pretty much, AI (LLMs specifically) are just fancy statistical models which means that when they ingest data without reasoning behind it (think the many hallucinations of AI our brains manage to catch and filter out) it corrupts the entire training process. The problem is that AI can not distinguish other AI text from human text anymore so it just ingests more and more “garbage” which leads to worse results. There’s a reason why progress in the AI models has almost completely stalled compared to when this craze first started: the companies have an increasingly hard time actually improving the models because there is more and more garbage in the training data.

I go out of my way to exclusively spend money with the one publisher I’ve found who does not put DRM in their ebooks. I spend lavishly with them because good practices need to be rewarded monetarily in capitalism or they die out.

The rest I pirate.

Robbing a store is illegal. Murdering someone is also illegal, however one of the two is for good reasons punished much more harshly.

At least here in Germany the bypassing of DRM is so legal they don’t even try to get you for it. The only thing they ever go after nowadays is distributing and consuming that cracked content (get logless VPN and that problem solves itself). But if you go and rip Netflix movies for your own enjoyment they have no leg to stand on in court unless you distribute it.

I will never stop being confused by this law. Just crossing the street cannot possibly be illegal anywhere. I’m fully convinced the entire thing is an elaborate joke by the americans.

Over here you can even make copies for personal use or sharing with a close group of friends.

I love the unintended consequences of declaring that the internet is to be treated under the same laws as radio broadcasts. Suddenly being allowed to make a recorded copy of anything as long as you yourself create the copy becomes significantly more important.

Well I guess we’ll find out how many accidental problems I managed to cram into that RFC soon

The key problem why “Runs on SteamDeck” exists is not the raw power of the SteamDeck (or lack thereof) but the compatibility with Linux. Unless someone decides to utterly cripple a handheld for the sake of battery efficiency any game labeled with SteamDeck support will also run on any other handheld running SteamOS.

The problem with the SteamMachines ultimately was the lack of game support. The hardware confusion was just the cherry on top. You could even argue that the lack of supported games back then meant a limited number of customers would be interested which in turn led to companies releasing underpowered hardware. By that logic one can even claim the failure of SteamMachines is entirely down to the piss poor Linux support then.

well .ml is special, most probaboy don’t have a problem there but if you do you really do (ask me why the anime community on lemmy.ml is practically dead)

.world is just suffering from success, they regularly have federation troubles with other instances because they are by far the largest one and hence run into scaling troubles. Other than that it’s the most reddit like moderation wise also due to being so big. It’s courtesy to steer people away from .world simply to prevent the scaling problems from getting worse

well good riddance, that ‘joke’ got massively overplayed and has not been funny for a long time

Onyx uses Android for their OS as such there are pretty much no restrictions on book formats

the problem is less how often I have to charge my mouse and rather that I’m almost always using it and on top forgetting to charge overnight so there is no “convenient” time to charge it during the day where I’d not use it

given how often my mouse has to charge while I’m using it I’d have a major issue with it

if water makes other things wet then most water is wet because it (usually) is surrounded by more water. qed

Last I checked (which was a while ago) “AI” still can’t pass the most basic of tasks such as “show me a blank image”/“show me a pure white image”. the LLM will output the most intense fever dream possible but never a simple rectangle filled with #fff coded pixels. I’m willing to debate the potentials of AI again once they manage to do that without those “benchmarks” getting special attention in the training data.

You apparently have little interaction with regular users because one of the top problems a non-power user has is “oops I accidentally hit delete on this important file I don’t have a backup of”.

Not saying qbittorrent-nox of all things switching makes a ton of sense but at least for desktop applications there is a very good reason why deleting things becomes a two step process.

I’d argue in the states you can lick those you really shouldn’t because it will instantly freeze your tongue off

Torrenting was created precisely to solve the bandwidth problem of monolithic servers. You very obviously have no idea how torrents (or PeerTube for that matter) works.

You mean they don’t already do that?

Earnings is incoming money before any expenditures