[talk-au] Edits vs edits

John Smith deltafoxtrot256 at gmail.com
Sun Aug 15 02:08:39 BST 2010


On 15 August 2010 08:15, Richard Weait <richard at weait.com> wrote:
> Are you suggesting that such a bulk edit has happened?   I'm not aware
> of edits that match that description perhaps you can link to the
> changesets for reference?

I'm not suggesting any such thing has happened, but if people are
going to abuse stuff it's possible that something like this might
occur by someone in support of CT/ODBL if/when they find out just by
shifting nodes by a random amount may give them the right to claim
they "created" the node even if it shares the node id with a previous
mapper.

> Presuming that regular mapper, A_Mapper surveys and uploads their
> neighbourhood, Mapperton, then other mapper, O_Mapper shifts
> everything by a random 0.1mm, a superficial look at the current data
> would show that O_Mapper touched all the data last.  A look at
> node/way history would show that A_Mapper created version 1 of that
> data.
>
> Given the example above, and presuming that A_Mapper chose to accept
> the license upgrade and O_Mapper did not, what would you recommend for
> the data?

You seem to have it backwards, A_Mapper refuses to accept, but
O_Mapper did all the updates so the data could be accepted under the
new license, even if A_Mapper refuses since the nodes would be
considered to be created by O_Mapper even if the node IDs match.

> a) leave as version 2
> b) revert to version 1
> c) delete data
> d) something  else.

You have to delete something, reverting may not help if my
understanding of some people's interpretation is valid in that the 2nd
version may still be deemed acceptable under ODBL, option a is morally
questionable in the same sense.

> What about the reverse?  What would you recommend be done with the
> data if A_Mapper did not accept the license upgrade, but O_Mapper did
> accept?

That is what I was taking about, Grant was the one commenting on the
reverse and as I understand it that sort of suggestion is what might
lead to people doing abusive edits, regardless if they are for or
against.

> There is a thread on osm-dev@ discussing how to model OSM data
> regarding ODbL/CT acceptance.  What you describe above sounds a little
> different than the current discussion on osm-dev@, what would you call
> it?  "edits vs. trivial edits", "user edits vs. automated edits" ?

I call it potential for abuse :)



More information about the Talk-au mailing list