Skip to Content

Lessons learned from the projectDB workshops

Printer-friendly versionPrinter-friendly version
Spring 2009

- Margaret O’Brien (SBC)

The 2008-2009 workshops to plan and construct the Projects Database (“projectDB”) employed a different model of collaborative development for the Information Managers Committee. In these workshops information managers dedicated time to developing a common solution for a common need rather than proceeding individually or using occasional meetings to advance a collaborative project. Participation was entirely voluntary; a total of ten sites were represented in the planning workshop, and six of these in the coding workshop. I believe that everyone involved agrees that using a workshop format was a good choice, particularly in three areas: speed of output, the general scope that a network-wide workshop accommodates, and that workshops both take advantage or our diverse skills and promote the development of new ones.

  1. Speed. Most obviously, it proved efficient for a group to dedicate a block of time to developing code. As expected, workshops resulted in much faster output than the model in which information managers try to advance a collaborative project with occasional meetings. This is not intended to detract from our use of VTCs, but to efficiently create code, there is no better way than to remove local distractions.
  2. Scope. Many site IMs had expressed the need for a database for their sites’ research projects, and some were considering development. At the first workshop, nearly half the sites were represented. This broad scope generated many use cases: a container for research products (citations, datasets, or images), a record of visitors and/or permits at a field station, or consolidated text for creating funding agency reports. All of these uses were accommodated by the schema. In considering the range of needs and required code, we also gained knowledge about the amount of work which could be reasonably accomplished in one week.
  3. Skills. By collectively designing a modular project, each participant was able to focus in a specific area. This meant that we could take advantage of our combined expertise in XML, EML, XSLT and various web development strategies. By choosing the EML-project module, the workshop participants gained in-depth knowledge about other components of that schema, and the IMC is actively contributing to EML’s further development. Additionally, the timing of the projectDB workshops, i.e., following the 2008 training workshop on XML technologies, means that most sites are now equipped to make use of the new database with little additional training (e.g., see the article by Wade Sheldon in this issue).

Coding workshops represent a different model of collaborative development to address our common needs, as opposed to relying on the usual 1 IM:1 site model in which one person covers all bases– a situation that will become increasingly difficult to maintain as internet technology grows. The participation of site information managers in the planning of network-level databases avoids the establishment of central, top-down repositories and their mandated contributions in favor of databases which both address the needs of the members and promote the network “look-and-feel”. Working collectively on a modular project allows each participant to further his/her expertise in a chosen area. This collaborative model could help to define standards and practices for developing network-level cyberinfrastructure.