This particular thing IS a common-sense technicality. Without it, every new instance, the second it came online, would start to try and mirror THE ENTIRE FEDIVERSE.

The ideal would be some kind of in-between, but that’s a lot more complex, and involves the developers making decisions for you in terms of what to pull in and what not to.

It’s not just getting your communities seen.

Content from other instances for you and your users won’t show up in the all feed unless someone on your instance (or a bot) subs to the relevant communities.

Only off-instance communities with at least one local subscriber will federate.

You also wont immediately see new instances and their communities, and they won’t see you.

Heck yeah!

I’m gonna make a couple celebratory posts on [email protected].

Also, a tip on quickly getting up to speed with federation, is to get your instance and communities set up on lemmy-federate.com.

Since content only shows up for users on instances with subscribers, the idea is to use a bot account on each participating instance to have that one required subscriber, so that posts from new communities actually show up on other instances, rather than just the originating one.

It basically allows you to kick-start federation, actually allowing users to organically discover your instance and communities.

edit: There is an active Balatro community at [email protected].

If you have access to another device, you can log into icloud.com and remotely log out of the phone.

Files “deleted” from icloud can also be restored on icloud.com/recovery for up to 30 days after deletion.

AAA publishers would irreversably delete all games more than about a year old, if they could.

Just to force you into buying and playing new games, regardless of whether you’d rather play something “old”.

OGGDO BOGGDO

Also google when scammers want to advertise, vs when a real, one-person-business is trying to get off the ground.

ASK ME HOW I KNOW

What’s lemmy

Oh. We exist.

You may need to be subbed to a relevant community (for me it’s anime girls) to repeatedly see a given user to the point that you start recognizing them. I certainly have users I recognize.

You might know me for running the dailycomic bot that has continued the CnH posting after the original mod stopped doing it manually, or the [email protected] community.

But I know for a fact most people who’d recognize my handle, consider me “the anime girl poster”. I use the same username in games, and have been asked “are you the MentalEdge from Lemmy” once, which was new.

From my perspective there are also people who comment often enough that I remember them.

There’s the guy that posts a daily collection of screenshots from games he is playing, along with commentary on how he likes the games.

There’s a couple big names over on lemmyshitpost.

Same goes for tenforward.

There’s Blaze, who I know as a general background engine of activity and advocate for good on the fediverse.

There’s a bunch more, but I’m not gonna bombard them with mentions.

The closest thing I can think of, is Soulseek.

You can find almost anything on there. People share their entire collections, and almost everyone has some niche stuff they like.

I’ve spent hours exploring other people’s curated libraries, finding stuff I’ve never heard.

I don’t see how this would work financially, tho. Soulseek doesn’t make anyone money, except when i go out of my way to buy something on qobuz or bandcamp when I really like something.

Music is art. Like visual artists, it’s simple enough for one or a couple people to produce, but unlike visual art, it’s less commonly done on comission. Which means freely sharing your music, doesn’t typically put food on the table.

Hence, musicians sell albums or singles. Preferably directly to their fans. Souncloud, YT, and Soulseek regularly help me find new artists I like… But for actual listening I pull up Symfonium, hooked up to my Jellyfin server, serving my carefully curated personal collection.

Ah. So this is the competitive ranked sex I’ve been hearing about?

Sorry, I must’ve misremembered about systemd. It’s how my installs start up, and the unit file is not in the usual location for systemd units I’ve created myself, so my assumption was it came with Kopia. There is no systemd timer though, and one isn’t needed.

Edit: Just confirmed no systemd file came with kopia on my system either, my mistake.

in the past week, it did not backup anything. Hence, there is no scheduler built into kopia automagically as described/ hinted in the docs.

Was Kopia running during that time?

If you run a Kopia command, then it will perform the instructed task, and then exit. It will obviously not do anything after completing whatever command was given, as the process will have exited, leaving no kopia process running on the system. This is for when you use it in cron or your own scripts.

The other way of doing things is to run it in server mode kopia server start, which will set it running as a background daemon. When running, it allows you to log into the web interface or configure it via cli to do whatever you like. And as long as the process starts along with the host system, that’s all there is to it.

How the daemon is set up to start, doesn’t really matter.

My current setup, is as follows:

Personally curated music I buy and organize using Picard into folder A.

Lidarr is configured with folder C, which is a mergerfs volume consisting of folder A and B. Folder A is read-only, and any writes on C go into folder B. This way Lidarr can “see” all my existing music, while any automated downloads go into folder B, keeping them separate from my organized files.

Lidarr actually works, because it is hooked up to Soulseek using Tubifarry with ytdl as a fallback. I also have an import list hooked up to my last.fm recommendations to automatically download new stuff I might like.

When I feel like it, I go through folder B using Picard, moving things I want to keep into folder A.

To access my music, I use Jellyfin, also through folder C. My clients are Feishin and Symfonium.

In Symfonium, I use smart playlists for discovery. These playlists populate based on stuff like “unlistened tracks” or “multiple plays without being favorited” and “recently added from favorited artists”.

My favorite feature however is the tag-based endless playback which allows me to pick a track to start with, and then swipe through music with at least some kind of logic to the progression. This is my main way to browse my library.

It works extremely well, with the exception of files that don’t contain many tags. Hence my main pursuit has been to find a good way too add at least some genre tags to ALL my files. I haven’t found a final solution.

For iOS support, look at Navidrome for the server and maybe SubStreamer for the client.

She writes her erotic stories on her phone?

Pretty sure that’s what OP is doing, as they mention the Deck along with a kitchen chair.

If this form factor for ID isn’t well known enough for them to know about it, I doubt they received training on its existence.

Which means having this card will make them assume it’s fake by default. “That’s not what a US passport looks like!” Thereby increasing their deluded suspicions, rather than reducing them.

Like you said, it might be impossible to avoid ascribing things like intentionality to it

That’s not what I meant. When you say “it makes stuff up” you are describing how the model statistically predicts the expected output.

You know that. I know that.

That’s the asterisk. The more in-depth explanation a lot of people won’t bother getting far enough to learn about. Someone who doesn’t read that far into it, can read that same phrase and assume that we’re discussing what type of personality LLMs exhibit, that they are “liars”. But they’d be wrong. Neither of us is attributing intention to it or discussing what kind of “person” it is, in reality we’re referring to the fact that it’s “just” a really complex probability engine that can’t “know” anything.

No matter what word we use, if it is pre-existing, it will come with pre-existing meanings that are kinda right, but also not quite, requiring that everyone involved in a discussion know things that won’t be explained every time a term or phrase is used.

The language isn’t “inaccurate” between you and me because you and I know the technical definition, and therefore what aspect of LLMs is being discussed.

Terminology that is “accurate” without this context does not and cannot exist, short of coming up with completely new words.

Yes.

Who are you trying to convince?

What AI is doing is making things up.

This language also credits LLMs with an implied ability to think they don’t have.

My point is we literally can’t describe their behaviour without using language that makes it seems like they do more than they do.

So we’re just going to have to accept that discussing it will have to come with a bunch of asterisks a lot of people are going to ignore. And which many will actively try to hide in an effort to hype up the possibility that this tech is a stepping stone to AGI.

Obviusly.

And like hallucinations, it’s undesired behavior that proponents off LLMs will need to “fix” (a practical impossibility as far as I’m concerned, like unbaking a cake).

But how would you use words to explain the phenomenon?

“LLMs hallucinate and lie” is probably the shortest description that most people will be able to grasp.

Yup. The way the article titled itself isn’t helping.