OpenStreetMap identifies catastrophic effect of Brexit on database right

The Guardian today reports that the open-source mapping project OpenStreetMap, described as "Wikipedia-for-maps", is contemplating relocating from the UK to somewhere that is still in the EU. Although there seem to be manifold reasons, one of them (and I should have thought a pretty important one) is the fact that databases of UK origin will no longer enjoy protection in EU27.

What we know (and possibly love) as database right in the UK is, of course, a creature of EU law. Actually, it's old enough to be EC law: Directive of the European Parliament and of the Council 96/9/EC on the legal protection of databases, implemented by the Copyright and Rights in Databases Regulations 1997, SI 1997 No 3032. Those regulations were amended in 2003 and further amended in the blizzard of secondary legislation that was caused by Brexit, by the Intellectual Property (Copyright and Related Rights) (Amendment) (EU Exit) Regulations 2019, SI 2019 No 605 (Regulation 28 is the place to look, if you are interested). I wrote about all this not long ago, here, but it bears repeating in this new context.

The amendments are not complicated, although it takes a lot of legislative verbiage to convey what they are doing: where the original legislation mentioned the EEA, it now refers to the UK. This means that where previously a connection with the EEA (being a national of an EEA Member State, or incorporated in one, or habitually resident in one - I am going from memory here, so please check the legislation if it's important for you) meant your database qualified for protection in the UK, from now on (for databases created after IP completion day, as 11pm on 31 December 2020 is called - "completion" being a form of Newspeak and actually denoting something very different) you have to show the same connection with the UK. For databases created before IP completion day, protection continues on the old basis and for the original term, but I imagine that will cause all kinds of interesting problems when the proprietor finds it convenient to argue that there has been a substantial change in the database such that protection can start running again - only to find that it potentially means the database no longer qualifies for protection.

Anyway, that's the other side of the OpenStreetMap coin. What OSM are worried about is the treatment of databases of UK origin in the EEA, and of course with the UK no longer being in the EEA, UK databases (if I may call them that) no longer enjoy the protection of the sui generis right created by the Directive. Pre-existing databases will still be protected, by virtue of Article 58 of the Withdrawal Agreement, but given the nature of databases that is a wasting asset. Arguably, the OSM database is a pre-existing one and (the OpenStreetMap Foundation being incorporated in England) it therefore enjoys continuing protection, but the term is only 15 years: the saving grace, for database makers, is that a substantial change to the database "which would result in the database being considered to be a substantial new investment" qualifies the resulting database for protection. Never having had to think about it particularly before, I had reduced that rule to "a substantial new investment in the database keeps the 15 term running", but clearly that is too simplistic: in fact, a substantial new investment means you have a new database which is protected for 15 years, and that makes it abundantly clear why OpenStreetMap are unhappy.

Case C-604/10, Football Dataco Ltd v Yahoo! UK Ltd, Stan James (Abingdon) Ltd and others

Back in the mists of time - as long ago as an expedition to Luxembourg takes - I wrote about the early rounds of this litigation (The Trouble with Databases). The Court of Appeal decided, on the basis of the BHB case, that the fixture lists in suit were not protected by database right (or sui generis protection, as it would be known to the ancient Romans were they still with us), but asked the Sages of Kirchberg whether they might be protected by copyright. Well, I always thought that the directive had all but done away with copyright in databases, and replaced it with the Latin right, but one must always expect the unexpected. (Be aware that the sui generis right is also the subject of a reference to the Court of Justice in another Football Dataco case, C-173/11, also from the Court of Appeal.)

Of course, we cannot expect a straight answer from Luxembourg - not because the judges are unreliable in any way, but because they are there to interpret the law for us and it is definitely not for them to apply it to the facts. They say that a database will be protected by copyright if the selection or arrangement of the data which it contains amounts to an original expression of the creative freedom of the author, which of course is a matter for the national courts. The intellectual effort and skill used in creating the database are not relevant, and neither is whether the selection and arrangement includes 'the addition of important significance to the data' [sic] - what that means is anyone's guess. Finally, 'the significant labour and skill required for setting up that database cannot as such justify such a protection if they do not express any originality in the selection or arrangement of the data which that database contains.'

I'd take that as a 'no'.

