[OSM-legal-talk] To calm some waters - about Section 3

[Frederik Ramm]

Hi,

Simon Biber wrote:
> I and many others need a firm commitment to ensure contributions continue to be 
> protected by attribution and share-alike in the future.

-1

(I mean, you may "need" that but you shouldn't get it. As an aside I 
also want to point out that the use of "continue to be protected" in 
your sentence does not fit with current wisdom about CC-BY-SA and our data.)

I am against trying to force our will on "OSM in 10 years". OSM in ten 
years will have a larger community and a larger data volume by orders of 
magnitude. I don't think it is right to force their hand in any way over 
and above the necessary minimum just because a few of us think so.

What exactly the necessary minimum is, is subject to discussion; I could 
imagine that the necessary minimum perhaps includes that we fix an 
attribution requirement, but a share-alike requirement would certainly 
be going too far.

It is bad enough if the share-alike minority force their will on the 
rest of the project now; we must not allow them to force their will on 
everybody who is in OSM in 10 years' time.

Oops. That wasn't exactly calming the waters, was it. But it needs to be 
said.

There is also a very practical reason against fixing anything, and 
*specifically* a share-alike requirement, in the CT, and that is that in 
order to make *clear* what you want you will have to write half a 
license into the CT.

Imagine that we put the phrase "a free and open license with 
attribution and share-alike" into the CT. Imagine further that, at some 
point in the future, a change to ODbL 1.1 is debated, and that ODbL 1.1 
only had minor changes over ODbL 1.0.

Then someone comes along and says: "Sorry guys, the CT say that the new 
license must be share-alike. But ODbL is not properly share-alike, see, 
it allows non-share-alike produced works, and it allows non-share-alike 
extracts if they are not substantial!"

Bummer. At that time, we'll have one hell of a discussion about what 
exactly qualifies as a share-alike license and whether ODbL 1.1 is 
covered by the CT.

To avoid that, you would have to write into the CT exactly what you mean 
by share-alike. By doing so, the CT would become much longer and more 
complex, and drastically reduce the choice of license in the future even 
within the pool of share-alike licenses. Inevitably, we would write what 
we *today* think is right into the CT - but the whole point of allowing 
future OSM communities to choose their license is that they may adapt.

Trying to force their hand - when their contributions will vastly 
outnumber ours, and they will be 10 or hundred times more than we are 
now, would be overbearing. I don't think it would be morally right. The 
amount of data we have collected and the amount of time we have invested 
will, in 10 years' time, be minuscle compared to what the project is 
then, and using that contribution to justify wanting to have a say in 
OSM for all time is just greedy.

I am aware that this is a moral statement and that it will be required 
to do slightly less than what is morally right, for practical reasons. 
And that's ok; we're all pragmatic.

Bye
Frederik