FLOSS Weekly 132

From The Official TWiT Wiki
Jump to: navigation, search
FLOSS Weekly
Episode 132


Dan Scott has been the systems librarian for Laurentian University since 2006. Prior to joining Laurentian, Dan worked at IBM in Software Group as a technical writer, information architect, and product manager for DB2 Universal Database. His first taste of Linux came in 1998 testing the skunkworks port of DB2 to Linux and made his first contributions to the open source community shortly after, first via The Linux Documentation Project, then getting involved with the PHP community by writing documentation and tests for the PHP Data Objects (PDO) extension. He keeps his toes active in the PHP world as the maintainer of the PEAR File_MARC library. Despite avoiding Java whenever possible, Dan co-authored a book about the open-source Java database, Apache Derby ("Apache Derby: Off to the Races").

Dan joined the Evergreen project in late 2006, shortly after Georgia PINES went live. After peppering the original developers with questions and patches, he was added as the first committer outside of the state of Georgia in September 2007. These days he is a jack of all trades for the project: providing support on IRC and the mailing lists, maintaining install documentation and developer-oriented tutorials, contributing code, wrangling bugs, and working on forming the Evergreen Software Foundation.

  • Mike Rylander

Mike Rylander is the co-founder and Vice President of R&D at Equinox Software, a company built specifically to serve the needs of libraries looking to implement Open Source software solutions, where he leads the technical design of much of Evergreen.

Before starting the development of Evergreen at the Georgia Public Library Service, Mike spent much of the preceding decade building large database- backed applications and enterprise monitoring systems, as well as system administration and security tools.

As one of the original Evergreen developers, Mike has brought his knowledge of database architecture, information search and retrieval, and low-level systems to Evergreen since its inception in August, 2004. He is intimately familiar with traditional library data standards and technology such as MARC and Z39.50, as well as emerging standards such as MODS/MADS and SRW/U. Mike is an active member of the library technical community who has participated in the design of both unAPI and the CQL Bibliographic Searching proposal.


Evergreen Library System is open source highly scalable software that helps libraries manage, catalog and circulate their materials.

  • Integrated Library System
    • Track books
    • Manage users
    • Notify users
  • Share data between different library systems
  • Share technical side with the software, save money on technical support/feature development
  • Originally designed to replace expensive Sun hardware running on solaris, the old system didn't use modern computer architecture
    • Spend the upgrade money on software (running on commodity hardware) that is up-to-date
  • Inventory control, built on Mozilla XML User Interface Language (XUL) framework
  • Built on OpenSRF (Service Request Framework)
  • Runs on Linux (initially on Debian)
    • Commodity x86 hardware
  • Build
    • Uses rpm/deb for dependencies
    • Uses tar-ball for actual core functionality
      • Looking for autoconf/automake wizards to improve build, and packaging experts to push out to the distros
  • OpenSRF runs on XMPP over JSON
    • eJabberd server
  • Main language - Perl
    • Some C optimization
    • JavaScript with Dojo for frontend (enduser and staff interfaces)
  • Evergreen supported by Equinox as well as several other companies
  • Evergreen Effect
    • During first usage, massive increase in cross-library sharing of material
    • Due to the fact that there was a way manage other locations materials in an automated fashion
      • Geographically aware, will find you the easiest/closest copy
  • Other open source library system Koha
    • Designed for Y2K problem
    • Designed for single libraries
    • Originally didn't do MARC
    • Not quite as scalable as evergreen
    • Share a lot of base libraries between Evergreen and Koha
  • Wikileaks document
    • FUD document] from a closed-source library vendor that was being spread around to raise doubt about evergreen
    • This release helped evergreen by pointing out that there weren't really big issues when they were raised in the light

External Links

  • IRC channel: #evergreen on Freenode

Questions from the audience


Production Information

  • Edited by:
  • Notes:
Info.png This area is for use by TWiT staff only. Please do not add or edit any content within this section.