[Osmf-talk] Future of DWG work, copyright, vandalism

[Guttorm Flatabø]

On Thu, Jun 7, 2012 at 12:13 PM, Frederik Ramm <frederik at remote.org> wrote:

> Dear OSMF members,
>
>   I would like to hear your opinion about two aspects of DWG work and how
> we would like to handle that kind of work in the future.
>

I think these are very important issues, and largely agree with you.


> DWG's daily work falls largely in three categories:
>

Coming from Wikipedia, the most important "drawback" as it currently
stands, that I see, is that the DWG has to handle all these issues in the
first place, and that they are handled on a non-web and less open medium
that a mailinglist is. I really miss to contact editors on "talk pages" and
discuss edits and objects publicly in a similar way to how it is done with
talk pages on Wikipedia (all pages on Wikipedia has a talk page).

The way it is done today is inaccessible and "enveloped" in that you first
need to contact users privately (risking to be filtered out as spam, etc.),
wait for a reply, find out about the DWG (which means what?), contact the
DWG, join the mailing list if you want to be involved in the discussion
(?), and follow up. If other users find the same error or another error by
the same user they will not know about the communication you've had.
Already the first step in a suspected case of wrongdoing could benefit from
talk pages. If the issue was handled on the "vandals" user talk page,
anyone could step in with their opinion and view, to resolve
misunderstandings and help new users do things the right way. Vandals could
be "marked" so others see there have been issues with them before, and you
wouldn't have to be a member of a mailing list (which is a pretty exotic
phenomenon for most internet users I'd argue) to be involved.

The wiki already have user talk pages, so a better integration could
provide those. As for discussing changesets and objects, I don't know how
things could be done, but if we don't have public records for these issues
connected directly to the data at issue, it seems to me that we risk a lot
of overlooked and forgotten errors. If I didn't have good memory ("todo
list"), this <http://www.openstreetmap.org/browse/changeset/11613749> would
be one of them (and I still don't know if the other edits the user has done
is vandalism).

[snip]


> We simply do not have the manpower to actually research cases like that.
> We need a simple policy that allows us, or ideally the community, to deal
> with such cases.
>

Indeed, we need to empower the community, not only to map, but to
effectively communicate about the data. The DWG (and the mailing list)
should be a last resort for the hard cases, or the cases where technical
assistance is needed.


> Such a policy could for example be one of
>
> * "Data for which no credible source is given in the changeset or on
> request by the contributing mapper is subject to deletion. The mapper must
> demonstrate, on request, that his source is legal. If the mapper chooses
> not to tell us his source then we must assume it is not legal." (In many
> cases the "credible source" could be "survey" but the boundary of a
> protection area might not always be surveyable.)
>

This seems very harsh, and could perhaps more or less justify vandalism
("he didn't provide a source and thus I can delete").


> * legal measures - pay lawyers to go to court and request the real address
> of the IP number that vandalizes our data, then send nasty letters to
> people and demand money. (I imagine that most vandals must be kids who
> would be in for trouble with their parents once they start getting recorded
> letters.) If the movie industry can do it, so can we, provided that
> vandalizing our data is indeed illegal. This might require "hardening" the
> process first, i.e. getting legal advice for updating our CT to include
> wording that makes it clear(er) that edits not founded in reality are
> considered vandalism, and it might also require lawyer-approved notices to
> be sent out by DWG (instead of my usual informal "hey, stop that, we don't
> like it").
>

Tweaking the CT would probably be good in any case. There is a "middle way"
similar to this though, and that is to try to contact the "real user" or
their network provider. In the Oslo incident we were pretty close to
getting his identity and contacting, but not close enough. Contacting a
school or work place with a log from the osm servers could be effective
though.


--
Guttorm
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/osmf-talk/attachments/20120607/85536a74/attachment.html>