[Openstreetmap] Which CC Licence for Data-Sets?
matt at matt-amos.uklinux.net
Mon Feb 28 21:55:06 GMT 2005
On Wednesday 23 February 2005 17:47, Saul Albert wrote:
> > Sure, it might be fun to make a license text, but it isn't fun to
> > be on the receiving end when there are hundreds of different
> > licenses to handle.
> Ok but I was just being silly about it being fun. The main point is
> to articulate what we want to happen to our data - and there
> doesn't seem to be a licence specific enough to do that.
the problem is that we (all) want to balance the openness of the data
with the necessity to make sure that the data is not exploited for
(too much) commercial gain.
> The main reason to articulate it is to solidify our own ideas about
> what we're doing.
i think thats the reason we're having this discussion ;-)
> I think there's a confusion here between openstreetmap/free map
> etc. as a *project* and the idea of open geodata as a framework.
> As a project, we adopet a licence, but surely the intention behind
> making a new licence is that it can be adopted by other projects.
i'm sure thats why RMS wrote the GPL and the CC guys wrote their
licenses, the BSD licenses, MIT/X11 licenses, and so on ad nauseam.
> By writing a licence we're talking about this as a framework -
> which has technical and legal and social aspects - just like the
> picopeering agreement is a document that articulates the principles
> of free networks, that many projects then adopt and adapt according
> to need - but maintaining the core principles that make the network
i don't really understand this one... we have some data: gps, aerial
photos, etc... which are contributed and can be licensed however the
contributor wants. if the contributor license is incompatible with
ours then we'll just have to find alternative sources of data.
then we have some more data: the roads and other features extracted
from the raw data. since this is also just data - "source code" if
you like - then we adopt a good license for this, e.g: CC-SA, GPL.
we (might) have some programs to "compile" the source code into images
or PDFs, but these can be GPL or whatever.
the only trouble comes if we want to restrict the use of the "binary"
images. i think that we're covered under the "share-alike" part of
the CC-SA (as the images would explicitly be considered derivative
works), so we wouldn't need an additional license.
> So I really think the case for making a licence is strong - because
> it opens this out from being a project to being something that
> other projects, commercial agencies and governments can sign up to
> and use to licence their geodata.
so my question is: why do we need another license? if we're clear
about what we mean, and we are sure that the CC-SA license covers all
the bases then all we're doing is confusing potential contributors
> I am also hoping we can come up with something really clear and
> simple - like the picopeering agreement, or the BSD-type licence,
> or the MIT licence or whatever, that fits into 1 screen, can be
> read in under a minute, and doesn't decend into the idiotic
most licenses can be summed up in a few sentences at most. the
"idiotic legalese" is there to try and bomb-proof the inherent common
sense that everyone takes for granted, but lawyers like to argue
some points worth considering are:
1) using a well-known, established license like GPL, CC, BSD, MIT/X11
and so on means that most people won't even read the license -
they'll accept it because many other people have seen it and not
complained. also, companies with legal departments can rest easily,
as they may already have examined the license. the converse is true
of any "custom" license.
2) we do need to firm up the precise situations where and when we
consider redistribution to be compliant with the license we adopt and
when we dont. a simple web page with some examples should be
3) what if we draw up a license without legalese and without
consulting an international IP lawyer... and it gets shot down? if we
change the license later that involves contacting all the
contributors and asking them to accept a change. even worse - the
data might effectively become public domain. i know that not even the
GPL has really been tested, but a number of very prominent lawyers
have read it and said its OK...
just some (rather long, sorry) thoughts,
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Size: 190 bytes
Desc: not available
More information about the talk