[OSM-legal-talk] Making OSM Public domain

[Frederik Ramm]

Hi,

Joseph Gentle wrote:
> What sort of support from the community would be needed to public domain OSM?

My answer applies to any license change, be that ODbL or PD or whatever.

The problem with that is that *any* change of license will require a 
concerted effort by many people in the project. Whatever change we 
attempt, we will have:

* a large number of people who like it
* a large number of people who dislike it but are willing to enter into 
a compromise
* some people who dislike it so much that they will not re-license their 
data and withdraw from the project altogether
* some people who cannot be reached for comment

A very important part of the puzzle, and one that has not been talked 
about a lot until now, is how to deal with the last two groups. This, 
also, requires a broad consensus - it is well possible that there are 
some who are prepared to go along with a new license but then quit 
because they are unhappy about the transition process!

> - How many people would rather OSM was PD at this time? How do we find
> this out? Would a web-based poll be appropriate?

There are two sides to this. One, what does the current community feel 
like. This can be found out with a web-based poll but of course such a 
poll will only capture those who feel strongly about one thing or the 
other, and not those who just want to map.


The other side is when you start thinking about safeguarding the current 
data set. First you have to have a transition strategy which could by 
anywhere between:

* delete anything ever touched (or even in the vicinity) of someone who 
declines to relicense or cannot be reached; it is never too late to 
start over, and even if this requires deleting half of our data it gives 
us a clean start.

* delete only objects where someone who declines to relicense has had 
significant and nontrivial influence, and say that any minor 
contribution is a fact anyway.

 > - How many would we need to make OSM PD? 50%?

Depending on what transition strategy you choose, you can then make an 
analysis and find out how many people you need to convince to relicense 
to keep enough of the data set to make it worthwile. Of course there 
will be major contributors whom you might try to contact personally, as 
Peter Miller has suggested.

> - Would moving the data to a public domain license be any more complex
> than moving to the proposed open-database license?

Probably not (the only thing more complex would be if the user was 
offered a choice between multiple licenses; this is not unlikely to lead 
to chaos).

There is one difficulty however; the Foundation, or more precisely, the 
server administrators, are currently the only people who can actually 
contact all contributors. As a normal user, you can contact individuals 
through the web site if you know their user name, but there are some who 
have not made their user name public and you have no chance of 
contacting these. You also have no bulk history data available, so if 
your transition strategy involved anything like "find out how many ways 
were ever touched by X", then you would practically not be able to do that.

We have to be a bit careful and civilized here. There's 50.000 mappers 
out there. My guess is that at least 45.000 of them have never even 
thought about the license, and many will perhaps not grasp the full 
details of all the options. If we want these people to re-license their 
contributions, then we must contact them with a very solid and united 
proposal, something like: "The foundation recommends so-and-so, it has 
been checked by 5 lawyers and the top 20 mappers in your country have 
already signed up. Please press the button below to make sure your data 
can remain in OSM..."

If we contact them with something like "Em, er, there's a lot of 
discussion and we're divided about where to go with OSM, some want this 
license and some want another...", then there's a big danger of getting 
less participation than we need.

So, as much as I'd love to see OSM go PD, I do not think that this is an 
option at this time. I think we should really do our best to create a 
united proposal that we can present to the mappers and ask them to sign 
- with no alternatives, just sign or not sign. Otherwise the whole 
process goes haywire.

I think the ODbL/FIL is a viable compromise. I dislike the share-alike 
stuff, but I like the acknowledgement that facts are basically free; 
that's a giant leap ahead from where we are now, claiming that there is 
some sort of intellectual property on the way you put a node in the 
database. Others will like the share-alike stuff and be a bit uneasy 
about a perceived loosening of the grip on factual information. But I 
think most of us, on both sides of the argument, are willing to go along 
with the ODbL/FIL, and that's our best chance at making a license change.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frederik at remote.org  ##  N49°00'09" E008°23'33"