Code4lib 2006
Focused as a highly technical event, the inaugural code4lib conference resulted from the combined efforts of the code4lib community, a loosely connected set of programmers, hackers, librarians, and library technologists who pursue technologies related to libraries and information. Over 80 attendees had the opportunity to learn what their peers were pursuing at other institutions, to share projects, hacks, code, and ideas, and to participate in breakout sessions and give lightning talks. Everything about code4lib proceeded from the involvement of the code4lib community and the attendees of the event, from the planning of logistics to the decisions about the programme, and the on-the-fly decision making and agenda shaping.
Genesis of a Conference
Code4lib 2006 [1] began as an idea that had been simmering in hallway conversations, email discussions, and chat room chatter for quite a while. In early November, the community in the #code4lib IRC channel [2] turned the rough idea of extending the 'hackfest' component of the Access conference held annually in Canada into a three day conference targeting coders and technologists who work in or with libraries. In essence, code4lib 2006 was created through a grassroots, open manner, with members of the code4lib community working together to plan, organise, and run the conference. Community voting decided everything from the conference location to the accepted presentations and the design of the conference t-shirt. Communication occurred mainly on the code4lib and code4libcon discussion lists, as well as the #code4lib IRC channel. (Conference details and information are available via the code4lib Web site [3]. Finally, members of the community helped run many of the events during the conference, including the lightning talks and breakout sessions.
Conference Overview
The code4lib conference comprised four distinct types of activity: keynote talks, proposed twenty-minute talks, five-minute lightning talks, and participatory breakout sessions. Attendees had the opportunity to participate in all of these activities, as code4lib was a single-track event (except for the breakout sessions, which ran in parallel). Information about presentations [4] can be found on the conference Web site and, by the time this article is published, audio recordings of the keynotes, presentations, and lightning talks should also be available.
Keynotes
The opening keynote speech was by the Evergreen development team of the Georgia PINES consortium. Evergreen is a state funded development effort to create an open source integrated library system. Because the Evergreen team could not attend the conference, they presented their keynote virtually with a speakerphone and slides shown in the conference hall manned by a conference attendee. This perfectly reflected the 'hacker' philosophy of the conference and also set the tone for the rest of the event, through both the breadth and technical depth of the work they presented and in how easily they overcame the technical challenges of presenting a lengthy keynote remotely. Indeed, the presenters were even able to interact directly with the audience, easily hearing and answering numerous questions.
Thom Hickey, Chief Scientist of OCLC's Office of Research, gave the second keynote highlighting the amazing things possible in as little as 1,000 lines of software code [5]. He demonstrated a number of tools created at OCLC in recent years and how they can be composed into extremely sophisticated new applications with a minimal amount of additional new software development. This keynote also provided the audience with a view into how inexpensive approaches to high-performance computing techniques may be implemented with commodity hardware. To top it all off, Thom delivered most of his keynote adorned in a shirt which itself consisted of an OAI-PMH repository.
Thom Hickey in formal code4lib attire
20-minute Talks
One of the aspects of code4lib much appreciated was the pace of presentations. Each day had a set of twenty-minute talks, whose length and subject matter kept the energy at a high level. These talks were voted in by community members from more than thirty proposals, which were solicited at short notice over several library-related email lists. Despite the short notice, the quality of all proposals was extremely high. The topics presented were diverse, ranging from practical solutions for current problems to forward-thinking digital library research topics:
- ERP Options in an OSS World (Art Rhyno) [6]
- Connecting Everything with unAPI and OPA (Dan Chudnov) [7]
- WikiD (Jeff Young) [8]
- Lipstick on a Pig: 7 Ways to Improve the Sex Life of Your OPAC (Jim Robertson) [9]
