On Thu, 12 Aug 2010 13:13:35 +0100, Ivor The Engine put finger to keyboard
and typed:

IANAL but to me 'reproduced, disclosed or disseminated to others'
would prohibit using the Avantix data set to populate queries outside
of the application. In other words, the fares themselves are not the
property of Atos Origin, but their compliation of them is. Discuss.

It's complicated. Really complicated. Intellectual property law is neither
consistent, intuitive nor (in many cases) logical. In this case, though,
I'm of the opinion that Atos Origin do have an enforceable database right
(not copyright) over the contents of their database.

To unpack that a bit, there are a couple of fundamental principles of
intellectual property (IP) law which are relevant here. Firstly, facts
themselves are not subject to copyright. However, their presentation may
be, and usually is. The copyright may be in the words used to convey the
facts - for example, it's a fact that (to pick a random headline from the
BBC news website), airport staff have voted to strike. But the article on
the BBC website is in their words, and copying that article, even if the
purpose is merely to convey the fact, is a breach of copyright.
Alternatively, the copyright may be in the means used to convey the facts -
for example, a computer programme which contains fares data is subject to
copyright, even though it merely presents facts which are themselves public
domain. Which is why you can't just distribute copies of Avantix, as that's
a breach of the copyright inherent in the software itself irrespective of
the legal status of the fares data it conveys.

Data mining the software, however, (or reverse-engineering the database
which the software reads) in order to extract the fares and then present
them independently of the software, is a different matter. In this case, no
copyright infringement has taken place (as the fares themselves are facts,
and not subject to copyright, while the software has not been copied and
hence no breach has taken place with respect to it). The relevant IP here,
therefore, is not copyright, but database right, which is a separate issue
and covered by separate aspects of legislation.

Database right protects a compilation of data from being reproduced or used
in an unauthorised fashion. Unlike copyright, database right can apply to
*any* collection of data, even if it doesn't meet the requirements for
being subject to copyright - so a collection of facts, such as football
fixtures, world records or railway fares, can, at least in theory, still be
subject to database right.

Whether or not a collection of data is protected by database right is
determined by the purpose for which the database was created and the amount
of work involved in creating it. The threshold for determining whether
database right applies was set in a binding judgement by the European Court
of Justice in a set of cases involving football fixtures and horse racing
data. The actual judgement is long-winded, and the cases themselves were
complex, but, in essence, the ruling is that data in a database is only
protected by database right if it has been compiled into the database
solely or primarily for the purpose of creating the database itself by the
organisation which owns the database and where the organisation which has
created the database has expended significant work in compiling it.

To give a couple of examples where this does and does not apply, I can
illustrate it with one of my own websites, www.motorwayservices.info (which
has recently featured in The Guardian and on BBC Radio 5Live). The site has
two major components: firstly, a database of every motorway service area in
the UK listing their locations and facilites, and, secondly, a database of
visitor comments about these areas. The first of these, the database of
MSAs, I can assert database right over - when I created it, there was no
existing unified list of MSAs (there were lists on each operator's website,
but no single list containing them all), so I had to compile it myself and
it took me a fair amount of effort to do so (interestingly, and rather
amusingly, when the Highways Agency needed to write a report on MSAs in the
UK, it used my website as a primary source!). The second database, however,
of visitor comments, I cannot assert database right over as I have put no
significant effort into compiling it - all I did was put a form on the
website and invited people to fill it in. (Being statements of opinion,
rather than facts, the visitor comments are subject to copyright, but that
copyright belongs to their authors, not me - I merely have a licence to
publish them).

Going back to Atos Origin and the fares database, therefore, the key
question is whether or not Atos Origin expended a significant amount of
work in compiling the database. And I think that they probably did, and
would be adjudged by a court to have done so. That's because the fare
structure is sufficiently complex, and has a significantly large number of
anomalies created by easements and the like, that any commercial
publication which gives access to fares will require a signficant amount of
verification and cross-checking in order to avoid errors, even if the
underlying data is readily available from the sources (the TOCs). That work
is enough to make the resulting compilation subject to database right, even
though no individual item within it - eg, an individual fare - has any IP
protection at all.

I'm assuming, of course, that it is Atos Origin which has actually compiled
the database. It may not be - it may well be National Rail. But, whoever
compiles it, it's subject to database right. If it's compiled by NR and
then licenced to publishers (such as Atos, or the online ticket vendors),
then the publishers in turn are bound by any conditions attached to that
licence - which almost certainly forbids further redistribution. So it's
irrelevent whether Atos own the rights to the data, or are merely licensing
it from the owners, they still have an enforceable claim against anyone
infringing those rights.

Mark
--
Blog: http://mark.goodge.co.uk
Stuff: http://www.good-stuff.co.uk