No EU Rights in Fixtures Lists -- Court of Justice Judgment

The long-running claim by the English and Scottish professional football leagues, Football DataCo, and its sub-contractor PA Sport has finally come to an end, four years after it started: Football Dataco Ltd & Ors v Yahoo! UK Ltd [2012] EWCA Civ 1696 (12 November 2012). The Claimants told the Court of Appeal that, following the Court of Justice's decision, they accepted they did not have rights in the fixture lists.

The Court of Appeal declared that neither database copyright nor sui generis database rights subsists in the EU in fixtures lists. It ordered the claimants to post this message on their respective websites for a month clarifying the position:

Following the decision of the Court of Justice of the European Union dated 1 March 2012 in Case C-604/10 Football Dataco Ltd v Yahoo! UK Ltd & Others, it has been determined that neither copyright nor sui generis database rights subsist in the EU in fixture lists for the English and Scottish football leagues."

The claimants suggested at the hearing before the Court of Appeal that they might find a way to create fixture lists in a new way in future that attracted protection, but they do not seem to have any concrete ideas at present. Given that the Court of Justice found that the activity conducted by the Claimants in creating the fixtures should be disregarded when assessing whether the fixture lists merited protection, they sound very optimistic indeed.

Presumably now there will be rather a lot of claims for repayment of licence fees - especially for the period since the Court of Justice decided that there were no rights to support the claimants' licensing activities.

The trouble with databases

The first problem I have with databases at the moment is that there are two cases going through the courts, both of which have produced references to Luxembourg from the Court of Appeal, and in both of which the first claimant is Football Dataco (and the other claimants are mostly the same, too). Football Dataco Ltd & Ors v Yahoo! UK Ltd & Ors [2010] EWCA Civ 1380 is the first and Football Dataco v Sportradar [2011] EWCA Civ 330 the second: the expeditions to Luxembourg began respectively on 9 December last year and 29 March this year.

Of course, that's not the only trouble with databases. Other problems stem from the uncertain language used in the European Community's directive on the legal protection of the things, 96/9/EC, implemented in the UK by the Copyright and Rights in Databases Regulations 1997 (SI 1997/3032). The directive harmonises copyright protection for databases by requiring that to secure protection they must, "by reason of the selection or arrangement of their contents, constitute the author's own intellectual creation". Whether the stuff in the database is protected by copyright doesn't come into it: the directive gives that stuff no protection, though it might well have it anyway, and the directive (Article 1(2)) defines a database in terms that envisage a collection of material that has its own copyright protection as well as mere data that don't (and always remember, dear reader, that "data" is a plural, and resist the process by which usage is making it a singular noun, because where would that leave the word "datum", as in "Ordnance Survey Datum" and how would we then know whether sea levels were rising?):

... a collection of independent works, data or other materials arranged in a systematic or methodical way and individually accessible by electronic or other means.

In Case C-444/02, Fixtures Marketing Ltd v Organismos prognostikon agonon podosfairou AE (OPAP), the Court of Justice held that "independent" means "separable from one another without the value of their contents being affected", or alternatively (and it is not clear where the Court got this from, and it doesn't appear in the ruling at the end of the judgment) it means that the data have "autonomous informative value". Given that all these cases involve databases of football fixtures, let's be specific: there are dates, times, and names of teams in the database, along with venues perhaps and in due course scores (though at the stage in which the gambling industry is concerned those are obviously in the future).

So, I think it's fair to say that date, time and teams constitute independent materials when taken together. What isn't clear from the Fixtures Marketing case is whether that material constitutes data, but it's hard to see what else it might be. It's certainly not copyright works, and once the directive had specifically enumerated works and data the addition of other materials seems to serve little purpose - other than to allow the Court to leave us partly in the dark at this point.

What also isn't clear is whether the date, the time, and the teams are themselves data or  materials, and if so (and surely they have to be one or the other) whether they are independent. They have autonomous informative value, surely, although taken singly they convey no information about actual football matches.

In the Yahoo reference, the Court of Appeal asks whether the intellectual effort and skill of creating the data should be excluded when applying the "own intellectual creation" test. However, that test looks at selection and arrangement rather than creation per se - so there's a subtext here, with the court asking whether the creation of data (and perhaps it should have said "independent data", although it has left it open for the Court of Justice to elaborate on the difference) might constitute selection and arrangement. How about: selecting and arranging odd little bits of data with no autonomous informative value of their own, like dates and names of teams, so as to create what the Court of Justice has identified as independent data, can satisfy the test? There's a further complication, because the legislation talks at one moment about contents and at another about independent data, but I see no insuperable problem there - the database contains independent works, data or other materials (and the adjective seems, on the basis of the Fixtures Marketing case, to qualify all three nouns). But if it contains independent data it must logically contain non-independent data too, so perhaps the question boils down to whether selecting and arranging the non-independent data so as to create independent data can satisfy the "own intellectual creation" test.

