[OSM-talk] Disallow Anonymous Edits NOW

[Andrew Loughhead]

Steve Coast wrote:
> On 14 Oct 2007, at 23:13, Richard Fairhurst wrote:
>
>   
>> Steve Coast wrote:
>>
>>     
>>> +1 to be fashionable
>>>
>>> It's also fashionable to give up freedom for security, which is
>>> pretty much what this is. Cutting off semi-anonymous (they still need
>>> an account) editing is just putting off attacking the real problem,
>>> which is rollback.
>>>       
>> Hm. I'm working on rollback, but it is undeniably a Hard Problem and
>> in Potlatch UI terms needs to be lumped with the even Harder Problem
>> of redesigning the tag editing panel. That's why it hasn't happened
>> in Potlatch yet, not because I'm lazy (well, not especially).
>> Frederik committing five-line changes to two or three controllers
>> doesn't really delay this work.
>>     
>
> No, but it provides the illusion that people won't break data. I can  
> still go in and delete all of Oxford, right? And in the mean time, a  
> basic freedom has been lost.
>
> I know I sound a bit Stallman, but I think it's a shame.
>   
Would a fair analogy be SVN / CVS commit access on open source / free 
software projects?  Do any such projects, even those managed by FSF, 
ever allow completely anonymous commit access?  In that context they 
actually have working rollback.  What seems to be asked for here is only 
that people are probably contactable before granting them commit access, 
which is much weaker than the usual software project model of 
participate, submit patches, gain trust, etc, and a long way short of 
demanding identity. 

Apologies in advance if I'm misunderstanding your point or the issues. 

cheers
Andrew.