[OSM-legal-talk] feedback requested

[andrzej zaborowski]

On 27 December 2011 15:31, Frederik Ramm <frederik at remote.org> wrote:
> On 12/27/11 14:53, andrzej zaborowski wrote:
>>>
>>> * treat any tags contributed by a non-agreeing mapper as harmless if
>>>  these tags are not present any more in the current version
>>
>>
>> Did you manage to address your example of a user fixing a typo in the
>> tag name (individually or for a large number of objects)?
>
>
> No. It would be possible to say that a constribution is only harmless if
> neither the tag nor the value are present in the final version but then
> there will again be examples where this is wrong.
>
> I think a good way to deal with such "but what if..." situations is not to
> make sure they never occur, but to produce some kind of quantitative
> assessment. If we have reason to believe that the new rules produce
> something like a hundred errors then who cares. If it's more like a million
> then it needs to be fixed ;)

Right, I asked because I think this case is quite frequent.  Some
changes, like natural=wood -> landuse=forest, shop=dentist ->
amenity=dentist, and the other way are often done massively by people
(not that they're very useful..).  It's possible that in absolute
numbers they're more frequent than genuine changes to the map.

>
>
>> A similar case is where a mapper adds many ways in a city, but another
>> mapper thinks all of the objects were misaligned and offsets them en
>> masse.  Is your assumption here also that this takes all IP away from
>> the first mapper?
>
>
> Currently in OSMI, this mapper will continue to be viewed as the author of
> all the ways; it is just the node positions that he got wrong and where his
> contribution is overwritten by the change.

If we talk about individual node positions then he got them wrong, but
if we talk about "geometries" then the later change is a detail.

But actually this case is probably quite rare so your earlier argument applies.

Cheers