Showing posts with label database. Show all posts
Showing posts with label database. Show all posts

10 March 2013

The Continuing Disaster Of Open Government In Germany

Recently, Techdirt noted that the European "database right" could pose a threat to releasing public data there. But that assumes that central governments are at least trying to open things up. A splendid piece by Sebastian Haselbeck on the Open Gov Germany blog, with the self-explanatory title "German government screws up open data," underlines that things can fail because the government itself sabotages transparency moves. 

On Techdirt.

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.

29 June 2009

Watching the Watchers

I read with interest this morning the following:

“Snitchtown” is an essay by Cory Doctorow that first appeared in Forbes.com in June 2007. This SoFoBoMo project is an attempt to illustrate that essay with photographs of some of the 4.2 million CCTV cameras currently estimated to be active in Britain.

It got me thinking: how about setting up a database - a surveillance of surveillance database - that has pictures and locations of CCTVs in the UK? It could be crowd sourced, and anonymous, solving problems of scaling and legal issues. If nothing else, it would put the watchers on notice that they are being watched....

11 February 2009

India Fights Patents with Huge Prior Art Database

One of the many problems with the patent offices around the world is that they are often unaware of prior art, granting patents for so-called inventions that are, in fact, common knowledge. In the computer world, there have been a number of efforts to provide prior art to patent offices, either after a patent is granted, in order to have it rescinded, or – even better – as part of the examination process. That's fine for a community with easy access to online source materials, but what about other fields, where prior art exists in other forms like books, or perhaps orally?

This is a particularly thorny problem for the sphere of traditional medicine. Substances derived from plants, for example, may have been in use for literally thousands of years, and yet patents may still be granted – especially in Western countries ignorant of other ancient medical traditions.

Perhaps the best-known example of this is the case of turmeric, commonly used as a spice in curries, for which patents were granted in 1995 on its wound healing properties by the US Patent Office, even though these supposedly novel uses had in fact been known for millennia.

To combat this problem, and to prevent its huge traditional knowledge basis being exploited in this way, India has created the Traditional Knowledge Digital Library (TKDL) database, which was unveiled on 2 February, and is now available to the Patent Examiners at the European Patent Office for establishing prior art in case of patent applications based on Indian systems of medicine.

Here's some background information on how the database came to be created and was set up:


The genesis of this maiden Indian effort dates back to the year 2000, when an interdisciplinary Task Force of experts was set up by AYUSH and CSIR, to devise a mechanism on protection of India’s traditional knowledge. The TKDL expert group estimated that about 2000 number of wrong patents concerning Indian systems of medicine were being granted every year at international level, mainly due to the fact that, India’s traditional medicine knowledge exists in languages such as Sanskrit, Hindi, Arabic, Urdu, Tamil etc. and was neither accessible nor understood by patent examiners at the international patent offices due to language and format barriers.

The TKDL breaks these barriers and has been able to scientifically convert and structure the information available in languages like Hindi, Sanskrit, Arabic, Persian, Urdu and Tamil, in open domain text books into five international languages, namely, English, Japanese, French, German and Spanish, with information contents in 30 million A4 size pages, with the help of Information Technology tools and a novel classification system - Traditional Knowledge Resource Classification (TKRC).

This is a huge, multilingual resource – something that could only be put together with governmental support and resources. It is also fairly specific to the domain of traditional knowledge. Nonetheless, it's a great example of how an extensive prior art database can be created and then made readily available to the patent authorities in order to help prevent patents being granted unjustifiably. It's a pity that we are unlikely to see anything quite like it for other knowledge domains.

02 March 2007

Googling World of Warcraft

WoW - or rather World of Warcraft - real search for a virtual world:

The Armory is a vast searchable database of information for World of Warcraft - taken straight from the real servers, updated in real time, and presented in a user-friendly interface. Since the Armory pulls its data from the actual game servers, it is the most comprehensive and up-to-date database on the characters, arena teams, and guilds of World of Warcraft in existence.

(Via Clickable Culture.)