Can we please agree not to call apples "oranges" just because we like how "oranges" sound, and we also happen to like apples?

These are different kinds of fruit, and calling one using the name for the other is simply misleading.

Same with "anti-capitalist licenses" being called "open-source licenses".

They're not. They're different. The difference matters - if it didn't, there would be no reason not to use open-source licenses!

@bookwyrm is not "open source" (doesn't mean bad!):

I am of two minds about "anti-capitalist licenses" that started springing up in random places.

On one hand, I understand why people feel we need them. I, too, take precautions against software I write being (ab)used by Big Tech.

On the other, I feel it might be counter-productive.

AGPL, I believe, provides enough protection against Big Tech using software licensed using them. At the same time, having clear rules of licensing compatibility is powerful -- it allows us to collaborate and build together.

Finally, all "anti-capitalist licenses" I've seen so far use very broad terms that can easily be interpreted in a lot of different ways. They likely do not offer the protection they claim to offer.

In the end it might turn out that by creating a myriad of "anti-capitalist licenses" that are incompatible with each other and with the larger body of FLOSS code, we handicap ourselves - by making it effectively impossible for different projects to be used together to build more complex and more useful systems.

@rysiek The anti-capitalist licenses are interesting. I think someone should make a similar license, call it "the coop license" and get it vetted by a copyright lawyer. If such a thing existed and carried real legal clout (i.e. could be used in a court case without getting laughed out) then this would be a more favorable tool against the status quo.

We definitely do need to be finding better ways so that BigTech doesn't keep stealing from the commons and grifting on our collective labor, and so that we can build up counter-organizations which are focused on real needs.

@bob @rysiek

Yeah, actually I like this better than kosher free software and open source licenses. Free software can't do nothing to avoid the big tech monopolies world we live in. Rather, it was an enabler. If you want to change that, you can't use the same set of tools. I hope it gets legally standardized, as Bob says.

@tagomago @bob I remain unconvinced.

It really feels like playing the "wack-a-corporate-ownership" game is a losing strategy (I'm skewed, worked too long for people investigating byzantine corporate ownership structures...).

At the same time, AGPL seems plenty effective against both Big Tech and techbro startups, is well-understood, vetted by lawyers, and has a huge body of code already licensed under its terms.

@tagomago @bob
I also note that all "anti-capitalist licenses" I've seen so far curiously miss the single most anti-capitalist measure possible and available: a requirement to share back any modifications made.

Which is something I don't get. Clearly people who write them are opposed to capitalism. Can there be a better way to combat rampant capitalism than, well, sharing broadly?


@rysiek @tagomago @bob Agree, “Share Alike” must be a core tenet of any post-capitalist tech.

· · Web · 1 · 2 · 6

@aral @rysiek @bob

Turns out knowledge is not the power source in capitalism. Of course, without knowledge free distribution, it gets worse. But in capitalism, capitalists have a huge advantage given the uneven distribution of material resources.

How does AGPL or any free software license prevent that?

@tagomago @aral @bob by undermining their very business model. AGPL is *toxic* to cloud-based Big Tech, just as GPL was toxic to the previous generation of Big Tech.

Software is not knowledge. Software is a tool. By creating useful tools that are unusable by capitalists we build our common toolset and expand our abilities, without providing these tools to the capitalists to use against us.

Anti-capitalist licenses are trying to do the same, of course, but in a way that undermines this community-building.

@tagomago @aral @bob think about it: an "anti-capitalist license" would not have given us Linksys WRT-54GL OS's code, and thus would not have jump-started the ecosystem of projects focused on embedded devices, like OpenWRT and the like.

If Linksys had used code under an anti-capitalist license, the lawsuit (if any) would result in Linksys being forced to stop using the code and pay compensation to the original developers, that's it. No code would have been published. No common toolset made available.

@rysiek @tagomago @aral If I understand it correctly, if Mastodon used an ‘anti-capitalist licenseʼ, there would be no contributions from pixiv/pawoo developers

@rysiek @aral @bob

Tools and tools' improvement are the product of knowledge, it's the same thing to me. Anyway, let's not turn this into a logomachy.

