[talk-au] Fwd: [OSM-talk] That license change link

[Kai Krueger]

Richard Weait wrote:
> 
>>On Thu, Jul 8, 2010 at 1:20 PM, Liz <edodd at billiau.net> wrote:
>>> On Thu, 8 Jul 2010, Richard Weait wrote:
>>>> But to get back to the point of your question about the percentage
>>>> required for changeover.  Do you have a preference or suggestion for
>>>> percentage of contributors and or percentage of data that would be
>>>> right for the changeover?
>>>
>>> I find the notion of voting for or against a changeover without these
>>> targets
>>> being defined quite wrong.
> 

I too think answering the modalities of when sufficient support has been
gathered to proceed with the change over is incredibly important and must be
nailed down much more before the procedure begins. Otherwise, it is just
asking for trouble, bias and non rational decision making in the likely
flame war that will happen around that time of results. 


>And yet you haven't suggested what you think would be an acceptable
>target.  I don't mean to pick on you, others haven't stepped up to
>suggest the bright line limit either.  But really, if it is important
>to you to have a bright line, why not suggest one?  Now a method to
>measure it would be good too.

I won't be able to give you a good answer either. It is no doubt an
incredibly difficult thing to set those targets, but that is precisely why
it needs to be discussed and decided beforehand and not once you are biased
by the results and your strong wish to get the result you were hoping for.

I fully acknowledge the issue and the criticism against defining limits
before hand of that it is pretty much impossible at the moment to forsee the
full consequences and thus it is impossible to fully understand the
situation until the results are in and the scripts have been run to see how
things cascade through the data. With that, I can also see that there needs
to be some flexibility in judgement, but at least the ball park, or an
absolute minimum requirement needs to be set beforehand.

>Simple percentage of contributors might be too limiting as a bright
>line as contributions follow a long tail.
>See the lowest graph on http://wiki.openstreetmap.org/wiki/Stats

As an starting point for discussions, I personally would probably suggest
something like no more than 10% of data (both nodes and ways) (excluding
tiger) may be removed and no more than 20% of data may be modified / rolled
back. In addition probably something like 80% of active contributors must
agree. (I don't think we really care about the 100.000 or so registered
accounts that have not yet mapped  a single node or way) 

I don't agree with your opinion of that contributors are orders of magnitude
more important than data (in this criterion) though. Not because
contributors aren't much more important, but because they are linked. If 99%
of contributors agree, but still 90% of data gets deleted, that will be
hugely demoralising and a good fraction of those 99% who agreed will
probably stop contributing as they may think that their contributions aren't
respected or valued and might get deleted again in the future. And that
reputation can even carry on to potential new members thus being even worse
than loosing current mappers.

So let us give ourselves a clear target of what needs to be reached so that
we can then work hard towards convincing enough people that the license
change is necessary for legal reasons and that the new license is no worse
than the old, so that people can agree and not care if they don't want to.

Kai 


-- 
View this message in context: http://gis.638310.n2.nabble.com/Fwd-OSM-talk-That-license-change-link-tp5267634p5272104.html
Sent from the Australia mailing list archive at Nabble.com.