More abstraction of user interface from data processing code. Since I know this will mean nothing to the non-programmers amongst us, it basically means that things like search routines return a fixed set of data every time, and _never_ concern themselves with how the data is going to be presented to the user. This should probably be taken into consideration while axkit development is taking place...
Yep, good call
1. Search interface needs to be cleaned up. The @stuff arrays confuse me. :)
Yeah I have a grand scheme for the search stuff, in terms of setting up some way so that ppl can customise the search. IE when u install koha, you can say I want to be able to search author, title, keyword in the opac. Or I want the search off the front screen of the opac to search only author's and titles. I was thinking that as mysql is heading towards stored procedures, and most other rdbms support them already it would be the way to do it. So we have some kind of customisation/admin program that can generate the stored procedures for doing the searches. Thoughts/Comments/Youve got to be joking remarks?
2. Cataloguing API - This is what I started on, but have gotten badly bogged down with. Syncing the MARC and KohaDB backends got ugly in a hurry. I need to rethink how I approached it, I think. If there are programmers on the list now who are interested in helping with this, drop me a line.
I could probably spend a little time on this, if you have some tasks, you want to throw my way.
3. Soundex searches (ie searching for elefant will turn up elephant results)
Ohh yeah, I know the librarians at HLT would love that Chris -- Chris Cormack Programmer 025 500 789 Katipo Communications Ltd chris@katipo.co.nz www.katipo.co.nz