That wasn't quiet what I was asking. I'm wondering about how you want to rate
the results of searches, beond simply matching the keywords. For searches to
work well, it is not really enough that one gets a list of a bunch of keys that
claim to pertain to search keys, but that they are rated somehow. 

So if a person requests the keywords "Teletubbies" and "pictures", and he gets
a list of 20 keys, he probably does not want to request the first 10 to just
find out they are falsely indexed porn adverts. 

Your "unrequests" are a form of voting in my opinion, but I don't see how it can
be directly translated into rating of the actual relevance of search results.

On Thu, 18 May 2000, Ian Clarke wrote:
> > How do you plan to combat littering of the search results? Like the way 
> > that a
> > lot of spam pages will attach just about any popular keyword in the 
> > meta-data
> > on the web?
> 
> Because a hit on some data's search string won't actually benefit the
> data unless someone, having retrieved the search string, requests the
> data.  This does raise the issue of how to prevent "stray" search
> strings which point to deceased data, and I think the answer probably
> lies in unrequests (if the data is not found an unrequest is sent
> immediately).  Unrequests are actually something else I want to get
> stuck into, their implementation should be straight-forward.
> 
> > I'm guessing you want to see some sort of voting, which might work, but I 
> > don't
> > know any search engines that work by having the users vote on the validity 
> > of
> > the hits, so it is hard to say exactly...
> 
> Voting is dangerous as it risks censorship by majority, so I think
> unrequests are a good solution to this, and work sorta like voting.
> 
> Please say everyone agrees with unrequests? (I like getting consensus
> opinion on things, but I hope this doesn't result in paralysis as people
> insist on giving their two cents regardless of whether it slows down
> implementation).
> 
> Ian.
> 
> _______________________________________________
> Freenet-dev mailing list
> Freenet-dev at lists.sourceforge.net
> http://lists.sourceforge.net/mailman/listinfo/freenet-dev
-- 

Oskar Sandberg
md98-osa at nada.kth.se

_______________________________________________
Freenet-dev mailing list
Freenet-dev at lists.sourceforge.net
http://lists.sourceforge.net/mailman/listinfo/freenet-dev

Reply via email to