[Osmf-talk] Blanket copyright licence in Contributor Terms

80n 80n80n at gmail.com
Sun Dec 13 15:14:02 UTC 2009


On Sun, Dec 13, 2009 at 2:28 PM, Matt Amos <matt at asklater.com> wrote:

> Gervase Markham wrote:
> > On 12/12/09 16:33, Matt Amos wrote:
> >> and also that the Contents are effectively PD, but anyone downloading
> >> planet.osm is effectively downloading a database. so each individual
> >> Content item has a very liberal license, but when used in Substantial
> >> amounts is governed by the ODbL.
> >
> > So if OSM accepts a contribution from a 3rd party "under the ODbL", what
> > is the licence of the individual data items? Anything they like? Even
> > something restrictive? Are we back to the position where we effectively
> > have to get the original owner of the database to sign the Contributor
> > Terms in order to take their "ODbLed" contribution?
>
> yes, the contributor terms means that OSM wouldn't be able to accept
> such an ODbL contribution. imagining that we were accepting
> contributions under the ODbL, then it would be another point of
> difficulty that the Contents licenses would have to be compatible.
>
> > One of the big points of share-alike licences is that you can, without
> > asking permission, incorporate other content under the same licence into
> > your content. The more we look, the more we seem to be discovering ways
> > that this isn't in fact true of the currently-proposed licensing scheme.
> > And if it's not, then it's not a suitable licence to switch to if
> > "preserve the share-alike" is one of the design goals - as we are told
> > it is.
>
> there's really two parts to share-alike: 1) you can import other
> people's data, and 2) other people can import your data. (1) isn't true
> of the proposed scheme. basically, it's not possible to have both the
> ability to react in the future to the need to change the license and the
> ability to accept ODbL + permissive Contents contributions. as i've said
> several times; my opinion is that the former is more important. your
> opinion is clearly for the latter.
>
> we're not setting a precedent for this; FSF is already unable to take
> GPL-only contributions, since they require rights assignment. we aren't
> proposing rights assignment, but the effect on imports is the same.
>
> > I'm entirely in favour of switching to a licence which is a) like
> > CC-BY-SA in share-alike, and b) is better suited for data. The LWG has
> > made a good case that the ODbL is b), but it seems increasingly
> > uncertain that it's actually a).
>
> well, we've got half of (a); it's share-alike downstream, just not
> upstream ;-)
>

Well, it's not really share-alike downstream either, is it.

If someone takes an ODbL licensed planet.osm and adds their own data to it
then there are two reasons why we may not be able to have these
modifications back:
1) They can impose some incompatible license conditions on their added
content.
2) They refuse to agree to the Contributor Terms

What attributes of share-alike are successfully implemented by ODbL+CT?  Is
it really a proper share-alike license in practice?

80n
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20091213/32bc18db/attachment.html>


More information about the osmf-talk mailing list