Search engine.. with a difference

[Mrblogs]

Now, we all know that google has some serious computational power behind it, and that is how it can index so many sites, and keep so much of its information up to date.

But, how about if the Sites that want to be indexed Also perform part of the work?

Let us assume that the central search server has some code, that is quite time consuming (as it has to do all the algos and stuff for each site), and being as it is only one server - would be quite slow at indexing anywhere near the amount of sites google does.

But how about if the sites that want indexing merely upload a small file to their server, and then their server indexes itself and merely sends the central search server the finished information?

of course - the main issue there would be security, making sure that the central server didnt misuse its privileges on the indexing server - but theoretically, it may be useful?

More so, if you consider then that it could perhaps just store "general" information on the central server, and then request further information off the actual servers on which various sites are located. So say someone was searching for "Holidays Abroad".. it would have a list of websites that are used for that keyword.. and then go off to the websites to bring back more information about their usage with that keyword.