Code4lib 2007, Day 3, Talks 2007 March 2
Terry Reese: LibraryFind
The development team is Jeremy Frumkin and Terry Reese and two offsite hackers Dan Chudnov and Tami Herlocker. LibraryFind is a Ruby on Rails application. LibraryFind is a metasearch tool(where items are harvested and federated), openURL resolver, openURL server, Web-service, Open Source and “one component of OSU’s vision of the library as a platform.” Some of the unique things about Library Find is that it has integrates OpenURL resolution both harvested and federated, utilizes a metadata-based knowledge base system, and has caching to make the federated search appear faster than it really is. Currently OSU’s instance of Library Find is harvesting and indexing 65 databases ever 5 days.
Michael Doran: The Intellectual Property Disclosure Process: Releasing Open Source Software in Academia.
Michael discussed the issues of creating OSS in Academia. Intellectual Property in aunt shell is basically “Intellectual property… blah blah blah…. belongs to the University” and “The exceptions that you thought belonged to you, don’t.” Michael said you need to use some common sense to see if it is relevant to your software. Nobody cares about your 50 line bash backup script, but if it is something bigger such as LibraryFind, you need to go through the process. Michael went on to describe the whole process of doing this at UTA. He described the issues he had with releasing software. You need to explain how the OSS model works and why it is useful for the Library (and the university as a whole).
Casey Durfee: Open-source Endeca in 250 lines or less.
Casey discussed the idea that the number of bugs increase exponentially with the number of lines of code. He keeps the programs short by using building blocks that others have put together. The building blocks he used are Solr, django, and other open source tools.This program looks pretty neat, but isn’t in production yet. It still needs more work.