From what I know, I don't think AGPL is not usable by capitalists, it's just that they can't exploit any labor invested in the tool to sell it as an end, as a capitalist product by itself, but it can be used freely as a, well, tool, for their capitalist goals. Correct me if I'm wrong...

@tagomago @aral @bob technically it can. Practically it is so toxic to their business models that Google outright bans not just including AGPLed software in their own projects, but also bans their employees and contractors from even *having any AGPLed code on their machines*!

Straight from the horse's mouth:

> Do not install AGPL-licensed programs on your workstation, Google-issued laptop, or Google-issued phone without explicit authorization from the Open Source Programs Office.

@rysiek @tagomago @aral Which is pure fear-mongering on their part. I’ve worked with the legal department of a FAANG-level corporation on determining what different licenses actually mean because they tried to forbid the use of GPLed software, which would’ve meant a lot of workstations and all our servers 😋 Bottom line, which should surprise no one: None of the *GPLs places any burden whatsoever on any user of the software (which is why it’s bullshit to put them in the license-agreement popups, you don’t have to agree to anything to use the software). They only come into action once you distribute (or SaaS, in case of AGPL) it to third parties (no, your employees are not a third party), and even then it’s only really relevant if you modified it, because for unmodified software you can just point people upstream. The whole concept of the *GPLs boils down to “If you make a change and let others use it, you have to also give them the source (under the same conditions)”. No change? Nothing to do. No “others”? Nothing to do either.

@aexiruch @aral @tagomago you're right, of course.

That said, I don't see it as fear-mongering, I see this as risk-avoidance. If an employee/contractor has AGPLed code on their workstation it might accidentally be used in a project that then goes live and (from Google's perspective) "taints" a lot of code.

That would expose Google to lawsuits from users demanding they release all of the "tainted" code, which is a huge risk if you're a Big Tech surveillance capitalist.

So, they play it very, very safe.

@rysiek @aral @tagomago Fair, it may be actual fear, instead of deliberate fear-mongering. In our case it indeed was some lawyers getting beFUDdled themselves. Took a while to show them the intentions behind the licenses. I think they got it when I showed them the WTFPL, Beerware, and Unlicense on actual useful code 😉

@aexiruch @aral @tagomago haha!

I always loved the JSON PL. JSON PL is/was very similar to the whole "anti-capitalist license" scene in that:
1. it wanted to do some good
2. it did it in a way that made it a non-free license

But it did say "The Software shall be used for Good, not Evil." and that made IBM's lawyers queasy.

So they asked for an exception, and the author of JSON gave it to them: "IBM and it's customers are also allowed to do evil with the Software".

That's some awesome legal trolling!

I perceive some misplaced fears in this description
it's not like AGPLed code automatically makes itself interactive with remote users. if you use it, it's likely that it has features to interact with remote users (and point users at sources through that interface), and if you modify the program, you must make it point to corresponding sources as well, but imagining an accidental situation seems misleading at best

furthermore, lawsuits from users would likely be quickly dismissed. users don't generally have standing to sue for copyright infringement, which is what a failure to comply with the license terms amounts to. only copyright holders can enforce their copyrights over the work.

anyway, IANAL

@tagomago @aral @rysiek @bob To answer that, you first need to ask, why capitalists could profit so widely from GPL-code without sharing back.

The answer: When you put GPL-code in your own webservice, it acts like the weakest of BSD licenses, because you never propagate your code, only results.

The antidote is AGPL which fixes that hole.

For Javascript that’s different, because you actually ship it. Why do you think do corps go for lax licensing there, but don’t care about GPL on the server?

@tagomago @aral @rysiek @bob AGPL prevents part of that huge advantage, by forcing them to share.

Decentralization prevents the other part of of the problem. But there the regular GPL is effective, so big corps can’t take this route as effectively as the webservice-route where they can profit off free labor without giving back by circumventing the terms set by idealistic volunteers.

Sign in to participate in the conversation
Aral’s Mastodon

This is my personal Mastodon.