- many controls are now functions/variables (that's good!), and some change name, for example `{% csrf_token %}` -> `{{ csrf_input }}`
- calling methods without parenthesis does not work (that's good!), for example `query.all` -> `query.all()`
- in tests` response.context` is replaced by `response.context_data`, which is not set when calling `render` directly (need to return a proper `TemplateResponse`)
- template folders need to be renamed from `templates` to `jinja2`; there may be a way to change this behavior, but i did not find it, and this change is written so small in the docs i lost an entire day over it
Well, i guess that's what i get for trying to sympathize with someone i disagree with. I have also been this person who doesn't fit in a group but only realizes too late. My bad for thinking we were having an interesting and honest conversation.
Sure, i linked the docs about private repos specifically. Public repositories of FLOSS are the mission of codeberg, minus slopware and shitcoins (the subject of this thread). I thought the FAQ was pretty clear about that, but if that's not the case, i guess PRs are welcome to improve it.
> You say yourself you didn't even see all the comments.
Yes, and i'm not claiming you're wrong in your feeling that some people were excessive about this. Though i could argue that LLM deployment has been excessive too and this is a very minor comeback compared to the pressure people have faced to adopt LLM tools (eg. at work) or let new AI datacenters ruin their neighborhoods and electricity/water networks.
> it's in-group implicit consensus masquerading as formal process
That's indeed a problem in some circles. However, i fail to see what's the alternative. Letting a debate open for years before opening a formal vote sounds reasonable. That's far from the entryist tactics of building consensus behind closed doors in a secret affinity group then faking organicity and taking everyone by surprise with an "urgent" formal decision. I'm interested if you have feedback on how to organize more "democratically".
> I'm glad to be departing from such people, but no thanks for wasting my time.
I'm glad you feel relieved, and i'm sorry you lost time and energy in a collective where you don't fit in the end.
Not in fact. On systems with permissions/privileges, you always return a 404 when a page does exist but the user doesn't have permission. Because sometimes leaking the existence of an item is a problem in itself.
Overall: there are humans performing the moderations. I trust them to have good judgement. This is not a witch hunt proposing to delete every repository where Claude has ever committed.
Personal interpretation: i think, from a very outside perspective (but that's what i'd do if i were in their place), that they'll inspect projects sorted by CI minutes consumed, then individually seeing if that's useful CI for useful projects (eg. fuzzing for tor or whatever) or if that's just slop CI for slop projects running countless steps on 100 commits an hour for software that's used by exactly 1 person (or even 0, who knows). From there, i'd start by failing the CI (and pushes) with a message inviting the user to read their mailbox, and send an email saying the user has been detected as malicious and should move away from codeberg, that existing code will remain available for a year, but no push/CI/issues will be permitted.
This will not remove 100% LLM-assisted project (which then again, was not the case), but might reduce codeberg resource usage by 10-99% depending on how bad the situation already is.
> run agents on it and hide the "slop"
I'm not sure how an agent would be able to hide that it's committing 100 commits an hour with 10k lines each, adding and removing tests as a mere inconvenience. But sure, let's assume so. That's just going against the rules, so the user may get banned.
The first link has been removed (or may be private).
The second link and quotes sound bad, but placed in context they actually sound more reasonable arguments to me.
The "limits of consent" was somewhat misplaced, but in reply to someone arguing in bad faith about bad-faith interpretation of the rules. So it's not comparing LLMs use with rape, but comparing testing the limits/interpretations of rules with rape.
The violating of group values (which in my understanding has semantically nothing to do with rape/consent) was in reply to someone saying if you violate the rules a little, "noone will say anything" (so it was on point).
The nazi analogy was not an analogy at all. Noone implied LLM users are nazis. It's about how to interpret what is slop: the general rule of thumb that the only people worried that "anti-nazi" rules in communities are a slippery slope are the nazis themselves. The author was suggesting the same rule of thumb would apply to slop: those worried about the rule's interpretations are the ones making the slop. (a point i don't entirely agree with but more reasonable than the quote out of context)
All in all, i must say i cringed a little reading some discussions on this topic. And maybe i've missed the worst you witnessed due to deleted posts. But maybe you were also having a strong emotional reaction and the comments taken from afar are not as dramatic as you suggested?
