I’ve got a pebble in my boot. :(

Efficiency does little for your wallet and the environment if you need to buy/produce a new machine every few years.

(Not to say that we shouldn’t strive for efficiency.)

Not much different in Europe either.

Well, except for that one(?) time where a comedian registered a party to take a jab at politics and “accidentally” got voted in. Hehe.

LLM er en subkategori af en subkategori af en subkategori af en subkategori af AI. Så ja, LLM er en yderst bestemt type af AI.

Selvom et skateboard er et fartøj, så er det måske ikke det mest praktiske valg, når chefen på byggepladsen siger at der kommer nye fartøjer.

Så hvor jeg ville mene at AI sagtens kan bruges til at effektivisere papirarbejdet, og jeg ville være forbløffet hvis vi ikke allerede bruger det i stor stil, så ville jeg mene at LLM ikke hører til i staten.

Jeg håber inderligt at de mener AI, og ikke LLM såsom chatgpt.

I’m afraid to say that I too have been corrupted by VSCode.

It’s widely used, easy to get into, has LOTS of extensions, and works mostly the same across OS’es meaning it’s easy to setup by and explain to others.

The two extensions I’m missing most in other IDE/text editors would be the “Remote - SSH” extension by Microsoft, which gives unparalleled integration when working remote, and PlatformIO which, while it can be used independently in its core form, just works way better in VSCode.

Besides this, I’ll use Nano for small tasks and vi on embedded devices where Nano is unavailable, though, I’ll need a vi cheatsheet for anything more advanced than basic editing.

It’s still in beta and audio appears to not always work when streaming. Though, there’s recent activity on the related issues, so hopefully it gets out of beta before Discord alienates the regular user.

I tested it a few days ago and besides the audio problem it appears to work very well.

That’s weird. I just tested it with a friend (I’m on Endeavour, she’s on Win11, the server is VPS with Debian running the newest Synapse and Element-web). Audio works fine both ways with no mic config required, streaming is a little laggy when viewing the screen and stream next to each other, but that’s all.

EDIT: No, you’re right. Audio within streams seem to fail. I remember Discord having the same problem (hence why I use Vesktop), but if Windows also suffers this shortcoming? I’m pretty sure I remember it working a month ago, so there should be a bug report in Synapse (or element).

Element has audio issues on Linux? Didn’t notice it when I tested whether Matrix had what I needed (a month past). I’ll see if it screws up if I try again now.

Element already has desktop streaming as an experimental feature. Worked fine last i tested it. Currently planning how to trick my social circle into using it.

I also want to go check out the new TeamSpeak, it’s supposed to be a decent Discord alternative - Even though Discord originally replaced it.

Go tell 'em! Why have alternatives if we can just put all our eggs in one, holey, basket?

I think it came with a pop-up once when it first got added, and I’ve just kept removing it since.

To this day, I’m still not sure what problem it was supposed to fix, or what feature it was supposed to add.

Found the answer in the parent thread, thank you @Quibblekrust@thelemmy.club:

That’s a mention, not a tag. A tag is a private description you save about a user. Only apps have this fearure.

It’s a little weird that they took a well established term (in social media context: tag, id by which to mention a user, also known as ‘tagging’) and gave it a wholly different meaning (tag: label).

This is what we are talking about, right? Tagging others?

But the other comments seem to be talking about some kind of labelling. Did Lemmy add a new feature that I’m unaware of?

Kommissionen kan holde kæft og dy sig.

Vel og mærke, så er det dejligt at vi har den, så vi kan samle alle tosserne, og altid have en idé hvad man ikke burde gøre, men nu bliver det altså for dumt.

What kind of concussions? Knocked out, bleeding from the head, and forced to stay in bed for a few days, or ran head first into door/wall/table and started crying with a mild headache?

I feel there’s a start difference between those two, and the article doesn’t appear to specify.

Was about to point this out. I’d just go to one of my IRL friends and have him send me an E-Mail/PM/whatever while i watch him do it.

The claim above was off the top of my head, but I’ve found multiple pages of results describing the panic that ensued.

Now, Microsoft (Copilot and Github) are less than clear on what exactly is used for training, but the general consensus seems to be, that they don’t train on private repositories. Though there appears to be some confusion about this, especially regarding Microsoft’s honesty about not using loopholes (this article might be faked, I haven’t tried confirming it, though, this topic is a shit show ripe with miscommunication, misinformation, and quite a lot of confusion and fear regardless).

It appears that the specific issue I was referring to required a human error for copilot being able to train on the private repositories. Namely, some unfortunate fool temporarily making the repository public (in which case it obviously isn’t private anymore, and therefore free for grabs by scrapers). Usually this wouldn’t be a problem, since no indexer or scraper can check all of Github all at once all the time, so the chance of a briefly exposed repository being cached is rather small, albeit always there.

That said, Copilot, Bing, and Github are likely better integrated than Bing simply wasting resources on continuously scraping Github for new repositories. I personally imagine that Github saving resources by sending a signal to Bing when a repository is made public isn’t entirely unlikely (that’s something I might do, harboring no ill intentions), meaning that it is possible (though in no way confirmed) that Bing punishes briefly exposed Github repositories instantly by forever caching them.

Is this 100% Microsoft being predatory? No, obviously not, since it requires a user error to happen in the first place, and since Copilot is technically only trained on public or exposed data. Though, Microsoft learning about this rather scammy behavior and simply classifying it a “low-impact-severity” and disabling the Bing cache for humans (but apparently not Copilot) doesn’t sit right with me. I’m sure that they knew exactly which kind of data they were working with during dataset sanitation, so they could have chosen not to use sensitive data or at least inform exposed clients that they are adding their cached secrets to Copilot.

Wasn’t it revealed that Microsoft was training their Copilot on Github repositories, including private ones such as paying coorporations believing their source code to be safe and secure, resulting in secrets suddenly being made semi-public?

I feel that there were other incidents too, though I can’t remember them off the top of my head. Definitely not a place I’d recommend anyone to keep anything they love, even if they keep to best practices and don’t store secrets in their repositories.