--- title: "Subscriptions for software and commodity fetishism" layout: post image: feature: header_software.png --- I was thinking idly today -- and probably in a wildly unoriginal way -- about some of the disputes about subscriptions to software and the politics of this model. It's no secret that Richard Stallman, perhaps _the_ core philosopher of the open-source software movement, is a problematic figure, most recently so in his [comments about Marvin Minsky](https://www.theverge.com/2019/9/17/20870050/richard-stallman-resigns-mit-free-software-foundation-epstein). Gabriella Coleman has also conducted a [lengthy study of the politics of open-source software](https://press.princeton.edu/books/paperback/9780691144610/coding-freedom), concluding that it is 'liberal' in its politics (that is to say: not radically left). One of Stallman's [recent pronouncements](https://archive.factordaily.com/richard-stallman-india/) caught my attention though: > Another malicious functionality I should mention is subscriptions. About 20 years ago you would simply have installed software on your computer and run it. Nowadays, the software requires users to get a subscription and identify themselves. That includes Microsoft Office. It includes a lot of software that artists use. So this means the chains are getting tighter. This appears to be a prevalent view among open-source/free-software developers; you should "own" the software that you have the right to and to be able to inspect it. Fair enough on the latter part. But the "subscription" model above is actually a more progressive model in some ways if you think about supporting the labour of programmers. The view that people should "own" some output is akin to a type of "commodity fetishism", where people believe that what they are paying for is the end output product, rather than supporting the labour that made it possible. This is what always seems missing to me in the supposed radical freedom of free software. Freedom can only come, in our world, whether we like it or not, with some degree of concomitant economic freedom. Companies resort to developing closed-source software in order to make money (and lots of profit in some cases). Lots of open-source/free software projects also make money by licensing commercial versions etc. But the practical challenge is how to value and pay for labour. I believe that a subscription, that recognises the _ongoing work_ of software development (which is never done), is actually a _good_ way of paying for software. It's just that many people will not pay for software if they can get it for free, which is where the coercive practices of uninspectable code etc. come back in,