Yesterday I finally read Unlocking the Potential of Aid Information. It’s good. Really good. Some new things to think about, some re-discovering of old friends, and the feeling that there have been some very smart people thinking about this for a long time — and that perhaps they were ahead of their time. Yes, Bellanet (current incarnation), I’m talking about you. Again. Sigh. (To keep watching: CGIAR’s ICT-KM group, they do a lot Bellanet-ish type stuff too.)
As usual, I have more to say on this. Especially in relation to publishing aid information in open formats. We waste a lot of time and money on publishing research and project findings using technologies and formats that make no sense. I’ve always opposed the way publications production works from a process/resource standpoint — but now I understand it also creates accessibility issues. More ammunition!
In the meantime, here’s a bit of a roundup mixed with some to-do notes for myself:
- Read the paper. (DONE!) It’s short and well-written. Just go do it so we’re all on the same page please.
- Pay attention to the International Aid Transparency Initiative (IATA)
- Pay attention to aidinfo
- Lurk on the Open Development mailing list
- Watch the Open Knowledge Foundation — and especially their Open Development Working Group (happy to see some familiar names there!)
- Figure out how the Comprehensive Knowledge Archive Network’s (CKAN) development information repository works
- AiDA (sad link) has now morphed into AidData — check it out
- Keep tabs on progress of the International Development Markup Language (IDML) — is it dead? — leads me to lots of dead Bellanet links from its IDRC days and idmlinitiative.org is down
- Tag opendevelopment resources more consistently
I realize from delving more into this that several of us have come to the same conclusion. I’m repeating myself but here goes: It’s not about building a big repository. Stop that. It’s about aggregating, not centralizing. Making it easy to find, bring together, and mash up. And this means, among other things:
- Open — not just accessible! — content and data
- Standard licensing frameworks (legal mechanisms to share and build on others’ work)
- Open, machine-readable file formats
- More thinking about how to integrate semantic tagging into our workflows
- Standard data schemas and metadata
(Oh oops, that’s pretty much the table of contents for the paper that got this whole post going…) Along with this goes something that I have not seen much writing about yet but that I think is really crucial:
Citation practices for open datasets and instruments.
To be clear: Academics should get credit for the data and instruments that they share. It should be right up there with publishing in a kick-ass, prestigious journal. (Make that an open journal! Love to ITID and First Monday). This HAS to count for them professionally. Toward professorship, toward tenure. This gets back to the way the game is played now. Want things to be different? Look at incentives. Re-think the rules of the game.
I’ve got more info on datasets and citations… later.