[Openstreetmap] London locations

[richard at systemed.net]

Quoting Lars Aronsson <lars at aronsson.se>:

> Richard Fairhurst wrote:
>
> > TeleAtlas (who provide the relevant Google mapping), and
> > potentially _their_ suppliers (which most probably means
> > Ordnance Survey), have database right in the gazetteer used to
> > draw placenames on the Google map. This is the big database
> > table which contains information like your Glasgow lat/long for
> > every single town and village in the country, and OS sell this
> > as a commercial product.
>
> Ah, yes, now you are talking database rights.

Indeed - first thing that sprang to mind at 6-something this morning. :)

> That kind of
> protection is valid if I copy a larger number of facts, but not if
> I copy a single piece of information (a single coordinate).
> Furthermore, I must copy the information, at least down to a few
> decimal places.  If I just look at the map and estimate the
> coordinates, I'm not copying the informations in the database.

If you're "just looking" and estimating one single co-ordinate by sight, then
yes, I'd agree you probably have a case.

Placeopedia isn't doing that. Placeopedia has _lots_ of co-ordinates
georeferenced directly by people clicking on Google Maps.

I don't think that the distinction as to "a few decimal places" - or indeed, any
form of accuracy limit - is upheld in law. If I were to publish a song that
declaimed "the answer, my friend, is blowing in the breeze", Bob Dylan would
still sue me.

> If you can provide me with a U.K. court case where database rights
> have been claimed and *upheld* in a broader sense than this, I'd
> be interested.

As would I!

This one appears to be the test case (thus far) for the database right:
http://www.it-director.com/article.php?articleid=12460

The database right was not upheld in that case, so it doesn't fulfil your
criterion. But two relevant extracts:

"The acts of extraction and reutilisation will only amount to an infringement of
the Database Right where they... are carried out “repeatedly and systematically”
in relation to “insubstantial parts” of that database."

"...“reutilisation” means making available to the public the contents of a
database. These types of acts will be preventable where they result in
significant detriment, evaluated qualitatively or quantitatively, to the
investment made by the database owner. It is irrelevant whether the acts are
carried out for commercial or non-commercial purposes, and it is equally
irrelevant whether or not the owner of a database has already made the contents
of that database accessible to the public."

>From another summary of the same case
(http://www.managingip.com/?Page=10&PUBID=34&ISS=12521&SID=472704&TYPE=20):

"...direct access was not required for infringement by extraction or
reutilization. Extraction or reutilization indirectly via intermediate sources
(for example from a print medium or the internet) would suffice."

To my (non-lawyerly) eye, all of that looks like a pretty comprehensive
undermining of Placeopedia. But I would love to be proved wrong, as it's a
great project.

As per Frank's message, it's better, anyway, to proceed on a basis of "what do
we know is legal?" rather than "what can we get away with?". OSM, in
particular, needs to be whiter than white.

(The case above does look interesting and hopeful in some other respects and I'd
like to study it in more detail.)

> Furthermore, database rights expire sooner than the life+70 years
> term that is valid for copyright.  I think they expire 15 years
> after publication in Sweden, but I'm not sure how it can vary
> between countries.

Yes, I think it's 15 years over here, too, and I think that's a EU-"inspired"
term.

> [snip]
> The really interesting case, that will prove if "database rights"
> lawyers can out-think the Internet generation, or if it is the
> other way around, is this: If I copy just a few facts from some
> database, and you copy just a few other facts from the same
> database, and thousands of us come together at a fact swap-fest,
> will the resulting smorgasbord of facts constitute an infringement
> of the database rights?  We should call this test ... "Wikipedia".

:) Agreed.

At present we're stuck, for the most part, with pre-Internet laws and lawyers.
All I can say with any confidence is that no lawyer would ever clear Wikipedia
for publishing in print, irrespective of Wikipedia's own licence.

> > Daniel's point about electoral boundaries is a useful one. Those
> > electoral boundaries are derived works, too. They've mostly been
> > obtained by drawing lines on an Ordnance Survey map, which is
> > copyrighted. For the local council, there's absolutely no
> > problem in this, because they get their OS mapping free through
> > the Pan-Governmental Agreement. It just hobbles any of the rest
> > of us wanting to do good works with it.
>
> So are there any U.K. court cases where the O.S. has defended
> their claims of copyright to electoral boundaries?

Don't know, sorry. (The MySociety folks might.)

> Electoral boundaries are just a list (database) of coordinates, so
> they cannot be covered by traditional copyright, but they can be
> subject to database rights.  Do electoral boundaries change more
> often than database rights expire?  (Say, 15 years.)

Right, I really _don't_ know anything about this, but my understanding is that
UK electoral boundaries are not just a list of co-ordinates. They're usually
defined (in statute) by a map appended to the relevant legislation, like this:
http://www.bristol-city.gov.uk/council/ward/pdf/cabotmap-high.pdf . The map is
copyright Ordnance Survey for the next 50 years; the electoral boundaries are a
derived work; and therein lies the problem.

Richard

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.