[OSM-talk] Attribution guideline update

[Christoph Hormann]

On Wednesday 19 February 2020, Simon Poole wrote:
>
> The updated document can be found here
> https://wiki.openstreetmap.org/wiki/Draft_Attribution_Guideline

I appreciate the draft document being available on the wiki - although 
the lack of an edit history makes this fairly pointless for the purpose 
of analyzing what has changed compared to previous drafts.

Anyway - while i am not surprised about this it is sobering how little 
of the feedback provided in previous conversation - in particular from:

https://lists.openstreetmap.org/pipermail/talk/2019-August/thread.html#83068

has found a substantial reflection in the document.  I am therfore 
reluctant to newly review the document in detail because it seems a 
waste of effort.  The process of designing the document appears to be 
dominated by a negotiation of lobbying interests rather than by 
arguments and reasoning.  I have no interest in attempting to compete 
in a lobbying campaign against corporate stakeholders who could easily 
mobilize an army of lobbyists to fight any voice of reason that stands 
in the way of their interests.

Apart from the various specific issues and self contradictions in the 
document the overall most questionable thing is that it claims to 
present recommendations how to be on the safe side w.r.t. attribution 
of use of OSM data - yet the document then almost exclusively presents 
supposed exceptions from the attribution requirement of the ODbL.  This 
is a fundamental design flaw of the document.

I won't go into details again about why it is a bad idea to from 
official OSMF side poke holes into the ODbL and what the specific 
issues are with the specific holes created.    I already discussed this 
at length in August/September.  The funny thing is that where 
apparently things have been re-formulated in an attempt to dodge 
previous critique the new formulations are often worse than before. For 
example one criterion for visible attribution being required under the 
guideline is now if OSM is the "most significant data source".  That is 
ridiculous.  I can design any map in a way that i can argue OSM is not 
the most significant source.  It would without doubt lead to many 
corporate data users *removing* existing OSM attribution from maps 
where they so far did not dare to do so.  But that is just one of many 
specific issues with the various attribution exceptions being claimed.

Not to mention the most blatant attempts at sneaking corporate wishlist 
items into the guideline are all still there - like the 10000 m^2 map 
area limit that has been conjured out of thin air apparently and the 
section on machine learning models which is completely out of place in 
an attribution guideline and which indicates that some corporate data 
user wants this kind of "blank check" really badly.  I can guarantee 
that should the OSMF adopt this it is going to blow up in their face. 
Trying to sneak either of these into a guideline on attribution is just 
reckless and in complete disregard of the reason why the ODbL has a 
share-alike provision.

If anyone wants to make a serious attempt at actually writing a guidance 
document how to practically design attribution in a way that is in line 
with the mapper community consensus interpretation of the ODbL and its 
intention as well as its function in the social contract between 
mappers and data users i would gladly help with suggestions and 
feedback.  But what is being presented here is something very 
different - it is essentially the attempt of slighting of the 
attribution requirement to an 'if you want you can visible attribute 
OSM when using its data but if you want to avoid that here are some 
hints how you can bury the mentioning of using OSM data where hardly 
anyone will see it' and we, the OSMF, promise to look the other way.  
Or in other words:  It is the preemptive surrender of the OSMF in front 
of massive corporate interests.

-- 
Christoph Hormann
http://www.imagico.de/