Sitting between a rock and a hard place

In the course of the last two days, two major aspects of the OOXML standardization efforts have been revisited by renowned experts. Unfortunately, what sounded like gestures of goodwill by Microsoft is being seriously challenged on a legal and technical point of view.

The Software Freedom Law Center (SFLC), the legal arm of the FSF, has published an in-depth analysis of the Open Specification Premise (OSP), the licence or covenant covering both OOXML and the now famous documentation on the binary formats used by the different flavours of Microsoft Office. The result is apalling. Basically, they found out that the OSP was vague, unclear, very limited in its coverage (both in the scope of coverage and the covered extent) and above all, was going against the letter and the spirit of Free Software.

Going against the « letter » of Free Software, meaning going against the four freedoms is being shown by the assertion found inside the OSP that it only applies to the specification. What this ends up meaning is that you’re not being covered for the code you write; this not only leaves room for a highway of legal interpretations, this also strongly dictates how you should use your code, and that goes strictly against the terms of the GPL. The OSP is also found to be vague enough to cast some serious doubts about the possibility to write GPL code against it; to be fair, that’s even ackowledged by Microsoft in its FAQ. It should be remembered here that Microsoft leaves the choice between the OSP and a RAND (Reasonnable And Non Discriminatory) licence, which, despite its name, is fully incompatible with GPL and other Free Software licenses. I’m sure Novell will have a cunning answer to that, but what this means, as I had written before, is that the OSP is legally unsafe and is a weapon against Free Software , without even mentionning Open Standards.

Another interesting note is that the OSP only applies to OOXML in the present version. Nothing prohibits Microsoft to withdraw it when a « OOXML 1.3 or 2.0 » would be published. And by the way, we don’t even know what’s the actual version of OOXML, since there is no definite specification coming out of the BRM yet.

On a more technical level, Stephane Rodriguez has posted a blog showing how much the whole buzz about Microsoft « opening up » and releasing its former binary formats’ specifications doesn’t sustain a careful examination. His blog post is quite long, so I will spare my readers all the details, except perhaps Stephane’s own words found inside his text, and I will quote them in guise of a conclusion:

«  What is meant by that is that, from Microsoft point of view, those documents may be “reference” documents in the sense that if you are taking a look at the corresponding source code, those documents are handy. But only they can do that. For everybody else, those documents are hints , not actual references . It is fire and motion, and it’s been going on for the best part of two decades.  »