The peculiar case of data sharing: are intellectual property rights relevant to begin with?

Hans Graux

This is the second article in the series of opinion pieces that dive into data sharing as a service. Will data services remove intellectual property rights from the picture, and at what cost?

In the previous article of this series, the shift from an ownership economy towards a service economy was introduced. In order to assess whether this shift may be occurring for data sharing as well, it is firstly relevant to assess to what extent intellectual property rights are relevant to data. Intellectual property rights exist in many forms: copyrights, database rights, patents, trademarks, and others. For the topic of data sharing, copyrights and database rights are particularly relevant. 

Copyrights generally apply when a work can be deemed original, in the sense that it can be considered an individual expression of its author(s). At the EU level specific copyright law exists, which define the rights of the copyright holders, and a harmonised series of exceptions to these rights. Databases are protected under the "Database Directive" which established that databases could qualify for copyright protection if they constitute the author's own intellectual creation. In the absence of such personal originality, no copyright applies. Separately, it established database rights for the makers of databases, irrespective of whether they qualify for copyright. This right is granted whenever the maker had to undertake a substantial investment in either the obtaining, verification or presentation of the contents of the database, which enables makers of databases to benefit from legal protections even when there is no particular originality to their work.

To what extent do intellectual property rights then apply to digital data? Obviously, traditional original works – such as music and films – benefit from copyrights irrespective of whether they are rendered as digital data, and the same applies for software (including video games): any computer programme that shows sufficient originality qualifies for copyright protection.

But what about non-original digital data on which data sharing discussions normally focus? This includes measured data, observed data, metadata, and any other kind of factual data that intends to capture an objective reality, devoid of significant creativity of any author. Common examples include geographic data, statistical data, administrative registers, ecological performance data, product descriptions, component lists, economic indicators, socio-economic and demographic data, and so forth. The central question is whether these qualify for intellectual property rights, principally copyright or database rights.

With respect to copyrights, the answer is relatively straightforward: none of these examples of data show signs of originality from any author, and therefore the existence of any copyright is doubtful in most cases. A protection by database rights under the Database Directive on the other hand is more plausible. The only criterion for the applicability of database rights, as mentioned above, is that a substantial investment was needed to create the database. That should mean that database rights would usually apply, since the creation of these databases usually requires significant investments from public authorities.

But the issue is not that simple, since the scope and breadth of database rights has been somewhat eroded over the past two decades. In a series of decisions 1 2 3 4  the European Court of Justice stressed that the database right does not apply to databases that are the simple by-products of the main activity of an organisation. As a result, it could be argued that database rights would not apply broadly to any databases created by public administrations or private companies that were merely necessary for their administration, products or services to function, including e.g. most records created or kept by public administrations, or any automatically captured or machine-generated data (such as data from IoT devices), since such data is principally a by-product. If that logic is indeed accepted, then for such data sets, licensing based on copyright or database rights would not be an appropriate choice.

In this environment, are intellectual property rights still relevant for data sharing in the public sector? Given the description above, intellectual property rights – and specifically copyrights and database rights – play an important role for digital data that shows a certain level of originality from a human author, or for databases that required a substantial investment independent from the main activities of their creator. Where these requirements are not met however – and as the examples above show, this comprises a rather broad range of digital data in the public sector – intellectual property rights don’t apply, and licensing intellectual property rights is therefore not legally possible.

Data sharing as a service however, based on subscription agreements, would certainly be possible, since such agreements don’t require the existence of intellectual property rights. In the next article in this series we delve into this. 

Stay tuned for the next article in the series: Will data services remove intellectual property rights from the picture, and at what cost?

The peculiar case of data sharing: are intellectual property rights relevant to begin with?
Image credit:
2017, Giammarco Boscaro via Unsplash

For questions and comments, please visit our forum on Futurium.