you won't win at copyright

For women, the need and desire to nurture each other is not pathological but redemptive, and it is within that knowledge that our real power I rediscovered. It is this real connection which is so feared by a patriarchal world. Only within a patriarchal structure is maternity the only social power open to women.

Audre Lorde, The Master's Tools Will Never Dismantle the Master's House

The open software ecosystem is not in a good way. We’ve seen escalating abuse of our commons by the very entities we were trying to provide an alternative to. Our default tool against this, copyleft, is repeatedly showing itself incapable of rising to these challenges. understandably, many find it difficult to see a path forward. In this context, many are attempting to double down on copyleft, with some even going as far as making their work proprietary in an attempt to maintain control of how it’s used. Herein I hope to share a different vision.

some context on copyright

A comprehensive history of copyright is out of scope for essay, but I think it worth highlighting a critical point in its conception:

The Licensing of the Press Act 1662 in England was drafted in the wake of the printing press to provide publishers with a monopoly on printing registered books. Its purpose was to protect publishers from competition, ensuring their profitability, and to provide a means for censorship. Notably unrepresented in this consideration are authors, who could not register books themselves. The act was temporary, and in 1694, unable to secure a continuation from parliament, publishers were in need of a new approach. It was only then that they chose to highlight authors’ rights, to make the proposed legislation palatable to a public tiring of their greed. Yet, they did this knowing that even with authors holding these rights, they could do little with them other than sell them straight back to the publishers from a disadvantaged bargaining position.

The publishers have spent the centuries since clawing back as much as they can from this small concession. Authors have hardly noticed the change. Before copyright, creative workers primarily made their living by commission and patronage. Today, creative workers continue to primarily make our living by commission and patronage.

That’s not to say we haven’t had our wins. Copyleft was a big one. But I’m reminded of Lorde’s wisdom:

Those of us who stand outside the circle of this society’s definition of acceptable women; those of us who have been forged in the crucibles of difference—those of us who are poor, who are lesbians, who are Black, who are older—know that survival is not an academic skill. It is learning how to take our differences and make them strengths. For the master’s tools will never dismantle the master’s house. They may allow us temporarily to beat him at his own game, but they will never enable us to bring about genuine change. And this fact is only threatening to those women who still define the master’s house as their only source of support.

Copyright is one of the master’s tools, that’s carved into its bones. It was intended to protect power and capital, yes, but more importantly for us the means by which it does that protecting is also power and capital. One enforces one’s copyright by making threats, and hiring lawyers to back up those threats, ideally very expensive lawyers. Under that superficial enforcement lies the state, setting the legal framework. The state is made up of politicians and bureaucrats who can be plied, bought, or blackmailed, and derives its power from its ability to imprison or do violence to you. In the long term, I don’t think it possible to use such a tool without perpetuating the logic of power and control it’s built on.

the form of our work

Lorde’s advice is to take our differences and make them strengths. I think there’s more to that I haven’t seen how to apply to our trade yet, but I can at least take a superficial reading:

I remember as a child my parents were paranoid that my sister or I might get tattoos or unconventional piercings. This wasn’t because they had any particular dislike or prejudice toward such things, or even that they were worried about discrimintation or prejudice from individuals. Their worry instead was that it might make us “unemployable”. We routinely cut off parts of ourselves, to make us generic and palatable enough to be fed into the maw of industry, and we advise others to do the same. I’m not sure I can advise everyone to stop cold turkey in good conscience, as most of us do still rely on an employer to make sure we’re fed and housed, but I’m not willing to do it to myself outside that context, and even within it I find it necessary to push back on some of the conventions of professionalism. I’m definitely not willing to do it to others.

I’d like to praise Soatok for the example he’s set for me on this front. He’s talked before about regularly getting complaints that his blog is too “unprofessional” to share at work and, like, good. Fuck your company.

If you don’t want your project used by a company, filling it with personality seem a more promising way to achieve that than a license. Of course it’s not perfect, your work can always be whitewashed, but if the roots go deep enough those stains can be hard to shake. In any case, it’s more work for the company downstream, especially if they want to track your changes.

Who would have guessed a decade ago that sarcastic commenters on Reddit would be such a big stumbling block for LLM training.

the structure of a technology

I recently had to implement OIDC auth in a Rust service at work, and so had the pleasure of dealing with the trait soup that is the openidconnect crate. It crossed my mind that I could write a better version of this crate, or at the very least some more serviceable documentation. Judging by what others have to say in the crate’s issues, there’s demand for it, so I’d be scratching my own itch and helping other programmers out. Myself of even a couple of years ago would have done it. What I wouldn’t have stopped to consider then is why we’re being asked to implement this protocol, and who is served by the protocol becoming easier to implement. It’s a tool for centralisation, lock-in, and surveillance, and I really don’t need to do Google’s legwork for them. Every bit of extra friction in using it is a good thing, in hindsight even my struggles were good. The extra implementation time is a cost my employer paid for making a bad decision, and the compile time burden of all those traits is another cost they will continue to pay until they rip the feature out. If the openidconnect crate didn’t exist, they probably would have given up on the feature entirely.

It’s not always possible to know how a technology will be used before it exists, but we can make some pretty good guesses. A technology like BitTorrent functions to advance freedom of information, and because that purpose is set so clearly in it’s structure, it has proved resistant to subversion. Indeed it seems resistant to attempts to do anything else but share information, including its creator’s attempts to monetise it.

mutual aid

This is not a only a hobby for most of us, but also how we sustain ourselves. An undercurrent in the complaints about the state of open software is that it’s hard to make a living writing open software, but I think that’s somewhat by design. The thing we came here to oppose was the control companies exert on the use and development of their software, and the externalities that exerts on the users; they do this mostly as a means to extract money. If we were to turn our goal towards extracting money, it would indeed make sense to start exerting more control on the use and development of our software, but at that point we have to question why we’re here and not in a startup lab.

The old advice to have a day job and do this on the side still carries weight, but I dream of a world where sustenance is not synonymous with profit. I don’t see why I shouldn’t be able to write software that benefits others, and have others do work that benefits me, without either feeling the need to control and capture. We have achieved that within software, but I can’t eat software.

When most people today hear “mutual aid”, they think of charity, and charity is often an ugly thing. Those with many resources “giving” some away to those with few, to ease the suffering caused by their lack of resources, while ignoring the reality that the lack of resources is often solely due to enclosure of the resources by those with many. It offers no real answer to the injustice it purports to address, and instead primarily functions to assuage the guilt of the perpetrators, and give them an excuse to avoid having to contend with the root of the problem. We can do better.

Mutual aid is about seeing systems that produce bad outcomes and working together, whether we are directly affected by those systems or not, to address those outcomes systemically. It is about involving those “in need” (which includes you and your need to eat) in how those needs are met. There are people out there doing things you need, with expertise you don’t have, who would benefit from some software help. The connections you build with this kind of work often turn out to be worth more than a paycheck, and it gives you a chance to practice making software that molds to its users’ needs rather than the reverse, which has given me a needed ray of sunshine amid the bleak outlook this industry engenders on our profession.