Then finally there's perhaps the most important question of all:  "does 'author’s own intellectual creation' require more than significant labour and skill from the author, if so what?". In Case C-5/08, Infopaq International A/S v Danske Dagblades Forening, the Court of Justice stated that a work would only be protected by copyright if it was its author's own intellectual creation, a proposition which it took not from Directive 2001/29 but from the simple use of the word "work". That seems a shallow foundation on which to build a challenge to a fundamental principle of copyright law - but perhaps it will cut back some of the excesses of copyright protection that have grown up in the recent past. Crucial to determining what the scope of this newly-found requirement is, will be the answer to that question. If the "author's own intellectual creation" test is more rigorous than what we are used to in the UK if not elsewhere (as it surely must be) the answer to the Court of Appeal's question will show us the future of copyright law.

The Sportradar case involves another part of the directive, the one that created a sui generis right to protect databases - which in the UK we called "database right", the use of legal Latin being malum prohibitum. (But calling it a sui generis right, using a generic term for a particular example of the genre, is so extraordinarily unhelpful that one has to prefer the UK's approach.) The right covers unauthorised extraction and re-utilisation of the contents of the database, and re-utilisation means "any form of making available to the public all or a substantial part of the contents of a database by the distribution of copies, by renting, by online or other forms of transmission" (Article 7(2)(b)). The key issue in the case is jurisdiction, as there are proceedings on foot in Germany and the UK, and where the contents of the database are being made available determines which court can proceed with the case: is it Austria or Germany, where the servers are situated, or the UK where punters log on to the Internet to interrogate the data on the servers?

"Making available to the public" is also an important concept in copyright law, so the answer to this question will be important for more than just database right - and might be as revolutionary for the copyright system as the earlier question about intellectual creations.

Copyright yes, database right no

The other day, I did my standard half-day course on copyright for CLT. Less than half as much time as I need to cover copyright, I'd say, but OK for a sprint along the main straight. Far better to have time for a relaxed Marathon, appreciating the scenery, the birdsong, and the byways of the subject. But modern life doesn't allow time for that.

I talked about the difficulty - even the  near-impossibility - of finding anything in a database that could be protected by copyright, in the post-Directive world - the problems of 'author's own intellectual creation'. I outlined the sui generis database right, and told the delegates that the BHB case smashed a big hole in that with its requirement that the 'substantial investment' referred to in the Directive had to be directed to the right sort of activity, creating a commercial database rather than governing a sport.

Then I read Football Dataco Ltd & Ors v Brittens Pools Ltd (In Action 3222) & Ors [2010] EWHC 841 (Ch) (23 April 2010) and felt the need to eat my own words. Do I have an excuse based on the fact that my talk was only six days after the judgment was given? I don't believe it helps. In this new case, Mr Justice Floyd held that there was copyright in the fixture lists, the preparation of which had involved 'very significant labour and skill in satisfying the multitude of often competing requirements of those involved.' The process was not 'entirely deterministic' and not everyone would come up with the same answer:

Some solutions will better accommodate the requirements of the clubs and rules than others. The more sophisticated the compilation process, the more permutations it will be able to consider and the more requirements it will be able to satisfy. Judgments have to be taken as to the relative importance of certain rules in comparison to others. On occasions rules will have to be broken.

This work is not mere “sweat of the brow”, by which I mean the application of rigid criteria to the processing of data. It is quite unlike the compiling of a telephone directory, in that at each stage there is scope for the application of judgment and skill. Unlike a “sweat of the brow” compilation, there are some solutions which will simply not work, and others which will be better.

There is clearly more to making up a fixture list than I had ever thought - enough, it seems, to make it an original literary work. Just as well, from the point of view of the compilers, because there's no database right in it - the FML cases in the Court of Justice make that clear enough, and given the subject matter of the present case there was little hope of distinguishing the cases. Fixture lists (like the BHB's information) are created by first creating or collecting the data: the creation of the database does not involve anything like a substantial investment. An interesting judgment, which is going to reward further reading.