Interesting. I’ve only ever had success with cellular PPP when using #777 for carrier based DUN back before mobile broadband was a thing.

Yeah. Part of that is the same kind of nostalgic for me, but also I guess I miss the feeling of the internet being somewhere you go, deliberately, rather than always-on, always-connected, pinging me with attention-sucking notifications constantly.

Like when you sign into AOL messenger in the dial-up days. That was an indication you wanted to chat and had set aside time for it; it was like flipping the sign from “closed” to “open”. Now, you’re just always expected to respond.

I have no idea about the why, but I’ve seen the same thing. All new accounts spamming content to all of those (and at least one known-good user posting links there; maybe thought they were legit?). I’ve banned/removed them all locally on my instance, and I think some other admins are doing similarly.

All of those, if you go to their base URL, seem to be part of the same campaign (spampaign?): The Freedom Project / The Freedom Estate Project

I may just start blanket-banning *.pages.dev for now and deal with any false positives case-by-case.

major record labels consider it a parasitic threat

Well, they would know. Takes one to know one.

What do record labels even do anymore? Broadcast radio is dying and even when it was less dead, it was the same 100 payola songs on repeat. It’s never been cheaper or easier to self-record and publish, and there are all kinds of online platforms artists can use to distribute their work and get paid.

Hell, the last few new artists I’ve found were from YouTube/Spotify/Band Camp and had Patreon accounts.

The music industry needs to die off already.

Lemmy UI constantly pissed me off, Photon didn’t quite do what I wanted, so I forked it and learned Svelte. lol

You know anybody who can network 8 connection machines and debug 2 million lines of code for what I bid for this job? Because if he can I’d like to see him try.

How far we’ve come, indeed. Lol.

One major change was the introduction of an image proxy. This means that images in newly added posts (and new user/community avatars) are no longer directly downloaded from the federated servers, but rather first mirrored on our server.

I really wish the image proxy wouldn’t federate out the proxied thumbnail image URLs, though. My instance isn’t fetching the thumbnails of those itself (from 0.19.4+ instances) and always uses the poster’s home instance’s proxy URL. Just a gripe I’ve had with how they’ve implemented that feature.

Should turn around and sue Nintendo for facilitating the facilitation of piracy by making the consoles in the first place.

I run the same 500kb limit on my instance. It was becoming restrictive, so I added a feature to Tesseract that optionally preprocesses uploads to webp, client side, before uploading (with a user configurable quality level) which has worked well. Might be something to consider or perhaps can patch a similar preprocess step into your default UI.

I never really ran into that, and part of the curriculum for that class was editing / correcting other people’s documents.

Was that only if you imported a document from a different format, perhaps?

There were a few times I encountered slight formatting glitches, but we were taught to just turn on Reveal Codes to easily find/fix those.

I miss old-school WordPerfect. Our school was largely using Office 97, but one teacher preferred WP, and we had to use it in that particular class.

My biggest takeaway from it was that, contrary to what MS would have you believe, it is absolutely possible to put formatting options in logical places in menus. Everything about WP was just so intuitive.

I think cartographers also used to put fake things on their maps to detect unauthorized copies.

My takeaway from that is that I should start putting canary bugs in my code 😆

I just put a caching layer in front of my /pictrs path. Problem solved.

I use Nginx for that, but some instances use Cloudflare which can be configured similarly.

Not sure then. I think Emby, or at least the way mine’s setup, only uses tvdb and moviedb for metadata. Assuming JF is similar. So if the show doesn’t exist there, it may just pull the next closest match regardless?

AFAIK, you only need to set one of those. I usually set IMDB ID since tvdb and moviedb can pick up on that.

Maybe someone else has some tips.

Not sure in JF, but in Emby I can click the menu for the show in the library and choose “Identify”.

Usually setting the correct title there is enough, but if it keeps getting mis-identified, I’ll set the IMDB/TVDB/ etc IDs directly.

True, but that doesn’t stop them from being propagandized as “piracy”.

Everything that gives people non-gatekept access to any media is considered “piracy” by the powers that be.

That propaganda image is from the 80s.

Re: NFL

I don’t want to ask any kind of LLM anything at all. Ugh.

Yeah, I’ve never had a good experience with Linux on any Peintium 4 (though II and III worked great). The last time I tried was on an old P4 IBM ThinkCenter. Like you said, could be GPU, but even non-graphics tasks were sluggish on it.