Fair, Reasonable And Non-Discriminatory...Ain't

One of the inescapable facts of free software is that it involves a lot of law - far more than innocent hackers might expect when they settle down for a light bit of coding. That's in part because it is built on the foundation of licences like...


One of the inescapable facts of free software is that it involves a lot of law – far more than innocent hackers might expect when they settle down for a light bit of coding.

That's in part because it is built on the foundation of licences like the GNU GPL, which depend upon copyright for their efficacy (although that doesn't mean that free software couldn't survive without copyright – see my earlier discussion of this point with Richard Stallman.)

Unfortunately, it's also because software is impacted by the surreal world of patent law, which seems to spend most of its time considering how many angels can dance on the head of a pin, and what exact rhythms they are tapping out with their angelic plates of meat.

Even apparently dry-as-dust areas like standards can turn out to be legal minefield for free software. The key problem has to do with any software patents (which, of course, don't exist "as such" in Europe, but anyway...) that may be involved (as holders of such patents would like.) Obviously, any patents that might impact on standards will need to be licensed (otherwise no one in their right mind would adopt the standard); the question then becomes: on what terms?

There are two main options: Royalty Free (RF) and Fair, Reasonable and Non-Discriminatory (FRAND). The former does what it says on the tin: it means that there is nothing to pay, and people are granted a licence so that they can simply use the technology that somebody claims they patented without further ado. FRAND is more subtle.

At first blush, Fair, Reasonable and Non-Discriminatory sounds great. After all, how could anyone argue with something that promises to be fair and reasonable? The trouble with a FRAND licence is that it can indeed by eminently reasonable and eminently non-discriminatory, but if it is non-zero - even if it is extremely small – it is useless for free software licences.

The problem is that even a tiny licence fee, if levied on a per-copy basis, is impossible for free software because of the way it is distributed. There is simply no way to know how many copies may be shared, and hence no way to collect all those eminently reasonable and non-discriminatory licence fees.

This means that any standard that adopts FRAND licensing automatically excludes free software; the corollary is that for a standard to be truly open – that is, open for all to use, not just those able to count and pay for copies – it must adopt royalty-free licensing (as the World Wide Web Consortium does).

Thus the decision about something as apparently abstract and dry as the licensing terms for patents that may be involved in standards actually has an enormous knock-on effect on free software. If FRAND is adopted, those standards cannot be implemented by the latter, and so any moves to adopt these nominally "open" standards actually lock out real openness.

That fact makes an online "Study on the Interplay between Standards and Intellectual Property Rights (IPRs)", currently being conducted by the Technische Universität Berlin on behalf of the European Commission, extremely important.

As the FSFE pointed out in an email sent out to journalists (not online as far as I can find shortened online version now available), there are grounds for concern about this apparently neutral academic work:

This survey is a key component of the study that will play the major role in the EC's reform of standardisation policy.

That is, this matters very much in the real world. Moreover, the FSFE points out:

Some of the people producing this study also give some cause for concern. Particularly worrying is the fact that Benoit Müller is involved. He used to be the director of software policy for the Business Software Alliance, a lobby group one-sidedly devoted to the promotion of proprietary software. Furthermore Prof. Knut Blind, the study's coordinator, has long advocated licensing terms for patents in standards which are hostile to Free Software business models.

That means we run the risk of seeing FRAND, for all its reasonable unreasonableness, being adopted in EU standards. To avoid that risk, please fill in as much of the online form as you can: I've done so, even though it's not really addressed to individuals.

But given the likely weight of the push against the Royalty-Free option, we really need to make the case for it as strongly as possible, and that means answering the questions as best we can. According to the FSFE, the deadline is 17 September, so you have plenty of time to wade through them.

Follow me @glynmoody on Twitter or identi.ca.

"Recommended For You"

A Great Indian Takeaway FRAND Is Broken: Time To Bin It [Updated]