Some thoughts on open source

Historically my attitude to open source has been one of avoiding the GPL like the plague and sticking everything under a permissive license.

There are a couple reasons for this:

  1. Generally speaking I want people to use my stuff. The GPL acts as an obstacle to that.
  2. I want to spread my ideas. The GPL acts as an obstacle to that too, although less so.
  3. Most of my open source code is not stuff I’m emotionally attached to and frankly if someone wants to take it off my hands that’s great – it means I don’t have to maintain it
  4. I’ve used a lot of open source code in my commercial work that I wouldn’t have been able to use if it was GPLed. It seems churlish to not return the favour.

I’m… starting to rethink some of this attitude.

Here are some links:

  1. How to capture an open source project
  2. Github is not your CV
  3. The Ethics of Unpaid Labor and the OSS Community

Fundamentally, open source may claim to be about freedom, and most people who do it may even have that intent, but when judging systems intent is irrelevant, what matters is the structural effect.

And the structural effect of the current relation between the vast quantity of permissively licensed open source work and the industry is one of overvaluing people who are capable of putting in work for free, devaluing everyone’s labour, and allowing the industrial scale parasite of VC backed startups to leech yet more value from the commons without putting anything back at all.

And it’s at this point that I go have an existential crisis and go gibber in the corner for a bit.

Because this all ends up falling under what I regard as the fundamental question of my life: How can I make the world better while not also making the world worse and make enough money to eat at the same time?

Answers on a postcard. Or a blog comment. Or anything really because if you’ve got the answer can you please let me know? Because I’ve no idea.

Because the stuff I give away for free is the closest I’ve come to achieving the first even if it falls down hard on the latter. It sounds terribly megalomaniacal, but the point of this blog is at least partly to make the world a better place, and the point of Hypothesis is explicitly to raise the software quality bar for software written in Python (This is part of why I’m putting in quite so much work on the edge cases and things like cross platform compatibility. I don’t want people to be able to say they can’t use Hypothesis).

But… I increasingly feel like giving things away for free is also failing pretty hard on the second criteria: I may be making the world a better place, but I’m also making it a worse one by furthering an exploitative system.

And I’m not really sure what to do about it, but I think at the bare minimum it’s time to start pushing back a little.

And I think in the spirit of that bare minimum I’m going to be moving away from licenses that are quite as permissive as I’ve historically chosen. Hypothesis is under the MPLv2, as will all future open source work I do until I decide I have a better idea. It manages to strike a good balance between the desire to get other people to use it and taking a bit of a stronger stance on things. I’m also requiring a CLA assigning all copyright to me, though I’m less convinced that that’s a good general principle.

I’m also going to try to figure out how to get paid for this. It feels… embarrassingly self-serving to think in terms of being morally obligated to get paid for my work, but I think it’s valid. I’m only able to do this free work because I’m in the privileged class of people who have enough time and money to do so. For everyone else it’s not a moral issue, it’s just a flat out issue. It may not feel like it, but providing push back on doing free work is at least partly an attempt to abdicate some of that privilege.

(To be clear: I also want to be paid to work on Hypothesis for purely self serving reasons)

This doesn’t mean I’m not going to work on Hypothesis if I can’t get paid to do so, and it doesn’t mean that the wellspring of random half-baked ideas that is my Github account is going to dry up, so maybe this is a promise without teeth. I don’t know. Getting paid for open source work outside of a few big projects (there are a lot of paid Linux devs) seems to be one of the great unsolved problems of our industry, and I don’t really want to predicate something I regard as as important as Hypothesis on solving that problem, and until Hypothesis starts to see the sort of traction I’d like it to, getting people to pay me for it might be something of a hard sell.

Which basically puts me back where I started, just with a slightly less permissive license that if I’m being honest doesn’t really affect much of anything. It’s not a solution, but maybe it’s at least the start of one.


This entry was posted in Uncategorized on by .

2 thoughts on “Some thoughts on open source

  1. Alaric Snell-Pym

    I think you’re seeing an instance of a more general problem: How can somebody afford to feed themselves and a family without either working stupidly hard, or having special skills that are rare, or being lucky?

    We live in a society where people who can’t work unpleasantly hard, don’t have special skills (which is a combination of “not being born with a book-learning personality” and “not having had rich parents to provide you with educational resources”), and aren’t lucky enough to win their way out somehow (inheritance, lottery ticket, find an unusually well-paid undemanding job, supported by a family member/spouse, etc), tend to live in decidedly substandard living conditions.

    This, in turn, makes it very hard for people to spend much of their energy on things that don’t contribute to earning cold, hard, short-term cash.

    People who are paid to work on open source either get paid less than a useful wage, so still have to make it a hobby; just one that brings in a little pocket money, one that has to play second fiddle to the day job – or they have to find a way to be paid a lot, so they can live off of it. And the latter is, clearly, quite hard, because the threshold is so high.

    So, I think the solution here is to drive down the cost of living, because the high cost of living sets such a tough baseline difficulty on this problem – not to mention thousands of other nasty consequences – that it’s going to be hard to do much about it otherwise.

    Renting a property low-end flat near where I live in Gloucester costs about £300 a month. Minimum wage is £6.50 an hour. That’s about two hours’ work every weekday, if you don’t have to pay any tax – just to rent a bottom-end studio flat, on top of which you’ll be wanting some utilities, food, and so on; and saving up so you can establish a safety buffer against unemployment. The amount you’ll need to work just to not have to worry about basic living expenses will leave you with little time and energy left to do anything else, and if your anything else requires money (say, to buy and maintain a computer), you’ll need to work even more.

    If you somehow have a large lump sum to live off of, then it’s a lot easier; you can invest months of your time, living off of savings in relative comfort, building up a new business idea until it’s profitable (or, if it doesn’t become profitable in time, ditch it before you run out of money and pursue other avenues). Without that, you have to take whatever work you can get Right Now, which is tremendously limiting.

    I suspect that if a suitably rich individual or foundation threw up a prefabricated apartment block, somewhere nobody otherwise wants to live; put a canteen and a hackspace and shared computers in common areas, and encouraged a suitable culture; and let people come and live there for free, they’d be able to pay for it in a few years from donations and by investing in the startups that would form in it.

Comments are closed.