It is worse in HW prototyping where sometimes loose wire is all over the place

Holy shit that is one heck of a thing to do.

TIL about bus factor

To be honest, that is one rad use cases. Games usually don’t have privacy concern as much as day to day usage. The problem is the fucking recall works by denylist instead of allowlist

Instagram by Meta has started using user generated content for their training data. Naturally users are upset about this and are flocking to a new platform named Cara made by an artist which has principle and are actively fighting for artist copyright. Cara uses Vercel to host their platform. The problem is that the sudden influx of users means bigger bills to pay to handle those traffic (I think they are billed around ~$96k). As of now, it is unclear how they will proceed to pay the bills. The creator of Cara didn’t want to compromise his stance by accepting any money willy-nilly, but also acknowledged that angel investors are not easy to find.

Most probably, yes

Transferring ownership of the account also transfers the game license owned by the account. Still upset publisher

I didn’t specifically comment on GNOME ma dude. Just the other commenter that said most people don’t want choice. I think it’s not “choice” but it should be “choose”. Having to choose can indeed become confusing, so there should be a default or pre chosen choice. Having no choice means you are locked in. Hence my comment. I am having no problem with people not wanting to choose, but people that do not want a choice is when I am starting to have a problem.

What the fuck are you talking about? Do you think the license being used is not legally binding? What constitutes as legally binding to you?

If they don’t want to choose, that’s fine by me. But why tf they didn’t want choice? They could just stick to whatever is the default and let others who wanted different choice have their way.

Gold test

“What if the test fail?”

*Write a test for the test

deleted by creator

Why… why is it more secure? Does it mean AI training is actively abusing copyright law? And this is more secure because they can hide it better?

I wouldn’t dare call it mislabelling since there is no precedent yet. Just the other day a judge ruled AI generated CSAM is still CSAM. If it can be proven beyond a doubt that an AI output comes from copyrighted works without proper license, will that AI violate the copyright? Also, will AI count as derivatives work from the training material or will it be treated like software compiler? I think a lot of our current legal framework is not up to speed to answer those questions. So I would not call it useless nor misleading.

Also, lemmy doesn’t have EULA as far as I am aware of so the license of the content hosted on the instance is by default unlicensed. The user just notifies that to whoever wants to use their comment for whatever purpose, must abide by those licenses.

This also highlights the problem of extrapolating from a single data point.

Damn, this looks handy. Thanks for sharing

Android is open source so it can be customized. If a country decides to use Android I’m pretty sure they can spend a budget to modify and vet the source code being used by their device rather than re-inventing the wheel.

WebRTC can be used, and there is public STUN or TURN server

Alright, try it with cat