BakedCatboy

joined 1 year ago
[–] [email protected] 10 points 1 month ago

Not only is there the issue of getting approval from the video creators, there's the issue that most PeerTube servers aren't ready to handle a huge influx in uploads, as this would likely be a bulk operation.

Personally I think mirroring YouTube content would be more viable once ActivityPods lands and is integrated with PeerTube, which could potentially let you self host your PeerTube account data while still being part of a separate "home instance", which would greatly help with the storage issue for PeerTube as we could all bring our own storage.

[–] [email protected] 4 points 1 month ago (1 children)

Previously, a Meta executive in charge of project management, Michael Clark, had testified that Meta allegedly modified torrenting settings "so that the smallest amount of seeding possible could occur," which seems to support authors' claims that some seeding occurred.

It really seems like their defense is coming down to "we may have seeded a little bit but there's no proof"

[–] [email protected] 1 points 1 month ago

Sure, but you also don't need to give them full benefit of the doubt just because that's how the court operates. It's a perfectly reasonable stance to not believe their claim that they loopholed the law by not seeding, which I don't think is contradictory with supporting piracy. And comparing the mass ingestion of human creative work into an exploitative AI model to an individual person pirating for human consumption as if someone who is against one must be against the other is absurd.

[–] [email protected] 1 points 1 month ago (2 children)

My argument is that just because the courts may give Meta the benefit of the doubt, it doesn't mean that you need to as well. It shouldn't be any surprise to you that you're getting the response you're getting here when you seem to be bending over backwards to find any excuse to give Meta a pass.

And no - wanting Meta to be fully investigated on the basis that they most likely did break the law has no bearing on wanting to oppress the enemy lol.

[–] [email protected] 1 points 1 month ago* (last edited 1 month ago) (4 children)

I'm not a court so absent any actual evidence from Meta, I can assume whatever I want. Meta can suck a dick.

[–] [email protected] 1 points 1 month ago (6 children)

It's a distinction without a difference, because there is no reason to believe Meta's word that they blocked seeding when downloading. So whether it's always or usually makes no difference, because in either case, Meta should not be given the benefit of the doubt.

[–] [email protected] 3 points 1 month ago (8 children)

Both things can be true at the same time - you can get a letter for leeching only AND usually when leeching you are also seeding. I don't know what your issue is with that statement.

[–] [email protected] 2 points 1 month ago (10 children)

I don't know if this is news to you or not, but while you are leeching, you are also seeding.

[–] [email protected] 6 points 1 month ago (12 children)

I am the reader and I have made the determination that you are wrong. Plenty of people get letters for leeching only - just your presence in the swarm is all it takes, and that's all they check for before sending you a letter - at least in the US.

[–] [email protected] 2 points 1 month ago* (last edited 1 month ago)

Yes for gaming, but for LLMs I've heard that the bandwidth limitations of using system RAM as vram hurts performance worse than running on the CPU using system memory directly, since smaller models are more memory bandwidth limited.

I've never tried to run AI on an igpu with system memory though so you could try it, assuming it will let you allocate like 32GB or more like 64GB. I think you'll also need a special runner that supports igpus.

[–] [email protected] 4 points 1 month ago (2 children)

If you have a lot of RAM, you can run small models slowly on the CPU. Your integrated graphics I would guess won't fit anything useful in it's vram, so if you really want to run something locally, getting some extra sticks of RAM is probably your cheapest option.

I have 64G and I run 8-14b models. 32b is pushing it (it's just really slow)

[–] [email protected] 3 points 1 month ago* (last edited 1 month ago)

One related thing to watch out for is the state table size - one of my old cheap routers back in the day showed how full it was and it was hitting 100% a lot and seemed to grind the network to a halt when it did (I was in a house of 5 young people with lots of devices and multiple people torrenting behind a cheapo Netgear running ddwrt). That's what lead me to switch to high end or x86 based routers. Being able to see the state table stats really helps to know how likely it is to be a problem, it's so big when using opnsense on an x86 box that I don't think it ever goes above 1% now.

Edit: now that I think about it, if your VPN is working I wouldn't expect any states related to peer connections to show up since your router won't be NATing them, I guess I was just bold back in the day because it was a huge problem then.

view more: ‹ prev next ›