Defederating cuts off the whole instance. They just blocked those three piracy communities as far as I understand.

Remember that lemmy.world has to keep a copy of whatever content appears in a federated community on their servers, making them legally liable for the content. At least they just blocked the community instead of defederating.

A strength and a weakness. The strength, as you say, is being able to move to a different instance. However, the weakness is that Lemmy (the software) requires each instance to keep a copy of every federated post for its users to interact with. This means they have to host (and be legally liable for) data that they can’t police beyond blocking the community / instance.

Not really - it isn’t prediction, it is early detection. Interpretive AI (finding and interpreting patterns) is way ahead of generative AI.

There is no point to linking communities- if they are going to have identical content, just pick one or the other.

A better option would be for cross posts (using the Lemmy cross post feature) to exist as a single entity that is visible in multiple communities. This would allow for some differences in moderation which is the justifiable reason for multiple communities on the same topic in the first place.

If you aren’t paying for it, you aren’t the consumer, you are the product. It is ok if you are cool with that, but quite a few people are not.

If it won’t work in a docker container, I need a real server anyway.

The first thought that came to my mind was internal solid state power storage (good for an hour or so, but will outlive the rest of the phone) with an external MagSafe battery. Call it MC, but that’s definitely a more Apple UX than disassembling your water resistant phone.

Why do people insist that there needs to be (for example) /c/politics on every instance? Really, there are only 3 or 4 with any substantial traffic, and there are good reasons to pick one over the others, and they are the same good reasons for them to be separate.

There is a cross post feature, and the resuting post appears to be aware it was cross posted - it would be nice if Lemmy would consolidate those to one post that appears in multiple communities, or at least show you only one of them.

Again, my point is that the output is what can violate the law, not the input. And we already have laws that govern fair use, rebroadcast, etc.

My point is that the restrictions can’t go on the input, it has to go on the output - and we already have laws that govern such derivative works (or reuse / rebroadcast).

Then this is a copyright violation - it violates any standard for such, and the AI should be altered to account for that.

What I’m seeing is people complaining about content being fed into AI, and I can’t see why that should be a problem (assuming it was legally acquired or publicly available). Only the output can be problematic.

There is already a business model for compensating authors: it is called buying the book. If the AI trainers are pirating books, then yeah - sue them.

There are plagiarism and copyright laws to protect the output of these tools: if the output is infringing, then sue them. However, if the output of an AI would not be considered infringing for a human, then it isn’t infringement.

When you sell a book, you don’t get to control how that book is used. You can’t tell me that I can’t quote your book (within fair use restrictions). You can’t tell me that I can’t refer to your book in a blog post. You can’t dictate who may and may not read a book. You can’t tell me that I can’t give a book to a friend. Or an enemy. Or an anarchist.

Folks, this isn’t a new problem, and it doesn’t need new laws.

The fediverse is the name for services that use ActivityPub - a communication protocol. What you are saying is like saying “tech companies, banks and regulators need to crack down on http because there is CSAM on the web”.

Yeah, let’s talk about self driving cars…

In medicine, when a big breakthrough happens, we hear that we could see practical applications of the technology in 5-10 years.

In computer technology, we reach the same level of proof of concept and ship it as a working product, and ignore the old adage “The first 90% of implementation takes 90% of the time, and the last 10% takes the other 90%”.

As noted elsewhere, do everything you can to avoid handing your card to anyone.

Use tap to pay wherever possible, then chip - neither of those methods give the card number to the merchant. Do not swipe unless you absolutely have to, and then inspect what you are swiping to make sure nothing is attached to the card reader.

For online purchases, do everything you can to avoid giving your card number to anyone - use ApplePay / GooglePay / Amazon Pay / PayPal etc. wherever possible. These can be used to put charges on your card without giving your card # to the merchant. These are one-time authorizations (unless you explicitly identify it as a subscription / recurring charge), so they can’t reuse the transaction token they get.

More RAM will always give you more longevity than more CPU. It is my policy to never buy the default configuration of any Mac, because they always have too little RAM.

The cost savings is a supply chain wet dream.