> The issue is not so much whether *any* search mechanism has undesirable
> properties, it's whether this particular search mechanism will work.
> It's not clear that a search mechanism with these properties will be
> adequate to satisfy the things people want to do with searching.

My point is that the criticisms you are making (regardless of whether I
agree with them - which I don't) are applicable to any reasonable search
mechanism with the features I think a search mechanism will need.  Given
that we really need a search mechanism, we may as well try it, if it
doesn't work - so-be-it.  There are probably hundreds of arguments
behind why Freenet won't work, but we tried it anyway on the basis that
it was worth the risk.

> For one thing, it appears that the more keywords that are used during
> insert, the less likely the data is to be found.  (You haven't said much
> about how inserts would be routed, so I am not 100% certain about this.)

I did intend to be clear on the routing.  Basically the search is routed
to whatever node is referenced by the key which best matches the search
criteria, backtracking can also be permitted here, just like in normal
searching.  There will, of course, be skill involved in constructing a
search which can be found by the system, just as there is skill required
in constructing a search for any web-based system (eg. altavista).  Part
of this will be us configuring the fuzzy operators to work in a sensible
manner (for example, the and operator probably shouldn't be a simple
min() operator since this disregards any variation in the stronger of
the anded branches of the search).  

> The reason is because searches would need to use substantially the same
> keywords as the insert in order to route to the same place.  If I insert
> with 12 keywords and someone only searches for four of them, the search
> is unlikely to find the data (the first time) because those other eight
> keywords will influence where the insert went, and the search won't
> know them.

Yes, but you are failing to take into account the fact that the topology
of the network is actually determined by the process of searching
itself, not so much inserting.  Inserting should try to place things
where they are likely to be searched for (as this means there will be
less of a "settling in" time for information), but I agree that this is
more difficult to predict with a more flexible search mechanism such as
my proposal - still, we can do our best.  In this case, what this means
is that the system will probably adapt to route frequently occuring
searches (such as "'tori amos' or 'kate bush'") more effectively than
infrequently occuring searches (such as "'spice girls' or 'metallica'").

Ian.

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

Reply via email to