some of those words into Google and you will get a mix of results. Go through a few of them and you will almost certainly find the answer. A semantic search engine should, however, get you straight to the right answer.