If you're a vegan as your nick suggests, i would suggest to reconsider this point. Capitalism was already bent on sending us towards a 6th mass species extinction (starting right under our noses), but so-called AI has changed the scale and speed of the curb drastically in only a few years. See for example this article for recent Google stats:
Recent AI datacenter projects such as Stratos (40000 acres) are on a scale of industrial destruction my mind fails to comprehend.
I'm not 100% vegan, but i'm 100% "AI-vegan", and i definitely would recommend you considering it.
PS: All i said was just about the environmental impact of LLMs. I suspect the human impact to be just as bad (cognitive impairment, accrued exploitation, etc).
I'm sorry but you are completely out of touch with reality. This is a non-profit calling on their members to vote. Non-members didn't have a say, so a random member posting on Mastodon is not going to influence the results in the way that you imply (i.e. trolls raiding an online form).
Also, despite what your remote view suggests, members of non-profits usually have political opinions. Codeberg was founded for political reasons by people with political opinions: the members aren't waiting for someone on Mastodon to have an opinion.
Also, apart from this, people having opinions and encouraging others to defend their opinions is quite normal democratic process in a non-profit. Nothing to see here.
> Given they'll kick you off the platform if you're not aligned with FOSS views and act accordingly, it seems strange to have associated Codeberg with "freedom" at all.
It's different conceptions of freedom. For some, the freedom to oppress is not freedom at all, and building non-free software is oppressive.
> Side-note, but there is something that keeps popping up as an overlay on that page but I never end up being able to see what it is, as I click continuously as I read
Yes, clicking makes it go away! It's an overlay that opens when you leave the page, saying you should really not run Javascript programs sent by the web server because they can do pretty much anything.
That would be a commercial activity. Codeberg is a non-profit host. I'm not familiar with german law but i assume non-profits can have some commercial activity (as long as they don't compete with commercial actors), but that would open an entire can of worms.
I think the path they chose is better: stay on their mission to provide free services to FLOSS projects who need it.
I don't know you, but my feeling is you have the wrong analysis grid here. This is a non-profit taking a stand for what's right in a world where everything's wrong. The discussion has been ongoing for several years, and the rules introduced are not vague, and do not target random/unpopular projects.
I personally have trust in the human community behind codeberg. I understand if you don't. But my question is: why go with a non-profit host if you're not going to get involved in the governance? Of course governance is always happening in a project. For me, that power is shared and decisions are transparent are a criterium to trust codeberg even more.
Freedom of expression does not mean my non-profit has to spend its resources to give you a platform. I'm not a member of Codeberg eV, but i strongly approve their decision. We need more hosts with a mission, especially when it comes to AI and promoting human-crafted resources.
All IT infrastructure providers are activists. The question is what is their politics? Github pushes for centralization, AI slop, and gathering code for the military-industrial complex. Twitter pushes for techno-fascism under our great and only leader Elon Musk. FSF pushes for liberal left-wing copyleft.
If you don't see the politics behind economic actors, i would recommend taking a closer look.
I'm strongly anti-AI, but i upvoted your comment because it's the most reasonable one (in opposition to the AI ban) i've read so far. I haven't read those mastodon threads, but that sounds like a weird discourse. Are they from codeberg eV members or random passers-by (like me)? Would you mind sharing some links to some of those resources?
Bad example, you obviously didn't read the codeberg ToS. Codeberg banned projects that « mostly consist of code written by "generative AI"-tools ». subsurface is apparently quality software with different contributors, and in the recent commit history i only saw a handful of commits assisted by LLMs.
subsurface is not slop and is therefore not targeted by the codeberg ToS update.
A few notables differences:
- many controls are now functions/variables (that's good!), and some change name, for example `{% csrf_token %}` -> `{{ csrf_input }}`
- calling methods without parenthesis does not work (that's good!), for example `query.all` -> `query.all()`
- in tests` response.context` is replaced by `response.context_data`, which is not set when calling `render` directly (need to return a proper `TemplateResponse`)
- template folders need to be renamed from `templates` to `jinja2`; there may be a way to change this behavior, but i did not find it, and this change is written so small in the docs i lost an entire day over it