I'm responding to the koha-devel list as well to ensure that the people most involved with the things Sergey's touched on hear his comments. On Thu, 25 Apr 2002, Sergey A Yanovitsky wrote:
Hello Pat,
Monday, April 22, 2002, 11:26:37 PM, you wrote:
PE> Hi Jah, hi PE> thought I'd check in and see what your thoughts were on koha. Let me know PE> what you (and the others working with you) think, okay?
It's a nice system. You guys did lots of work there. :))
But it seems that it is growing without particular plan. Database sources are just dump from someone's (maybe your leader's) working DB and they really need normalization. It would be nice to make a diagram, that represents your database. Something like the one that i attached to this letter (i made it in Graphviz).
I know that some people are looking into the DB. I don't know what their plans are, so I can't really comment on them.
There is also no unified api for working with web forms. And it seems that it would be hard to extend your system :( I think you should look at something like Mason.
I agree, some kind of templating, and an overall shift to a separation of logic and presentation makes sense
Or, if you plan to shift to ruby -- mod_ruby+eRuby.
I think I'm the only rubyist working on koha, so I doubt a shift to ruby is in the works.
My expirience says that php will not help here. It's too messy for such a big project.
agreed
Unfortunately i can't say more by now. Too much work :( I asked my teammates to look at Koha. Later i'll send their opinions to you.
Please inlcude the koha-devel list, I'm afraid I'd step on information when I tried to pass it along and I'm sure the other folks working on koha would appreciate seeing them.
Also i'd like to speak with the guy/gal who wrote Koha's databases.
I'm not sure who it is, but he/she should be here on the devel list.
-pate
Hi ya,
Also i'd like to speak with the guy/gal who wrote Koha's databases.
I'm not sure who it is, but he/she should be here on the devel list.
I guess that would be Olwen Williams and I. Brief Koha history for those who arent aware. Koha was written because HLT needed a new library system. Their existing system had some serious y2k bugs and just wasnt going to work anymore. They needed something that would work over phonelines to their branches. Thus Koha was born. And the race to get it done before jan 1 2000 started. The database grew basically as a structure that would allow us to get the data out of their old system and into koha as fast as possible and without losing data. As Sergey has mentioned its a fairly organic structure, and needs some direction and normalisation. Where am I coming from? Well I really want koha to be used by lots of ppl and to be a useful and complete tool. But I also want it to still be usable by HLT. So any database changes need to allow the existing data to be mapped to the new structure. And that HLT continue to have a system that functions the way they want. (If that makes sense) Steve Tonnessen has been working on database changes to get MARC in and out of koha. And I think Paul Polain is keen to help out on that. I did draw some diagrams in the dim dark past, which only bear some relationship to the current db structure now. A UML diagram or ERD would be a great place to start at least describing the current structure. Then we can work on a new one Well thats my friday morning thoughts anyway :) Chris -- Chris Cormack Programmer 025 500 789 Katipo Communications Ltd chris@katipo.co.nz www.katipo.co.nz
participants (2)
-
Chris Cormack -
Pat Eyler