06 October 2009

Postcodes: Royal Fail

Here's a perfect example of why intellectual commons should not be enclosed.

The UK Postcode data set is obviously crucial information for businesses and ordinary citizens - something that is clearly vital to the smooth running of everyday life. But more than that, it is geographic information that allows all kinds of innovative services to be provided by people with clever ideas and some skill.

That's exactly what happened when the Postcode database was leaked on to the Internet recently. People used that information to do all sorts of things that hadn't been done before, presumably because the company that claims to own this information, Royal Mail, was charging an exorbitant amount for access to it.

And then guess what happened? Yup, the nasties started arriving:

On Friday the 2nd October we received correspondence from the Royal Mail demanding that we close this site down (see below). One of the directors of Ernest Marples Postcodes Ltd has also been threatened personally.

We are not in a position to mount an effective legal challenge against the Royal Mail’s demands and therefore have closed the ErnestMarples.com API effective immediately.

We understand that this will cause harm and considerable inconvenience to the many people who are using or intend to use the API to power socially useful tools, such as HealthWhere, JobcentreProPlus.com and PlanningAlerts.com. For this, we apologise unreservedly.

Specifically, intellectual monopolies of a particularly stupid kind are involved:

Our client is the proprietor of extensive intellectual property rights in the Database, including copyright in both the Database and the software, and database rights.

Here's what Wikipedia has to say about these "database rights":

The Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the legal protection of databases is a European Union directive in the field of copyright law, made under the internal market provisions of the Treaty of Rome. It harmonizes the treatment of databases under copyright law, and creates a new sui generis right for the creators of databases which do not qualify for copyright.

Before 1996, these sui generis "database rights" did not exist; they were created in the EU because lobbyists put forward the argument that they would offer an incentive to create more databases than the Americans, whose database publishers strangely didn't seem to need this new "right" to thrive, and so make the mighty EU even mightier - at least as far as those jolly exciting databases were concerned.

Rather wisely, afterwards the EU decided to do some research in this area, comparing their creation before and after the new sui generis right was brought in, to see just how great that incentive proved to be - a unique opportunity to test the theory that underpins intellectual monopolies. Here are the results of that research:

Introduced to stimulate the production of databases in Europe, the “sui generis”protection has had no proven impact on the production of databases.

According to the Gale Directory of Databases, the number of EU-based database “entries” was 3095 in 2004 as compared to 3092 in 1998 when the first Member States had implemented the “sui generis” protection into national laws.

It is noteworthy that the number of database “entries” dropped just as most of the EU-15 had implemented the Directive into national laws in 2001. In 2001, there were 4085 EU-based “entries” while in 2004 there were only 3095.

While the evidence taken from the GDD relies on the number of database “entries” and not on the overall turnover achieved or the information supplied by means of databases, they remain the only empirical data available.

So, the official EU study finds that the sui generis protection has had no proven impact on the production of databases; in fact, the number of databases went *down* after it was introduced.

Thus these "database rights" have been shown to stifle the production of databases - negating the whole claimed point of their introduction. Moreover, the Royal Mail's bullying of a couple of people who are trying to offer useful services that would not otherwise exist, shows the danger of entrusting such a critical data commons to commercial entities who then enclose it by claiming "database rights" in them: they will always be tempted to maximise their own profit, rather than the value to society as a whole.

Giving the Royal Mail a monopoly on this critical dataset - one that for all practical purposes can never be created again - is like giving a genetics company a monopoly on the human genome. That was attempted (hello, Celera) but, fortunately for us, thwarted, thanks largely to free software. Today, the human genome is an intellectual commons (well, most of it), and the Postcode data should be, too.

Follow me @glynmoody on Twitter or identi.ca.


Andrew Katz said...

Hi Glyn

You know about this?


- Andrew

Glyn Moody said...

@andrew: I did, but I'd forgotten - thanks.

The problem is, there's no hope the complete postcode database could be created in this way, so it's not really a solution....