Asheville hosted a fantastic code4lib. Here are a few highlights:
- Tuesday’s keynote. Cathy Marshall (Senior Researcher, Microsoft Research) discussed the nature of living digitally, where, for many, loss is an acceptable means of culling collections, where benign neglect is the de facto stewardship technique.
- Galactic glitter glue with space debris. There was a proposal for the code4lib community to pursue cloud4lib, a cloud platform that would enable libraries to build and use a common infrastructure and service layer, the glue to hold everything together. Development enhances the entire platform, not just a single product or installation.
- Public data. There are 3 cloud service models: infrastructure as service, platform as service, and software as service. Related to these service models is data in the cloud. One example is Google Fusion Tables, an experimental system for data management, collaboration, and visualization from Google Labs. For libraries, the cloud could include institutional data as a service.
- Agile development. This one is all about IT project management and development cycles. When priorities and requirements change frequently (or are undefined) and others see IT as a black box, is it possible to build both software and trust? Sprint planning and iterative development make it possible to set priorities and to commit to certain functionality collaboratively.
- Vampires vs. werewolves. How do you balance a stable production environment with a rapid upgrade cycle? How do you balance the needs of sysadmins and developers? You use puppet and nagios.
- Thursday’s keynote. Paul Jones (Director of ibiblio.org and Associate Professor, UNC SILS) talked about Dunbar’s Number, attributed source (where liking and quality are functions of the communication source), and the changing nature of liabilities and assets on the social web (attention deficit is to multitasking as jargon is to slang as idiocentric humor is to internet memes).