[OSM-talk] Anonymous login, yet again- was privacy policy

[Andy Robinson]

Nice summary David and I broadly agree with your opinion.

On the final point about permitting some anonymous edits I think this could
work. If anonymous edits were restricted to a special set of tags then those
tags could be available to editors for use in making corrections and to
users to see alternatives and suggested changes which have not gone live
yet. I've spotted this type of approach elsewhere. On Ancestry.com for
instance there is the ability to tag a poor census transcription with a
comment as to what you believe the transcription should read. These comments
get attached to the data rather than changing the original data entry and
new visitors can see this alterative information when the visit the same
census page and use it as they see fit. The same sort of approach could be
made with OSM. An anonymous user would be able to offer new or different
spelling for a road name for instance but this information would be attached
to the element in question rather than superseding the "name" tag. A logged
in editor could then make the correction or add the name if its verifiable.

We were discussing on irc last night how we can maintain confidence in the
validity of data in OSM. This might come by weighting data with confidence
statistics. User edits (both logged and anonymous) could plan an important
role in helping give the project continued confidence in the validity of
data.

Cheers

Andy

Andy Robinson
Andy_J_Robinson at blueyonder.co.uk 
>-----Original Message-----
>From: talk-bounces at openstreetmap.org [mailto:talk-
>bounces at openstreetmap.org] On Behalf Of David Groom
>Sent: 28 July 2006 00:47
>To: talk at openstreetmap.org
>Subject: [OSM-talk] Anonymous login, yet again- was privacy policy
>
>I think we are going to have to accept there is no absolute answer to the
>question of anonymous login.  But here are my thoughts for what its worth.
>
>1)  I can see no real downside to anonymous access to "read" OSM data.
>
>2) OSM data is not like words written on Wikipedia, so don't cite that as
>an
>example of how to proceed.  To explain:
>
>a) as Martyn Welch (and others) have said its far easier to spot an error
>on
>Wikipedia than it will be on OSM data;
>
>b) I'm not a believer in conspiracy theories, but I have to accept the
>possibility.  Something written in Wikipedia is in general not likely to
>cause commercial harm to anyone, and therefore:
>
>i) wikipedia "data" is unlikely to be corrupted simply to cause a
>commercial
>disadvantage to someone;
>ii) if something on wikipedia does come from a copyrighted source it is
>unlikely to cause financial loss to someone;
>
>However, at least in the United Kingdom, the data being produced by the OSM
>project directly effects the revenue earning potential of a number of large
>powerful organisations which therefore have a distinct reason to note any
>breaches of copyright, and take appropriate action.
>
>For these reasons OSM has to be more careful than Wikipedia in identifying
>the identity of contributors, and separating one contributor from another
>( i.e. no anonymous write access) so that problem data can more easily be
>removed.
>
>c) OSM data is more "viral". Suppose someone writes something on wikipedia
>that is wrong, it can be removed and this doesn't really effect the rest of
>the article.  However suppose someone (logged in as anonymous) puts in a
>few
>nodes that are derived from a copyright source, then someone goes and uses
>those  nodes to form a segment, and then someone else uses that segment to
>form a way.  In deleting those nodes we then have to delete someone else's
>work who created the segments and the way.  Not only that, but we then have
>to back out all data created by that anonymous source, even if it doesn't
>belong to the corrupt nodes.
>
>d) having said all the above, I can see the argument for the fact that if
>someone is looking at OSM maps and sees a name of a street is missing, or
>spelt wrong, that they should be able to easily correct this.  Such a
>correction would not be "viral" in the case of (c) above as it only affects
>the key/value of an existing entry.
>
>To summarise:
>
>A) Read access could be anonymous
>B) Uploading of gpx tracks, creation of ways, & segments MUST require an
>OSM
>user account
>C) I personally don't like the idea, but, adding key / values to EXISTING
>ways might be allowed by anonymous login.
>
>David
>
>
>
>
>
>
>
>_______________________________________________
>talk mailing list
>talk at openstreetmap.org
>http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk