<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p><br>
</p>
<br>
<div class="moz-cite-prefix">Am 09.09.2016 um 21:30 schrieb Luis
Villa:<br>
</div>
<blockquote
cite="mid:CAFjZtRfhwnNpFO7_D1ByewOfaW0yHux53X3Rxy2m+=GHgFqNRg@mail.gmail.com"
type="cite">
<div dir="ltr">...<br>
<div>
<div>
<div class="gmail_quote">
<div>If this sort of provenance issue is a real concern,
OSM should push data sources to document and review
their processes, and perhaps consider improving
provenance data in the database directly (so that the
problems are easier to fix once identified). Just don't
look to licenses as a solution (or in this case, see OGL
as the problem).<br>
</div>
</div>
</div>
</div>
</div>
</blockquote>
<br>
Given that documenting our data sources or rather the lack of
documentation is a pet peeve of mine, I probably can't simply be
quiet about it :-/<br>
<br>
On the database/OSM data model side there are lots of issues with
cleanly removing and documenting the provenance of data (the well
known issue is tracking way splits) and given that objects can be
re-used/re-purposed/shared at will, even if a geometry object was
created by an import there is no guarantee that it continues to be a
part of/derived from the original imported data. <br>
<br>
There's not very much we can do about that at this late date.<br>
<br>
However we could make it easier to document the sources (right now a
big, largely unmaintained wiki table
<a class="moz-txt-link-freetext" href="http://wiki.openstreetmap.org/wiki/Import/Catalogue">http://wiki.openstreetmap.org/wiki/Import/Catalogue</a> ) and make it
easier to track which accounts were used for which imports and at
the same be more strict about vetting licences before imports
happen. Implementing a system for this is however really quite
boring and likely only going to happen if somebody pays for it.<br>
<br>
Simon<br>
</body>
</html>