Is adoption of copyleft licenses really declining? Blackduck Sotware with its own secret recipe says so, as Matt Aslett reports. Whether or not this is true we can’t tell because BlackDuck doesn’t publish their method to collect and analyze its data. In any case, I think that if you look at this debate from a different angle, these numbers don’t count at all.
Twentyfive years ago there was no free/libre open source business and the GNU L/GPL licenses constituted the main corpus of free software available for use. The FSF used the strong copyleft mainly because of an ethical choice and also for practical reason to defend its code from appropriation by third party. Remember that at the time there was no ‘ecosystem’ and the notion of copyright to software was brand new.
Now free software is developed mainly by companies for business purposes: their choice is not only to defend code from appropriation but also to allow building thriving ecosystem for Open Innovation. The other reason I see for the large popularity of permissive licenses is the quantity of software developed by Apache Foundation. It’s amazing how far their code has gone from the original idea of a web server.
Whether BlackDuck has reliable numbers or somebody else does it’s irrelevant: numbers don’t count. I believe that if FSF wants to see more copyleft code out there, it needs to go back to its origin and get back to developing software that matters. The FSF needs to go beyond the GNU operating system and expand into the new areas of ‘cloud’, ‘social’ and ‘mobile’. Projects like MediaGoblin and GNU.FM are the first steps in the right direction but more is needed.