[OSM-legal-talk] Contact And Remap Campaign

[Michael Collinson]

On 13/02/2012 12:53, Simon Poole wrote:
> Am 13.02.2012 12:33, schrieb Frederik Ramm:
>>
>> This can be read - as Simon seems to do it - to mean "the CTs 
>> guarantee that required attribution will survive any future licence 
>> changes", but I think he's on thin ice there; in my reading, the CTs 
>> promise that OSMF will provide attribution, not that OSMF will only 
>> ever release your data under licenses that guarantee attribution down 
>> the line.
> My statement should naturally be read in the context of the statement 
> below: if we distribute your data, the attribution via website (and 
> further schemes that are being developed) will remain intact.
>>
>> But Simon is right when he says "data with such requirements would 
>> have to be removed". This means that if we ever wanted to go PD, then 
>> we'd have to find out which data has some kind of attribution 
>> requirement attached, and remove that data before we go PD. Since we 
>> don't require such data to be identified at the moment, that would be 
>> one hell of a job.
>>
>
>> In my eyes, this is a very sad development that undermines any future 
>> license change, even one to a non-PD license. Earlier versions of the 
>> CT basically required you to *only* contribute data of which you 
>> could surely say that it could be relicensed freely under the 
>> provisions of "free and open" and "2/3 of mappers agree". This as 
>> been whittled down to "you can contribute anything that is compatible 
>> with the current license and you don't even have to *tell* us what 
>> further restrictions it is under". Any future license change has 
>> therefore become very unlikely - except maybe a switch back to a CC 
>> license -, and not much remains of the license change provision in 
>> the CTs.
>
> While I've expressed my displeasure with every revision of the CTs 
> after 1.0 for exactly your reasoning, I don't believe that the 
> situation is quite as bad as you paint it. Come April the 1st the only 
> extra "string attached" to data that is in the database should be 
> attribution via the Website. Which implies that further data removal 
> would only be necessary if we wanted to use a distribution license 
> that didn't require any attribution at all, which is extremely 
> unlikely (not the least because of the necessary data removal).

And not even that. Using a distribution license without any attribution 
requirement and the OSMF obligation to provide first level attribution 
is compatible and works together.

*If* non-share-alike licenses become the future norm for open sharing of 
highly factual datasets, then I believe european licenses like the 
generic UK Open Goverment License [1] will become the template.  These 
require first level attribution only ... which is the primary reason for 
the CTs attribution clause. If anyone is interested on what I mean by 
first level attribution, I have a draft paper here 
http://docs.google.com/View?id=dd9g3qjp_103fdxjk3qt

Mike

[1] http://www.nationalarchives.gov.uk/doc/open-government-licence/