[OSM-legal-talk] [OSM-talk] ODbL vs CC-by-SA pros and cons

[Francis Davey]

On 2 September 2010 02:25, Anthony <osm at inbox.org> wrote:
> On Wed, Sep 1, 2010 at 6:04 PM, Francis Davey <fjmd1a at gmail.com> wrote:
>> "maps" are expressly treated as "artistic works" by s.4(2)(a) of the
>> Copyright Designs and Patents Act 1988 (to give a UK perspective).
>
> Pretty much the same thing in the US.  "pictorial, graphic, and
> sculptural works" are included as examples of copyrightable works, and
> "maps" are included under "pictorial, graphic, and sculptural works".

Yes - I didn't want to bore 8-), though there's a subtle difference in
that the US statute originates in a constitutional provision
permitting Congress to promote the progress of science, whereas the UK
Parliament can pass whatever it likes (as also in Australia). I don't
think that's relevant in this case, but it means that it is possible
to inject more policy into a USian debate.

Even in French law - where the fundamental object of protection is a
"work of the mind", the Code expressly includes "geographical maps" as
"works of the mind" (L112-2).

>
> Well, not really.  First of all, I'd say Mapnik tiles are clearly part
> of OSM, and I don't think there's any dispute that Mapnik tiles are
> maps.  But furthermore, when it comes to the OSM database itself, I
> agree with Assistant County Attorney Lori Peterson Dando that "a GIS
> database [is] essentially a computerized map" and "may be entitled to
> protection under copyright law, not only as a compilation, but as a
> 'pictorial' or 'graphic' work as well" (see Open Records Law, GIS, and
> Copyright Protection:  Life after Feist,
> https://www.urisa.org/files/Dandovol4no1-4.pdf).
>

I'm inclined to agree, at least for the UK. Dando's analysis of course
doesn't follow through (because we have no Feist) but I think UK
authority bears a similar conclusion. I'd guess Australia was the
same, but I don't have the same thorough knowledge of it.

>
> Well, in this case we were talking about the definition as used in CC-BY-SA 3.0.
>
> I'd certainly argue that "maps", as used in that license, include GIS
> databases like the OSM database, and I'd use Ms. Peterson Dando's
> comment that a GIS database is "essentially a computerized map" as
> evidence.  Ultimately, if it became a matter of dispute, and judge
> and/or jury would decide, and we can only make educated guesses about
> whether or not they'd agree.

Oh, that seems highly likely. The problem with CC-BY-SA 3.0 is not
whether a "Work" can include a map, nor even (in our view it seems)
whether "map" in the licence include the OSM database, but whether or
not CC-BY-SA 3.0 extends to works that are the subject of the sui
generis right or not. It is not clear whether "other applicable laws"
(in clause 2 say) would or would not include it, or even whether
"copyright" would be construed to include protections like copyright.
I think the express qualification of rights over database in the
definition of "Work" would suggest not.

If there's no copyright in the applicable law and CC-BY-SA 3.0 only
covers copyright it doesn't matter whether GIS databases are included
in maps as a matter of construction of "Work" since the licence
wouldn't reach so far.

But maybe you meant to imply all that and I wasn't reading carefully
enough. If so, sorry.

>
> On the other hand, it might not matter, as I'd also argue that the OSM
> database is a copyrightable compilation.  As to that, Ms. Peterson
> Dando says "in the context of copyright law, GIS databases are
> compilations which may be copyrighted".

That's something that is likely to vary more across the world I'm
afraid. In particular some GIS databases might not get over the "own
intellectual creation" hurdle.

>
> Finally, I want to be fair and point out that while (or even if) the
> OSM database is copyrightable, that doesn't mean the copyright on it
> extends very far.  Again quoting Lori Peterson Dando, "Even though a
> GIS database may be copyrightable as a compilation or a map, the
> protection afforded by copyright may be thin in light of the Feist and
> Mason decisions."

Right. That's even more difficult because, as you know, the approach
taken around the world to the way in which one assesses infringement
is complicated. When US lawyers talk about "thin" protection they
don't quite mean the same thing as we do and so on.

>
> To give a specific example, I'd say a routing database created from
> OSM data, suitable for running a shortest path algorithm and providing
> driving directions, would be completely public domain and
> non-copyrightable, in the US and in many other jurisdictions.

Right, because it takes only the factual content and not the work.
That seems plausible to me in so far as I understand the US
authorities.

>
> And that, I'd say, is a flaw in CC-BY-SA.  Because it means someone in
> a sui generis database rights jurisdiction could take OSM, make a
> routing database out of it, improve that routing database, and then
> sue people under database rights law for using those improvements.  At
> least, under CC-BY-SA 3.0, I think they could.

Yes, I think that is exactly right.

-- 
Francis Davey