Luke

It did load homedepot.com when I tried it just now, but I don’t have a mouse or keyboard attached, and the monitor isn’t touchscreen, so I have no idea how it performs when scrolling. Probably terribly.

IIRC, mine is an earlier version of this one: https://www.raspberrypi.com/products/raspberry-pi-3-model-b-plus/. It has 1GB of RAM, and a 64GB sdcard (which is honestly bigger than it needs), with basic Debian Bookworm installed. It runs essentially nothing except sshd, xwindows, and Openbox configured with the following autostart script:

xset -dpms
xset s off
unclutter -display 0:0 -noevents -grab
export DISPLAY=:0 && firefox-esr --kiosk $URL_TO_VISIT &> /dev/null & disown &> /dev/null

Where $URL_TO_VISIT is a panel on my local Home Assistant.

Granted, it’s not exactly doing much other than showing a single page all the time, and sometimes it does freeze and require a manual restart every few weeks (hence why I said it’s only “running okay”). It does work though, and I expect that an rPi 5 would be a good experience for actual browsing, especially if you used one of the 4GB or higher versions.

If you aren’t already, I recommend running a blocker like adguard on your network. Aside from making the internet more pleasant to look at overall, it might help with making sites more responsive.

Yeah, I have had firefox-esr running okay for years now on an ancient rPi 3 as a Home Assistant panel. I expect an rPi 5 would be able to run Firefox just fine.

Because you’re the account who posted what I’m responding to.

You didn’t make any substantive critiques about the journalism, so why would anyone be responding to that? All you’ve said is that you “don’t care about ffmpeg”, which is dismissive of the software itself, so yeah obviously people are going to be responding about the software.

Incompatible with every website in which browser? It works for years in both Chrome and Firefox. Is this a meme for Safari users only?

The fact that Google invented this format is the most annoying thing about webp, but the complaints in this image haven’t been an issue for a very long time in my experience.

There’s also GNOME Maps if you enjoy the GNOME look and feel (which I do):

I’ve been using n8n for years, but their constant (admittedly mild) nagging to upgrade is bugging me lately, so I’ve been looking into alternatives.

Kestra looks promising, but I haven’t played around with it much.

Anyhow, I’m appreciating the perspective of the article you shared. Thanks!

You have a lot going on there all at once that could be contributing to the whole thing failing.

If it were me, I would try to get caddy working independently of everything else first, since it was your original problem, then layer in the other containers one by one in case you’ve got configuration problems in them too.

Caddy by itself is super easy to verify with something like a browse directive pointed at an empty directory.

Then add your tailscale container and configs, and check that you can still access the browse page from caddy.

Then add your Jellyfin container and adjust the Caddyfile to proxy to it.

Also, Plex email blasted a few weeks ago about how nobody can share their libraries anymore without paying for a subscription. That was the push I needed to check out Jellyfin again, and the experience ranges from “good enough” to “that’s better than Plex” for me and my buddies.

Damn, that’s unfortunate, but good to know. Thanks for the info.

This seems like a really nice tool, congrats!

Too bad there is so much focus on AI though. The UI looks nice, and templating and being able to schedule posts would be super handy, but I don’t need an AI to write things for me. I find that using AI is ethically icky anyhow; I’d rather not have it in any of the tools I use.

Is it possible to disable all the AI features when using Postiz? Like, a boolean setting in the deployment configs would be great.

Couldn’t you do that with Firefox? It stores all that stuff locally already, doesn’t it? Just disable sync and use whatever software you prefer to watch those directories.

Edit: here’s how you can figure out where the settings and bookmarks are stored locally on your system: https://support.mozilla.org/en-US/kb/profiles-where-firefox-stores-user-data

People need to stop posting content to YouTube. Quit giving them new leverage.

Even the linked article whines about how they don’t want to use Peertube because “the audience for the content is 100x smaller” but that’s at least partly a self fulfilling situation. Of course they aren’t going to have a large audience on Peertube when they don’t post anything there. Mirror your old content there. Upload new content there instead. Advertise your Peertube channel instead of YouTube.

There’s not going to magically be a huge audience out of nowhere on alternative platforms, it takes content creators to migrate first.

Why use the AI in the first place then? Just search for the actual recipe sources from the start.

That’s working as intended; as the compose docs state, the command passed by run overrides the command defined in the service configuration, so it wouldn’t normally be possible to actually shut down all the containers and then use docker compose run to interact with one of them. Run doesn’t start anything up in the container other than the command you pass to it.

I’m not familiar with funkwhale, but they probably meant either to (a) shut down all the containers except postgres so that running pg_dump has something to connect to, or (b) use exec as you have done.

Personally, I do what you did, and use exec most of the time to do database dumps. AFAIK, postgres doesn’t require that all other connections to it are closed before using pg_dump. It begins a transaction at the time you run it, so it’s not interfering with anything else going on while it produces output (see relevant SO answer here). You could probably just leave your entire funkwhale stack up when you use docker compose exec to run pg_dump.

There are a ton of great UI libraries available, many with bindings for whatever our preferred languages are. We don’t need an LLM for any of that.

Humans also carry diseases and destroy the landscape and each other. By your logic, we shouldn’t care about anyone dying, or try to empathize with anything outside of ourselves. Seems like a sad perspective, IMO.

Ah yeah that sounds really frustrating. I’m sure you’ve tried this already, but does Lineage have support for the one-handed mode gesture? I’ve had that come in handy (no pun intended) now and then.

I read somewhere that the Thunderbird team was working on fixes for the new drawer UI, here’s hoping they take better accessibility in mind!

I’m not having any issues using it with one hand so far. Not even sure what is very different about the UI beyond a different icon set and the account switcher being fewer taps to get to. I’m curious what about the new app makes it hard for you? Maybe there’s something highly annoying that I haven’t run across yet.

Unfortunately, there doesn’t seem to be anything better than Calibre at the moment. (Though, I’m happy to be proven wrong!) Nothing against Calibre, it’s functionally amazing free software and it works very well; I said “unfortunately” because the interface is extremely dated and clunky and confusing to operate. Once you get it working, it’s very nice though. As long as you never have to go fiddling with it again, because every time you’ve gotta reacquaint with it’s weird UI. Still, it really is the best available at the moment, and it’s free so that’s awesome.

My favorite way to set it up is using the linuxserver image, which has a web-based VNC built into it, so you can remotely run the app on a headless server and then use your browser to interact with it.

I have Calibre configured to monitor a folder for new stuff I throw into it, where it’ll automatically fetch metadata and put it into the database. Calibre also has an OPDS server built in, to which I point a nicer frontend for reading comics. Currently that is Kavita which provides a decent web UI for both books and comics.

Anyhow, I believe you could enter data about your physical comics into the Calibre database, and then view the metadata with something like Kavita, though of course you’d be skipping the reading features.