[OSM-legal-talk] Circumnavigating Share-Alike through software / now and future

[Rob Myers]

On Mon, Oct 27, 2008 at 6:00 PM, Frederik Ramm <frederik at remote.org> wrote:

> Obviously. But I was not talking about restricting the rights of the OSM
> data distributed alongside the proprietary data; the super secret data
> set company would of course allow you to do anything with the OSM
> extract they give you. They would just restrict their own data. Or, in
> CC-BY-SA license terms, they would give you a collective work (a CD-ROM,
> say), that has OSM data and proprietary data sitting alongside each other.

If it's a way of trying to make a non-redistributable derivative at
arms length, I think it's very difficult to get around BY-SA.

Of course the company could just claim fair use. Google seem to get
away with that while making lots of money. ;-)

> Ok, we can make that change; it would probably not be too much to ask
> from the super secret data set company to build some automatic OSM
> downloader into their software.

I imagine that would be similar to Napster.

> That's an interesting point we are getting to here - the distinction
> between the end user creating a derivative of his own volition versus
> the end user being sort-of "guided" (or even tricked?) into making a
> derivative.

Yes this is the key distinction.

It absolutely should be possible for you to be able to make a mash up
of OSM data and whatever else in the privacy of your own home. But it
should also not be easy for a third party to be able to force you to
make a derivative of a BY-SA that removes the rights that BY-SA is
meant to ensure are inalienable.

> Probably extremely hard to find sharp wording for. "If you
> create a software with the primary aim of making derived works from OSM
> and another data source, you must make sure that the other data source
> is licensed such that the derived work can be distributed..." - but that
> would go much too far, it would for example make it impossible for
> anyone to write a generic GDAL OSM importer or an OSM importer for
> Google Earth!

Yes I see your point.

> In the end this is nothing else than what we're doing with the
> "layering" stuff right now. If someone says they want to display
> proprietary data alongside OSM, our answer is they can use something
> like an OpenLayers interface which we count as a "collected" work where
> each layer may be licensed differently. However, the end product on the
> user's screen is surely a derived work (imagine him making a screenshot
> and publishing it!). This, too, is a case where the creation of the
> derivative is pushed down to the end user: He may view both data sources
> on top of each other, but he may not re-publish what he sees.

The layers compromise is a consensual haullucination of OSM. ;-)

The scenario that you describe looks similar to the "distributing
proprietary object files for end users to link with GPL software"
exploit that I think was explicitly tackled in GPL 3. This scenario
should have been known to CC when they were drafting their licences,
and I'd hope they addressed it.

- Rob.