[Rebuild] Is an object created by a non-agreer always tainted, even if all original info has been deleted/changed by agreers?

[Woll Newall]

Sorry for the duplication, but I started this thread on the Legal-talk  
list, and didn't get many responses. http://lists.openstreetmap.org/pipermail/legal-talk/2012-February/006916.html

Anyone like to respond here? It would be good to get some consensus,  
so such cases could be recognised as 'clean' by the licence change  
tools.

---------------
What is the consensus on the legal status of an object that has been
created by a non-agreer, but all of the nodes and all of the tags have
been deleted/changed by agreers?

i.e.:
1) Non-agreer creates a way with tags 'name=A' and 'highway=tertiary',
and 3 nodes (with no tags).
2) An agreer then deletes 1 node and moves the other 2 nodes, and
changes the tags to be 'name=B' and 'highway=secondary'.

Is that way now clean or still tainted? I think that it is clean.

My thoughts:
Tags
I think that the changing of those tags makes the tags clean. The
existence of the 'name' and 'highway' tags (separate from the actual
tag values) is not "creative". The values of the tags could be
"creative", but the values have been changed.
I guess that it could be argued that the existence of some other
obscure tags might be "creative" enough. Can anyone think of any tags
whose mere existence on an object  (but with a different value)
carries enough "ownership" to make the way tainted?

Nodes
I think that the nodes are also clean.
However, I think that there was a discussion about this a while ago,
where someone argued that, if the new nodes/node-positions were
derived in some way from the original nodes, then they would still be
tainted. However, surely we are trusting agreers to only use odbl/CT-
compatible sources to enter those new nodes/node-positions, so they
can't be tainted? If the agreer was actually creating a completely new
way, we are also trusting that they only use compatible sources to
position the nodes, so it is equivalent surely?


Another thought - What if the scenario is actually:
0) Agreer creates a way with 3 nodes and with tags 'name=Z' and
'highway=residential'
1) Non-agreer changes tags 'name=A' and 'highway=tertiary', and moves
all 3 nodes.
2) An agreer then deletes 1 node and moves the other 2 nodes, and
changes the tags to be 'name=B' and 'highway=secondary'.
is the way tainted or clean?I think that it must be clean.
Is this conceptually any different from the first scenario?
The only difference in scenario 2 is that the way was originally
created by an agreer rather than a non-agreer.
If we accept that all of the agreers must be using odbl/CT-compatible
sources, then surely both scenarios must result in a way that is clean?

[Following the discussion in the Legal-talk list, it looks as if it  
might be difficult to create a test that can be used generally for all  
tags, without accidentally letting some tainted data through - we  
would probably need to create very specific tests (e.g. "If 'highway'  
tag has been changed from <one official classification> to <a  
different official classification> then it's clean"), so that things  
like correction of the original non-agreer's spelling mistakes don't  
get 'cleaned' by agreers correcting the spelling mistake.]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/rebuild/attachments/20120206/542b8d79/attachment-0001.html>