[OSM-talk] Import guidelines & OSMF/DWG governance

[Martin Koppenhoefer]

2012/9/18 Pieren <pieren3 at gmail.com>:
> On Tue, Sep 18, 2012 at 12:13 PM, Martin Koppenhoefer
> <dieterdreist at gmail.com> wrote:
>> not sure about this, but I definitely support the decision, because it
>> was a real problem in the past when imports could not easily be
>> distinguished from individual and original contributions.
>
> Excepted that in the mentionned case, the French cadastre building
> footprints import is localized (scale is a municipality, a town or a
> village) and the features are limited to buldings and possibly
> waterways. All imported elements are also sourced and uploads are
> limited to one or few changesets. So the problem to distinguish
> individual and original contribution does not exist here.


How does it help for distinguishing imports from original
contributions to have many small areas or small feature sets or many
small changesets? What might help is a uniform changeset comment or
component.



> We cannot ask each
> contributor to create a special account each time he is importing
> something into OSM which is not coming from Bing or its GPS.


nobody should be importing from Bing or converted GPS-traces ;-), the
distinction we do is: use different accounts for data you create
yourself and for data that you take from other sources (i.e. for which
you don't have the intellectual property rights).


> And if
> creating a new user account would be that easy, but it requires a
> special, different email account each time for each new account


well, not sure where this comes from and if it makes sense: I don't
see a real obstacle as email addresses are not a scarse ressource (you
get as many as you like for free), but I agree that it seems to be
better to allow the same email address for multiple accounts (would
make it more probable that someone is monitoring the inbox = more
likely you will be able to communicate with the mapper if he can use
his usual email address).


> What I would like to know here is if the DWG is allowed to block one
> contributor just because he is not following one of the requirements
> writen on the wiki guidelines, a requirement which was just an option
> few months ago


I'd put it like this: someone who didn't respect the import guidelines
valid for  almost one year was temporarily blocked. What's the
problem? That's what the DWG is for.


. The DWG is claiming that the import guideline is
> writen by the community. But how many people have been involved in the
> discussion deciding to change the wiki and make a separate account a
> "must" instead of a recommendation ?


how many have spoken up against it? I'd expect from every mapper who
wants to import something to read the current import guidelines and to
act accordingly.


> I agree with the concept of seperate accounts but only for large
> imports done by a single person in a short time. All the opposite of
> the small French cadastre imports done by the crowd since years on
> limited areas.


I agree that it seems not necessary in the case where the data comes
with no obligations (PD/CC0) and the mappers check manually every
single object they import.


> For all of these reasons, I would like to modify the import guidelines
> and make the separate account back to a recommendation which is not
> alsways necessary, especially in case of limited imports, in size
> and/or features.


-1 if there are other obligations (like attribution) associated with
the originals data license.

cheers,
Martin