Adonis Meeting

I was a meeting organized by the Adonis project (See TGE Adonis | Très grand équipement du CNRS pour les sciences humaines et sociales) to look at international collaboration. Adonis is running a number interesting projects:

  • Revues.org is a platform for e-journals in France.
  • Calenda is a shared calendar of events for French academics.
  • Hypotheses is a shared blog environment for news about projects.
  • Lodel is their content management system for publications.

Some other projects mentioned were:

  • Plume hosts and lets people discover open source software from university research projects.
  • SourceSup is a project management and code versioning environment for academic projects.

We are struggling with issues of international collaboration, archiving data, interoperation and so on. We all see the value to large national (or international) digital archives, but the funding is oriented to projects and not long-term archiving. Some of the issues that came up:

  • Lou Burnard made an important distinciton between archiving and backup. A lot of people want backup for their work or their project and think that archiving services will provide this; they don’t really understand that backup is not archiving. That doesn’t mean that backup isn’t important. Apparently in the student riots in Paris last year a number of computers with irreplaceable data were destroyed.
  • The limitations of centralized solutions. We are all tempted by the thought of long-term central funding to run services, but there are dangers to such centralization. If central funding is cut or shifted (as happened with the AHDS) then everything disappears. Can we imagine decentralized solutions? Would they work? I’d like to see more social research initiatives that support decentralized solutions. I think in the current economic climate we have to explore these.
  • David Robey made the point that we have to do a better job of explaining the value of digital resources and services. We need to educate ourselves to gather evidence of value and that includes the opportunity costs.
  • Paolo D’Ivorio argued that there are certain primitive functions that scholarly systems need including Citation (reliable ways to point to other works), Consensus (agreement in a field as to what is of value and how to assess that), and Discovery/Dissemination (ways of finding and getting at scholarship.)

You can follow some of the meeting is you search Twitter for #ADONIS.