-----Original Message----- From: Rhyno Art [mailto:arhyno@server.uwindsor.ca] Sent: Monday, September 23, 2002 6:57 AM To: discuss@callimachus.org Subject: Re: building a community

Wow, very impressive, Koha is doing great and is a much needed success story for OSS in libraries.

Thanks, it's always nice to hear kudos.

I am involved in two library-related open source projects, one called PYTHEAS <http://www.uwindsor.ca/library/leddy/people/art/pytheas> and another called WIBS <http://wibs.sourceforge.net>. PYTHEAS started out somewhat similar to Koha, it had some PHP instead of Perl, but used MySQL on the backend, and even had a middle layer written in C. I managed to get a 6 month study leave to work on OSS, about half of which I used to study the current trends in how to put enterprise systems together. Thanks to initiatives like Koha and Avanti, my feeling was that the options for small to medium sized libraries were becoming much better defined, but 15 years of working in a university environment had made me yearn for an option for large academic libraries, and one that plugged into other environments. It has always bothered me from a systems viewpoint that universities constantly replicate systems between departments and that mainstream software, such as LDAP servers, were very rare for university administrative systems.

I agree, it would be nice to see more emphasis on getting LDAP and similar tools into libraries. Perhaps turnkey system approaches built from OSS are the key to proving that this kind of tech not only works, but is a good fit for libraries.

Anyway, my study leave led me to look closely at Enterprise Java Beans (EJB), which latter lead to Web Services and SOAP. I was also trying to maximize the use of XML since one of my first tasks as a systems librarian was working on a MARC editor for Stanford's SPIRES database and rather than repeating the experience, I felt that XML was the key to freeing MARC from quirky editing applications. None of this is an attempt to convince anyone of the benefits of Java, XML, or middleware infrastructures like EJB, but one area where I think we could share some of our efforts is in how we map our data structures. For example, I think Koha's MARC work, at one time at least, used an early version of the table layout defined for PYTHEAS. Lots of work and mistakes since then have led to numerous changes in the layout and I would be very interested in seeing if some of this learning process would be of use to anyone else. For that matter, any of the projects that use an RDBMS on the backend could probably benefit from sharing ideas on mapping and information retrieval with relational systems.

I agree, this might be an excellent place for callimachus to shine.

WIBS was based on the MRBS (Meeting Room Booking System) and was a much simpler endeavor. In fact, it was mainly put together to avoid coming up with the outrageous amounts that the commercial options for workstation booking seemed to require. Like MRBS, WIBS is a PHP/MySql application but I am currently reworking it to use the University of Washington Calendar Server.

Is this an ICal (and related tech) thing? Do you have an URL for it?

Again, this ventures into Java territory, but my switch is motivated by similar concerns. We would like to pass booking information between the university's main calendar systems (Lotus Notes) and personal calendar systems, and a true calendar server provides the plumbing to easily support iCalendar and other standard data formats for calendar information. This could all be built by hand in PHP but my minimal amount of time for OSS activities is already shrinking and it didn't make sense for me to rebuilt what others have done. I am also keen to keep in line with the Java in Administration Special Interest Group (JA-SIG) <http://www.ja-sig.org/> and Java is a key part of this (though as I have told Pat, if I had my druthers, I would be working in nothing but Python!).

I have found that in a university environment, Java tends to open a lot of doors, not because of anything intrinsic to the language, but because it has such a rich and evolving set of APIs. Still, I have been through more grief trying to help people set up Java-based applications like Tomcat than anything I have seen with PHP, Perl, or other common types of OSS solutions, so I am glad to see a variety of options available for libraries installing systems. Unlike Koha, my sharing of code has largely been directly with other universities, and most of the projects I have seen from this have been very low profile. I am blown away by how well keeping the project in the public eye has served Koha and my hat is off to Pat and all the folks who have made this happen. Remember, I am still hoping at least one spokesperson for Koha might be at the upcoming Access conference <http://www.access.uwindsor.ca> and could give an update at the software rodeo we are trying to put together.

I'd like to get us there. Any volunteers from the devel list? Thanks, -pate

art --- Art Rhyno, Systems Librarian Leddy Library, University of Windsor Internet: arhyno@uwindsor.ca Tel: (519) 253-4232, EXT. 3163 FAX: (519) 973-7076 WWW: <http://www.uwindsor.ca/library/leddy/people/art>