Wally Grotophorst from GMU says: "OSCR uses Perl & MySQL to provide electronic reserves services at an academic library...." and that they are "entering their second year on the product." Download it yourself at timesync.gmu.edu/OSCR, it is under the GPL.
as seen at freshmeat: "Gutenbook is a personal multi-stage project to design, prototype, and develop a GTK+-based desktop application for reading Project Gutenberg Etexts. Currently, development consists of advanced prototyping in Perl/GTK+." According to the author, Lee, this is the original codebase from which the other Gutenbook project started, but this one is purely focused on perl/gtk+ instead of multiple (incompatible) codebases.
as seen at freshmeat, and at SIXPACK site: "SIXPACK is a free BibTeX and Reference Manager designed to edit, convert and manage reference files, search and sort bibliographies, import and export many different bibliography types. Sixpack uses the excelent perl package bp by Dana Jacobson to covert between formats." and it reads/writes bibtex, refer, endnote, tip, procite, rfc1807, and plain text along with one-way for several others. written in perl/tk... looks real good.
as seen at gnome.org: "This version is only a minor bug-fixes release." Btw I've been using this a bit and I'm getting ready to let go of Endnote for good. Wooh-hooh... :)
Word from Eric S. at OSU is that Prospero has been downloaded more than 150 times by folks from more than 10 countries... v1.03 includes several bugfixes and a significant performance improvement. What are you waiting for? :)
the Simon Fraser University Library Research Instrument (SLRI) is "a web to Z39.50 client interface" brought to you by the good folks at SFU. it's an adaptation of the web to Z39.50 gateway developed by Harold Finkbeiner at Stanford, licensed under GPL and recently spied at sourceforge.net as well.
as seen at freshmeat, improvements include "Added keybindings, page numbering, Etext title display and the shell of searching functionality. Sorry, searching will be in the next release. Also using make for building the tarball, installing and uninstalling." more available at Gutenbook site...
as seen at freshmeat: "Over 14,000 (65%) of serials cross-referenced now have Library of Congress subject headings, classifications, or Dewey classifications, and jake may be searched according to these, initial support for a complete data localization framework so libraries may experiment with customizing front-ends and results according to local holdings, and several bugfixes and UI tweaks, including easy-to-download database content lists." more at the jake site...