[OSM-legal-talk] Is the "data share-alike" road navigable?
80n80n at gmail.com
Thu Mar 8 12:59:35 GMT 2007
On 3/8/07, Tom Chance <tom at acrewoods.net> wrote:
> Apologies for the awful pun.
> Some people seem in favour of the following arrangement:
> 1. Database of GPS traces - public domain, possibly with a license to get around EU database rights to achieve the same effect
> 2. Database of nodes, etc. - share-alike licensed
> 3. Maps produced from data - up to the creator (proprietary, share-alike, whatever)
> Is this even possible? Let's get a legal opinion on the connection between 2 & 3 before we let this debate rage on much longer, because several people seriously doubt that you could get away with 3, since the maps would be derivative works of the data.
> If 3 is possible, then we can debate whether or not share-alike should cover maps produced and hosted by OSM, and cottage cartographers needn't worry about coming up with an innovative model since they could get away with keeping full copyright on their maps regardless.
> If it isn't, then we'd have to dual license the data to allow proprietary maps, and frankly that seems like a very counterproductive exception since nobody will be compelled to respect the share-alike clause. It would mean the "data share alike" road is just not practical.
> Either way, there's another quandary to explore relating to 2. Let's say I download the OSM data for St Albans, fiddle on my computer and produce a map with all of the bins and trees (hey, now I'm an eccentric). I distribute a book of these maps. Let's say we've discovered that 3 is impossible, and so my book is under the CC share-alike license too. Great, people can now share and build upon my book. But they don't get at the data because CC BY-SA doesn't require the release of source materials a-la GPL, so unless I voluntarily do so, or foolishly distribute the data as well as the book, the data will never go back into the OSM database.
However, if they do publish the data as a rendered map, then anybody
has the *right*  to trace that data and put it into the OSM
database. Faced with that reality any reasonable eccentric would see
that witholding the source data is futile .
An even more eccentric person might add all the bins and trees in St
Albans and then *not* publish what they've done at all. They don't
have an obligation to make the data available and their data will
never go back into the OSM database either.
 Whether or not we want litter bins and trees in the database is,
of course, another matter altogether.
 There would be a qualitative degradation in this case. Consider
for example, a low resolution rendered map that shows the approximate
location of speed cameras. This could be an advertisement for a
paid-for database containing the exact location speed cameras. The
publisher might be happy for coarse data to be reverse engineered and
placed in the OSM database, but understandably not willing to share
the original source data. I'm not sure that I see any problem with
> My point is - are we even really achieving outcome 2 with the CC BY-SA license? Surely the intention of the "data share-alike" road is to force people to contribute any data for nodes, ways, etc. back to the database?
> legal-talk mailing list
> legal-talk at openstreetmap.org
More information about the legal-talk