[cc-community] Computer-generated derivative

Gisle Hannemyr gisle at ifi.uio.no
Mon Mar 9 11:30:19 EDT 2009

On 09.03.2009 13:04, Richard Fairhurst wrote:
> We're having the usual lively discussion over on the OpenStreetMap  
> mailing lists and it was suggested (thanks Rob) that it might be  
> worth asking on the CC lists too.
> If Fred has a program running on his computer that downloads  
> OpenStreetMap data (licensed CC-BY-SA),

While CC-BY-SA 2.0 is the license you'll find  attached to the
OpenStreetMap data, "data" is not protected by CC-BY-SA 2.0.
The license defines "Work" as "copyrightable work of authorship",
and "data" falls outside this definition.

Also, "data" (and other facts) are not protected by copyright.
Slapping a copyright notice and a license on something that can't
be copyrighted has no legal relevance.

> then combines it with some
> proprietary, non-CC-BY-SA stuff, that's perfectly ok as long as Fred  
> doesn't then distribute the result. In fact, Fred isn't actually  
> _allowed_ to distribute the result.

In jurisdictions that does not protect sui generis rights (i.e.
in most of the world, including the US of A), Fred can do whatever
he wants to do with any data he can access, since data are
not protected by copyright.  As long as data are unencrypted,
anyone can use them for whatever purpose they like, and distribute
them in whatever way he want.

In Europe, where sui generis rights exists, Friedrich can do whatever
he want with a subset of the data (which will be adequate for
"derived" purpose, including the purpose described on the OSM mailing
list).  Database rights in the EEA/EU only apply if you copy
a substantial amount of the data, not if you utilise a subset and
add significant value.

> And therefore, I presume the same is true if the program is a Flash  
> app (running client-side, of course, albeit with a browser frame  
> around it) which outputs the result as a PDF - which Fred can then  
> save to his local hard drive and/or print.
> Am I right?

No - the premise of your whole discussion is wrong, because you
assume that the CC-BY-SA 2.0 license applies to OpenStreeMap data.
It doesn't.

> (See also http://lists.openstreetmap.org/pipermail/legal-talk/2009- 
> March/002215.html , in particular the second reply, and http:// 
> lists.openstreetmap.org/pipermail/legal-talk/2008-October/001654.html )

I assume that the "proposed new license" Frederik Ramm refers to is
the draft ODbL ( http://www.opendatacommons.org/licenses/odbl/ )
that has been created as an attempt to address the fact that it
is well known that CC-BY-SA 2.0 does not apply to data?

If so, and if the purpose of the ODbL is to enforce the SA-requirement,
I don't think it will be very effective for OpenStreetMap data. It will
only  be valid in jurisdictions that grant  sui generis rights, and
even there is will be very easy to  circumvent by subsetting (i.e. not
using a substantial part of the data), by changing schema, and by
adding value.
- gisle hannemyr [ gisle{at}hannemyr.no - http://folk.uio.no/gisle/ ]
    "Don't follow leaders // Watch the parkin' meters" - Bob Dylan

More information about the cc-community mailing list