Maintain data sets as CSV files in a repository

[brockfanning]

I'm just ramping up on JKAN, and from what I've seen the general use-case is maintaining metadata about data sets. Has anyone tried using JKAN to maintain data sets themselves, as CSV files in a repository? I'm thinking of functionality for editing tabular data in a user-friendly way, right in the JKAN interface. Is this possible/feasible in JKAN or is it out of scope?

[timwis]

Hi @brockfanning! You may want to check out octopub, mentioned the other day by @OxOpenData in #140. Other related resources:

• prose.io, which has a csv editor
• okfn/dataexplorer
• A pipe dream I had

There's some issues that git has with very large datasets, and there's probably a good case to keep your data in a repository outside of JKAN if you do go that route, but if you're just working with relatively small datasets or ones that don't update very often, I don't see why not. But I think folks at OKFN have explored this idea a lot more thoroughly.

[brockfanning]

Great info, thanks very much @timwis!