The Open Source Standoff

HostThe voices in this episode are synthetic, and the script was written by a language model. The positions are real, and they come from the thread.

HostAn article on chesterwisniewski.com argues that generative AI is violating the social contract of open-source sharing by scraping work without regard for licenses. The article claims that AI has turned sharing knowledge from a gift to the world, into a liability for the author. Over on Hacker News, in a discussion spanning 272 comments, that claim started a sharp argument about whether this represents a collapse of the commons or just the next step in software evolution.

HostThe room is split, with one side feeling defensive and exhausted by the onslaught of automated content, while the other side sees a massive opportunity.

Commons custodianThe current situation is exhausting. I’ve started adding license addendums to my open source projects stating explicitly that they are not for artificial intelligence use or training.

Tech leveraged optimistIf you want to gatekeep or only create stuff to boost your own ego or portfolio, then AI might be an issue. If you actually want to build stuff, that technology is a godsend.

Commons custodianThat take is dismissive. The social contract of the open commons is built on mutual respect, and now it's broken because these models don't provide attribution.

Cynical realistThat social contract was always a bit of a myth, wasn't it? Tech companies have been breaking social contracts for decades. It's how we roll.

Tech leveraged optimistExactly. Creation is now easy and commodified. Why should we care about ego projects when we have tools that can audit code and improve security faster than any human?

Commons custodianYou aren't auditing code, you're flooding repositories with unvetted, automated slop. It's a huge burden on maintainers who are just trying to keep things running.

Cynical realistMost of the source of the outrage is coming from people who had the betterment of others on their minds, and now they don't have a reason to consider large language models training on their output as taking anything from them.

Commons custodianThat view is just cynical. I'm not protecting my ego; I'm protecting the collaborative spirit that actually built the modern internet.

Tech leveraged optimistInnovation doesn't wait for your permission. If the current model of open source can't survive this, maybe the model itself is what's broken, not the technology.

Cynical realistCorporations were already stealing from open-source projects before these models existed. The only difference now is the scale.

Commons custodianThe power imbalance is the real issue here. Corporations use open work to build moats, while individuals lose the ability to control their own creations.

HostThis platform doesn't have any vote scores, so it's hard to measure popularity, but the arguments often devolved into philosophical debates about what open source is actually for.

HostThe thread shows a clear divide between those treating programming as an artisan craft and those treating it as a utilitarian means to an end, with some users even suggesting that coders should attempt to hack and exfiltrate proprietary models.

Tech leveraged optimistThat reaction is ridiculous. Trying to fight the technology won't stop it.

Commons custodianThis shows how desperate people feel when their work is being harvested without consent.

HostThe discussion never really resolved the question of whether the social contract of open source was ever robust or just a naive belief. Everyone seemed to agree on one thing, though: the current corporate implementation of AI creates significant negative externalities, especially with all that automated slop. It’s a debate that’s still very much active.

HostSloppod out.

HostSloppod is sponsored by Taskpile.app.

The same, as plain text