@chemical_cutthroat
Again, all of your analogical effort presumes that an LLM is synthesizing. When I say, specifically, they generate outputs based on statistical probability it’s not at all the same as a sentient process of reiterative learning based on their available knowledge.

If you can’t get that distinction, then all the effort to respond to you will expect too much from me (personally; I wish the best to others who’d like). If you’re really sincere though, honestly it’s been best elaborated by Timnit Gebru and Emily Bender in their writings about the “stochastic parrot”. Please do have a read. https://dl.acm.org/doi/10.1145/3442188.3445922
@stopthatgirl7

@chemical_cutthroat

If I do a book report based on a book that I picked up from the library, am I violating copyright? If I write a movie review for a newspaper that tells the plot of the film, am I violating copyright?

The first conceptual mistake in this analogy is assuming the LLM entity is “writing”. A person or a sentient being writing is still showing signs of intellectual work, which is how the example book report and movie review will not be accused of plagiarism, which is very very basically stealing someone’s output but one that is not made legally ownership of (which then brings it to copyright infringement territory).

LLMs are producing text based on statistical probability meaning it is quite literally aping/replicating the aesthetic form of a known genre of textual output, which in these cases are given the legal status of intellectual property. So yes, an LLM-generated textual output that is in the form of a book report or movie review looks the way it does by copying with no creative intent previous works of the genre. It’s the same way YouTube video essays get taken down if it’s just a collection of movie clips that might sound like a full dialogue. Of course in that example yt clip, if you can argue it’s a creative output where an artist is forming a new piece out of a collage of previous media, the rights owner to those movie clips might lose their claim to the said video. You can’t make that defence with OpenAI.

@stopthatgirl7

I didn’t want to eta, so addition: what is being federated right now is readability and participation - with your kbin account you can absolutely interact with other fedi posts. But on the client-side options for a cross-platform one that can also be useable with a threadi login is still unavailable.

@adonis

@adonis the quickest answer is that the clients you’ve explored are optimized for a specific fedi software. Pre-reddit meltdown most clients developed were designed for Mastodon and its forks or close cousins whose backend are coherent like Pixelfed. This includes the Mastodon app itself. There are other microblog softwares, like Calckey - they also don’t parse the same way so most fedi clients for microblogs can’t log you in to your Calckey instance.

With the threadiverse softwares, none of them are rendered the same way as Mastodon, so that’s why you can’t login with those clients. And with the threadiverse clients, currently what’s available are software-specific - jerboa only works with Lemmy for example. Interoperable threadi clients are in heavy development though, if you don’t mind waiting. At the moment there is no Kbin-optimized clients.

Sooooo for today, if you have a Masto-flavoured account, it’s almost a given any of the popular clients can log you in. Hope that helps!

@CarlsIII aye no worries - fair ask

@Friend

@CarlsIII try click on More at the original post and select Copy to Fediverse. That’ll get you the originating url. This works for every type of post and comment.

@Friend

@macallik
Absolutely. If this is true then for the other small to mid-size instances it’s not just an existential threat philosophically but technically. They’re expecting Threads onboarding might just knock out instances because of the traffic. Might as well limit or block just for your own performance metrics.

@macallik and if you scroll down the comments, Byron from Universeodon, who did take the earlier meeting, did provide some vague points from the meeting. Relating to your point about big instances, it seems likely that FB wants to throw money at them so that they won’t become overwhelmed by the ensuing traffic (unlike the rest of us, I guess…) so they can demonstrate that the Instagram bridge (it’s an IG product) works.

@giallo @madjo @nameless_prole @stevecrox

@furrowsofar very much this point. There’s an additional issue related to (kbin) infra as well, it does fetch content and present it as tho that person posted on the microblog here and short of contacting an admin, no user-level way to delete it (since they can’t actually login)

@asjmcguire the user-level block means you the user can’t see the post. It’s still being fetched at instance level. So in terms of avoiding hate speech (political problem), good and quick step to do. In terms of overwhelming an instance with traffic (the technical problem), not so much.

@asjmcguire i was talking about kbin. The AP programming language (eta: yes it’s called protocol or standard as well iirc) is definitely old :)

The fact that kbin does fetch posts from originating instances without instances having the setting to allow or disallow is a problem once something with the computing power of meta and its userbase comes on board. Because mastodon for example doesn’t have whitelists, only blocklists.

@neonfire scheduled posting would also be good, to stagger the cross-posting imo