Can 11 words breach copyright?

Thanks to Corus for alerting me to an interesting ruling on Copyright from the European Court of Justice - the highest court in the EU in matters of European law.
The ruling refers to a court case in Denmark between the Danish newspaper industry body DDF and a clippings service Infopaq over reproduction of 11-word snippets from newspapers for sale to clients.
Infopaq's clients identified keywords that they wanted monitored. Infopaq scanned newspapers and used software to locate those keywords and print them out with the five words on either side to provide a context.
Infopaq claimed that this copying was "transient" and so was covered by exceptions in the EU Copyright Directive. However the ECJ ruled that "transient" only covered the creation of files resulting from the process of converting image files which were then deleted automatically from the computer memory. Printing out these files produced a non-transient record which was not exempt.
The court then looked at whether these 11-word extracts formed a element that was protected by the law, in other words were they a substantial part? The court stated that "words as such do not… constitute elements covered by the protection". It is the "choice, sequence and combination of these words" that creates a protected element. It continues:

"That being so, given the requirement of a broad interpretation of the scope of the protection conferred by Article 2 of Directive 2001/29, the possibility may not be ruled out that certain isolated sentences, or even certain parts of sentences in the text in question, may be suitable for conveying to the reader the originality of a publication such as a newspaper article, by communicating to that reader an element which is, in itself, the expression of the intellectual creation of the author of that article. Such sentences or parts of sentences are, therefore, liable to come within the scope of the protection provided for in Article 2(a) of that directive."
The court did stress that the question of whether a particular sentence or part of sentence did come within the scope of protection had to be decided by national courts.

So what does this mean for information professionals, especially those that provide abstracting services or keyword alerting services?

Although the EU Copyright Directive has been implemented in to member's legislation we are still working with separate Copyright laws so you should not assume that all aspects of the Danish case apply to the UK.

The Copyright Directive removed the concept of fair dealing from copying for a commercial purpose. In a non-commercial context this may still apply.

Personally I am surprised that Infopaq tried to argue that their extracts were exempt because they were transient. Copyright expert Graham Cornish in his presentations and writings on copyright has indicated that the temporary copies exemption applies only if they are a necessary part of the technical process to transfer the information and the copy does not have any independent economic significance. However this does demonstrate that nothing in legislation can be taken for granted until it is tested in a court.

The issue of whether extracts form a "substantial part" of the whole is also interesting. Again Graham Cornish has pointed out that "substantial" is not defined in the legislation and that you cannot assume that a short extract is unsubstantial. The ruling confirmed that individual words cannot be copyright protected so if Infopaq had just printed a list of keywords and their location in an article this would not have infringed copyright. The problem arose from trying to put the keywords into context by printing the words on either side. If they had just printed one word on either side it would probably be unsubstantial and not infringe copyright but at the same time it would not provide enough context to be of any use. The paradox is that the closer you get to providing a useful context the more likely you are to infringe copyright. Of course, if you write your own abstract, or provide a context for a keyword in your own words, then you are not infringing copyright. The problem is that organisations are looking to reduce the amount of staff time spent on these professional activities by using automated systems. Possibly Infopaq or someone will come up with a piece of software that can generate a context for a given keyword without resorting to cut and paste - although this still seems to be something that the human brain can do better and faster than a computer.

A final point is that copyright legislation is only breached if you do not seek the permission of the copyright holder for these acts. In theory Infopaq could have asked DDF for permission on the basis that providing keyword alerts to its clients would benefit the newspapers by directing more customers to those titles. Of course DDF may well have asked for a licensing fee to permit this.

The main lesson from this case for librarians is that any copied extract, no matter how short, may be covered by copyright. If it is worthwhile to someone to copy it then the chances are that it could be a substantial part.

The full text of the ECJ ruling can be found at http://tinyurl.com/m6autk