- AHAH: When Good is Better than Best (Casey Durfee) [10]
- Standards, Reusability, and the Mating Habits of Learning Content (Robby Robson) [11]
- Generating Recommendations in OPACs: Initial Results and Open Areas for Exploration (Colleen Whitney) [12]
- Library Text Mining (Robert Sanderson) [13]
- Anatomy of aDORe (Ryan Chute) [14]
- Teaching the Library and Information Community How to Remix Information (Raymond Yee) [15]
- Two Paths to Interoperable Metadata (Devon Smith) [16]
- The Case for Code4Lib 501c(3) (Roy Tennant) [17]
- Quality Metrics (Aaron Krowne) [18]
- Practical Aspects of Implementing Open Source in Armenia (Tigran Zargaryan) [19]
- What Blog Applications Can Teach Us About Library Software Architecture (Casey Bisson) [20]
Proper laptop-to-attendee ratio
Breakout Sessions
A key part of the conference lay in encouraging all participants actually to participate. Prior to and during the conference itself, attendees were encouraged to propose breakout sessions where specific topics, projects, or concepts could be discussed and in some cases worked on. Instances were as follows:
- Building a Library IM Bot
- Developing unAPI and w/unAPI
- Developing an Interface for Federated Search Engines or for an OPAC Interface
- The Future of code4lib
- More Lipstick on a Pig: Improving OPACs
Lightning Talks
These five-minute presentations, wherein any attendee could speak on any topic for five strictly timed minutes, were the biggest hit of the conference. Of the longest of three separate lightning talk sessions, one held each day of the conference, one attendee later wrote elsewhere that "the session where 22 talks were presented in rapid succession was the most intense and gratifying conference experience I have ever seen." [21] Overall, thirty-three lightning talks were presented, by twenty-eight different presenters - or almost exactly one-third of the attendees. A complete list of the lightning talks can be obtained from the conference Web site [22].
Conclusion
Code4lib 2006 proved to be an unmitigated success. For a first-time event, having over eighty participants and speakers from the U.S., Canada, the UK, and Armenia (yes, Armenia) was impressive, and for the most part, things ran smoothly and effectively. The participants overwhelmingly agreed that the single conference track (in the style of the Access conferences [23]) succeeded in keeping everybody on the same page, and participants also discussed ideas for possibly expanding the next conference to include pre- or post-conference workshops, tutorials, or code sprints. Overall, feedback was extremely positive, and many people stated that they were already pencilling in code4lib 2007 on their calendars.
References
- code4lib 2006 http://www.code4lib.org/2006/
- code4lib IRC channel http://www.code4lib.org/irc
- code4lib Web site http://code4lib.org/
- Information on presentations available via http://code4lib.org/2006/schedule/
- 1,000 Lines of Code, and other topics from OCLC Research
http://www.code4lib.org/2006/hickey - ERP Options in an OSS World
http://www.code4lib.org/2006/rhyno - Connecting Everything with unAPI and OPA
http://www.code4lib.org/2006/chudnov - WikiD http://www.code4lib.org/2006/young
- Lipstick on a Pig: 7 Ways to Improve the Sex Life of Your OPAC
http://www.code4lib.org/2006/robertson - AHAH: When Good is Better than Best
http://www.code4lib.org/2006/durfee - Standards, Reusability, and the Mating Habits of Learning Content
http://www.code4lib.org/2006/robson - Generating Recommendations in OPACs: Initial Results and Open Areas for Exploration
http://www.code4lib.org/2006/whitney - Library Text Mining http://www.code4lib.org/2006/sanderson
- Anatomy of aDORe http://www.code4lib.org/2006/liu
- Teaching the Library and Information Community How to Remix Information
http://www.code4lib.org/2006/yee - Two Paths to Interoperable Metadata http://www.code4lib.org/2006/smith
- The Case for Code4Lib 501c(3) http://www.code4lib.org/2006/tennant
- Quality Metrics http://www.code4lib.org/2006/krowne
Editor's note: readers may be interested to learn that Aaron Krowne and Urvashi Gadi have contributed an article entitled QMSearch: A Quality Metrics-aware Search Framework to this issue, see http://www.ariadne.ac.uk/issue47/krowne/ - Practical Aspects of Implementing Open Source in Armenia
http://www.code4lib.org/2006/zargaryan - What Blog Applications Can Teach Us About Library Software Architecture
http://www.code4lib.org/2006/bisson - Comment on 5-minute talks
http://webvoy.uwindsor.ca:8087/artblog/librarycog/1140232323 - code4lib Conference: lightning talks http://code4lib.org/2006/lightning
- 2005's hosts were Edmonton, Alberta, Canada and provided links forward and back at:
http://access2005.library.ualberta.ca/