[Koha-devel] Website/Interface Design Focus Group

[Joshua Ferraro]

On Wed, May 25, 2005 at 01:57:03PM -0700, Corey Burger wrote:
> Enough rants, do we have a spec for BetterSearch, or should I go about
> dreaming one up?
We've been working on it and 2.4 will include BetterSearch ;-). How
that will go down is unclear at this point. I've outlined some methods
on the 2.4 scratch pad TODO list:
http://wiki.liblime.com/doku.php?id=koha24rmnotes
 (under "Searching" there are some links to ideas we've had)

Chris committed some basic Plucene textual database scripts for Koha
into HEAD (CVS for 'development stuff') and I've created a
Plucene index and got the searching up and running on NPL's
150,000 biblio database. Unfortunately, based on my initial tests,
Plucene doesn't offer much in the way of speed (which is one of
the biggest problems at the moment): with NPL's data it often
took 14-15 seconds to return a list of items which matched the
terms used. I haven't completely given up on Plucene yet as we
may not be optimizing the indexes and there may be some ways to
get more speed out of it.

The second option suggested by Indradg is Namazu which is also
a full text indexing/search engine but I haven't had a chance
to play with it yet.

I still think full text indexes are the way to go as they would
allow much more flexibility than SQL affords.

So this is an example where we've got some overlap between 'design'
and 'new features'. Corey, if you've got some ideas for how
searching could be improved, please let us know (you can email the
list or just edit the scratch pad I alluded to earlier) -- in the
meantime, I'll try to get a sample of Namazu up and running so we
can test it out. Does anyone else have searching ideas?

-- 
Joshua Ferraro               VENDOR SERVICES FOR OPEN-SOURCE SOFTWARE
President, Technology       migration, training, maintenance, support
LibLime                  Koha ILS, Mambo Intranet, DiscrimiNet Filter
jmf at liblime.com |  Full Demos at http://liblime.com  |  1(888)KohaILS