Keith Watanabe * NET 2.0

One of the worst things around the net is the amount of redundant articles or links.  For instance, let's say you want to find information on "ubuntu hardware support" and use that as your search terms.  You might pull up links that seem to replicate each other.  Or sometimes you might encounter mirrored sites like cpan.  The biggest problem with these points of redundancy is that for the user, they must spend more time sifting through large amounts of information, thus wasting their time.  For all of google's hype with Page Rank, you pretty much get the same thing and there's really not much difference in the quality of results.

Instead, you really need stronger ways of filtering results.  One idea off the top of my head is by solidifying similar search results into a grouping.  This is what I call "Search Result Coagulation."  Truth be told, search engines really have no way of easily discerning results, so it's probably up to humans to, initially at least, inform these systems of their error.  One way would be if people could group redundant results with each other.  So you would find one result, drag it on top of another result, and it'll form a grouping.

In that sense, you don't want to tag the group, although you could.  I mean, it's just a result group.  Maybe you could have a little plus by the side of the result group that inflates the other similar results.  When people attempt to locate such results again, they too might see this grouping.

Unfortunately, this methodology is prone to human asshole-ness.  That is, the typical netjerks who screw up results for other people intentionally by purposefully mis-grouping results.

Alternatively, you could create success result clouds, which would bundle up the most useful related results for a group of search terms.

I guess I'll have to check around to see if someone has done such a thing.  Certainly, I'm not the only one who probably thought of such a thing.  But at least, there aren't anything available that's widely publicized.