[Osmf-talk] attribution: board letter to facebook

Christoph Hormann chris_hormann at gmx.de
Fri Oct 11 10:27:22 UTC 2019

On Friday 11 October 2019, Frederik Ramm wrote:
> > Does this mean the LWG and the board consider interpreting the ODbL
> > in a way that would declare what facebook does to be compliant with
> > the license?
> Simon has posted a draft of what LWG are working on two months ago:

Yes, i know - and i have commented on it quite extensively.  But so far
there has been no substantial communication back from either the board
or the LWG regarding the critizism articulated on open channels by the
community.  Therefore i am gauging statements from anyone regarding
indications in what direction the journey might go.

In fact regarding the attribution community guideline i don't really
understand why we cannot simply have *all* discussion and deliberation
on this matter happen on open channels (meaning that anyone can listen,
not necessarily that anyone can speak of course).  This would make it
much more clear what the LWG and board are currently considering in
that regard.

> I think that, even if it is already clear that Facebook's current
> practice violates attribution requirements in some cases, it still
> makes sense to postpone any enforcement action until the re-worked
> guidelines are done because what we often hear in response to
> attribution requests is whining about "the requirements being not
> clear", and the new guidelines will hopefully make the requirements
> crystal clear so we can then say: "Maybe in the past, but here's the
> easy-to-understand new rules that tell you exactly what you have to
> do so please don't give us any of that everything-is-so-unclear
> anymore, thanks."

Frankly i think these 'the requirements are not clear' complaints often
seem like a chiffre for 'the requirements are inconvenient'.  If you
don't try to squeeze out the last bit of profit from a use case by
reducing the user attention to OSM as a data source to the absolute
minimum you can marginally get away with, correct attribution is not a
problem and what the OSM community asks for is quite clear and
documented in


If anyone disagrees - please show me a practical use case where it is
not clear how you can attribute OSM in a way that clearly satisfies the
ODbL requirements.

As i said elsewhere i am not opposed to clarifying this further, i am
opposed to clarifying in a non-conservative way that does not strictly
try to err on the side of caution.

Christoph Hormann

More information about the osmf-talk mailing list