We now note that that we do not waive our attribution requirements for Spritely Goblins, which are present in Apache v2 (and for that matter, CC BY, CC BY-SA, and most FOSS licenses) https://codeberg.org/spritely/goblins/pulls/998/files

@copiesofcopies no, it is not a value judgment, but it is similar to a debate over whether generative AI art is art. it's about the relationship between an underlying meaning/perspective and a produced output; LLMs simply do not have it. "slop" also encodes the types of issues and discrepancies that simply don't happen from sloppy human-produced code and art and that are, by design, tricky for humans to detect.

if an individual output at face value is all that matters, then no, this is not interesting. "slop" is associated with a context. the myth is that serious people can pretend there is no context and have an eye for the best outputs in a vaccuum. this is anything but serious.

@bkuhn @joeyh @cwebber @trashheap

The web server was not written in one go as a whole.
It was put together patch by patch
It was therefore A Patchy Server.

Being conscious of assorted things does not entirely rule out occasional flashes of humour, accretions of language, and mnemonic conveniences.

I hope.

@bkuhn @joeyh @cwebber @trashheap

(I'm told)

@bkuhn @joeyh @trashheap The name bothers me immensely. I'm using AFLv2 for our projects mainly because it's the most studied license of its approximate properties. But at one point I registered patchable.org because I was planning on making a fork of the license calling it the Patchable Public License, mostly just being a rename, to force the issue. However I was told that the Apache Foundation would be changing the name any day now and so I didn't continue with that effort. Years later, they still haven't.

@cwebber @LunaDragofelis it annoys me that ppl would consider a lack of enforcement to be a lack of interest;

i dont want to enforce them because the normal means to do so is via state violence and i might like attribution but im not about to hurt someone over it.

i would however say, leave an issue or message the guy or things like that, about not attributing, generally tell everyone "hey they didnt make that" which ideally? would get out everywhere and no one would really beleive then saying "i made this"

from a certain perspective i dont subscribe too; this could be considered a kind of "enforcement" i suppose,

tbh alot of instances of it are also a mistake somewhat, like not clicking fork or whatever so then it just doesnt include where it came from ..

anyway its not a copyright thing; its just basic courtesy really -- like pirates usually dont even claim they made the thing.

@yantar92 @cwebber

.. no the collection of training data is a problem because of the massive amount of strain it places on infrastructure set for the commons that makes it impossible for anyone else to access, and becuase its appropriating the commons then selling it back to you,

which still has nothing to do w copyright.

@aparrish @cwebber yeah .. it really should be social courtesy and done on a "best effort" basis ..

also just .. fuck laws

@Li @cwebber Not all the training data is putting strain. There are training datasets that were crawled respecting robots.txt.

I am most interested in the scenario when there is a training dataset (crawled once without hitting servers repeatedly), and used for open weight model.

Open data+weight still does not address attribution. That's why my question.

@Li @cwebber @LunaDragofelis the contrast to piracy is apt. I've seen anti-AI libs compare LLMs to piracy, and I've seen AI shills go "well, if you support piracy, how can you hate AI?" But one of the big differences is attribution. I don't mind people ""stealing"" my work to consume it. Art is made to be shared. I *do* mind people pretending they--or a statistical averages machine--made something that I poured my heart and labor into. Attribution matters, IP laws are still capitalist bullshit

@Li @cwebber Sure. But I am asking about very specific issue - attribution.

@raphaelmorgan @cwebber @LunaDragofelis another part of this is that it treats a corporation doing something the same as a individual doing it, which has different dynamics at play; but your not allowed to point it out

alot mention an idea of it being for-profit, but thats not really the whole of the issue there and i think not a very good look at things.

another issue is that AI is not piracy its taking the thing and trying to make some product their trying to sell .. piracy is just the thing distributed by someone else .. whos not approved .. and maybe modded to bypass DRM.

but to be honest i tend to avoid; copyright related reasons for saying AI bad for this reason; its not like thats all you have, its also used for mass survailence, the military are training it to murder people, its good at generating propaganda, it encourages reliance on it giving whoever made it lots of power and; also generally training is very intensive on infrastructure, but yeah.

@yantar92 @cwebber no your using someone saying attribution is an issue; then going "oh if we just do attribution then its acceptable" no, its not because its not the only issue

@Li @cwebber I am not interested to discuss other problems with LLMs here. I am only looking to see about attribution specifically.

@yantar92 @cwebber cool? then your question is dishonest- they still dont exist in a vaccum despite your insistence to say that they do,

@Li @cwebber Yet, the question about attribution can be addressed independently. I am not sure what is dishonest about this.

@yantar92 @cwebber your asking if you just add attributiion will ai be acceptable. the answer is no- because of reasons other than attribution, but then you refuse to address those. and say your "only about attribution"

how is that not dishonest?

@Li @cwebber @LunaDragofelis I think the for-profit point is definitely part of it (for me) but not nearly as much as some say. Like, I would still be upset if an individual took my work, modified it a bit (if only to remove my signature or what have you), and claimed they made it. But I wouldn't feel nearly as powerless or angry as when a corporation does it.

And yeah, talking about non-copyright-based reasons is very important, because copyright is yet another capitalist system oppressing us

@Li @cwebber I did not imply that solving attribution automatically resolves every other issue. I am asking about specific thing - AI not propertly attributing its sources used for the output. That's a concrete problem. I asked if attributing to *all* sources is good.engough to solve this particular problem.

@Li @cwebber I conclude from your answer that referring to all sources, even not really used in the specific output is practically solving attribution (at least, from your perspective). I'd like to point out that may or may not be enough for other people who are looking for their work to be attributed.

So, no, I did not say "consider one issue, then solve that issue". What I proposed may or may not be a solution.

@Li @cwebber Let's me give you exaggregated example - what if every slop code starts to attribute *every possible website on the internet*. Or even non-slop code using a code licensed with attribution does not attribute just to that code, but also to 1000 other codes (that have very little to do with this non-slop code). That feels not exactly right, although might technically fulfill the license. But maybe that's just me.