[odc-discuss] Open Database Licence

Rufus Pollock rufus.pollock at okfn.org
Wed Mar 4 18:35:33 GMT 2009

2009/3/4 Jonathan Rochkind <rochkind at jhu.edu>:
> This is interesting and odd to me, because my understanding was that
> providing a share-alike license was pretty much unworkable for data -- not
> feasible to make a legally enforceable contract to these ends accross
> jurisdictions. Which is what led to the PDDL/CC0/Science Data Commons
> approach.

I don't think anyone ever thought this was 'unworkable' for data: the
original license Jordan wrote (out of which came the PDDL) was an SA
type license.

There is a debate as to whether SA is a good idea (John Wilbanks of SC
is particularly critical) and in some jurisdictions is debatable
exactly whether (and what) 'DB rights' exist on which to base a SA
license. I wrote a blog post on this recently that goes over much of


This has a bunch of links to more info at the bottom. If you are
interested in the question of 'data rights' I'd point to our guide on
this matter:


There's also an appendix specifically commenting on Science Common's
"Protocol on Implementing Open Access to Data":


> But apparently it is legally feasible in some cases?  Any background
> material I should read explaining how ODbL manages to work at all?

See previous.

> I'm also not certain what the effective difference between the Factual
> Information License and the PDDL is. "MIT style" generally means that that
> the user is free to do pretty much whatever they want with it. So the user
> might have the same rights with the PDDL or the FIL, but the FIL doesn't
> actually put anything into the public domain?  I'll have to read the FIL
> carefully; again, my understanding was that it was legally very
> complicated/infeasible to do this with factual information _without_ just
> putting it in the public domain.

I agree here. I think the FIL and the ODbL were the original licenses
drafted by Jordan and formed a complementary package. With the
development of the PDDL I think the FIL is perhaps no longer needed
and can be deprecated in favour of the PDDL. However, Jordan would
need to comment on this.


More information about the odc-discuss mailing list