code4lib northwest, open source IR software

No question, I've been neglecting this space as I've gone bonkers with Twitter.

Today, I'll be attending a code4lib conference for the northwest region in downtown Portland. I attended the 2006 code4lib meeting in Corvallis. Some of the presentations and discussions at that meeting were quite intriguing (e.g., discussions on the future of the ILS and the need for better assessment options from ILS vendors). Some were crude (e.g., a virtual presentation on the Evergreen OS ILS, in which a presenter first bragged on the exclusion of librarians from the development team, then noted later in his remarks, when struggling with a point, that he had suffered a "brain fart" (sic). I witnessed heavy drinking at evening events and the ostracism of multiple attendees who didn't fit the model of a "progressive," late-30s, OS geek who's shacked up in his parents' basement.

Based upon my experiences since that time, I now see the pure OS position as analogous to the pure vendor-dependence position, with both entailing significant service risks for an institution. The implementation of OCLC WorldCat Navigator in the Alliance, as one example, illustrates why working with a cooperative is a more reliable development path for libraries, and it's not surprising that OCLC's published WorldCat API and efforts to integrate metasearch into its WorldCat Local product have come under attack from the open source community (http://oregonstate.edu/~reeset/blog/archives/623, point 6).

The archival software community is another case where the pure OS position has led to serious technical problems. Both the Archon and Archivists Toolkit projects are commendable efforts, but both rely upon mapping XML to a RDBMS; for AT, I spoke with a project developer who confirmed to me that an RDBMS was selected for storage because of developer preference.

The one place where OS solutions have taken the strongest hold in the region is in the institutional repository/durable digial repository area, with the four largest public academics in Washington and Oregon all using the DSpace software stack to support their services. My view: the regional OS community should build on this foundation and focus less on moving institutions from relatively low-cost commerical solutions that work to roughly-comprable OS solutions.