HADW,

I can see where your concerns stem from, & agree that some form of suitable source code to indicate where the data is derived from could mitigate any potential problems. At present there are some mappers, myself included, who are obtaining postcodes derived from the 'Land Registry price paid data' - we believe we are complying with the current legal requirements & are not introducing anything tainted into the database.

If we are subsequently shown to be wrong it could help any redaction process if we were all using the same tag to indicate the postcode was derived from the Land Registry Price Paid database. I'm currently planning to start using

source:postcode=Land Registry 'price paid' data

unless there is any strong guidance to indicate something more suitable.

As the postcode is the only part of the data I am using, another simple search would be for any postcodes, and then check their source tag.

Regards

Nick
(Tallguy)

On 29/09/13 01:01, OpenStreetmap HADW wrote:
On 28 September 2013 20:52, Rob Nickerson <rob.j.nicker...@gmail.com> wrote:
Oh I should add that I am a fan of source tags on the objects myself so I
add a "source:postcode=Land Registry 'Price Paid' data" tag (or =ONS
Postcode Centroids) to my edits. Not everyone agrees that source tags should
be added to the objects, preferring instead to add them to the changeset
comment.
I use a mixture of both, depending on the context.

Both are flawed, though.

Adding to objects doesn't work well when different tags have different
sources, and the geometry may have a different one too.  It is also
very vulnerable to people replacing the whole source with theirs.

Adding it to changeset comments means it doesn't survive splits and
merges.  The database has no knowledge that these actually happened
and doesn't record the audit trail necessary to find the true
provenance of a node, way or relation.  (For this reason, any
mechanical redaction is likely to be quite flawed.)

This lack of good traceability does worry me, as I see one of the
biggest threats to a cloud sourced project like this is people getting
over enthusiastic and importing copyright data, possibly in such small
individual amounts that no alarms sound, but when aggregated, enough
to get a copyright owner angry.  My feeling is that the upcoming
generation of contributors isn't so steeped in the concept of a map
that is untainted by material with restrictive copyrights, so will use
the easiest way of getting the data they want added.

The most important reason for sources may be to limit what has to be
taken down when a take down notice is received.

_______________________________________________
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb


_______________________________________________
Talk-GB mailing list
Talk-GB@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-gb

Reply via email to