@prologic@twtxt.net I can’t really commit to that. Don’t plan anything around me.

@shreyan@twtxt.net Same here. I work relatively late so I’m never up that early.

@prologic@twtxt.net Nice! Save some marshmallows for me.

@prologic@twtxt.net Any of the above

#QOTD : If you could redesign a fundamental internet protocol from scratch, which one would you choose and how would you improve it?

@rrraksamam@twtxt.net I’m looking forward to my all-SSD Btrfs RAID5 NAS. I think it’ll be a while, though. I just paid $6.92/TB for a couple of used 12TB HDDs.

@prologic@twtxt.net They’re shutting down after 7 years. It was a great place to buy Monero with cash by mail. https://localmonero.co/nojs/blog/announcements/winding-down

@aelaraji@aelaraji.com Nice. Compiling problematic software is my #1 use of containers on my PC. I use a handful of them on my server.

@lyse@lyse.isobeef.org Same here. Where does it not work, @movq@www.uninformativ.de?

@movq@www.uninformativ.de People just don’t ask these questions. It’s really a serious privacy issue, and I don’t see it brought up very often. Not even in privacy-minded circles. If you’re using a proprietary operating system on any Internet-connected device, you need to assume that the vendor can see everything you do on it and maybe even what you do on other devices as well..

Actually, it looks like notifications using Google’s service can be encrypted end-to-end. I don’t know if this is used much in practice or if you can tell if the notifications on your device are encrypted. There seems to be some conflicting information out there.

Even if the content is encrypted, though, you’re still giving quite a bit of metadata to Google by using their notification service.

It looks like ntfy.sh can work either through the OS’s notification service or by maintaining its own connection to the server in the background. For privacy, you definitely want to use “Instant Delivery” and self-host the server.

https://docs.ntfy.sh/faq/#how-much-battery-does-the-android-app-use
https://docs.ntfy.sh/faq/#what-is-instant-delivery

@movq@www.uninformativ.de I haven’t done any app development, but I know notifications on phones are indeed dependent on cloud services run by the OS vendor which talk to servers run by the app vendor on your behalf. This is supposedly better on battery life, but it conveniently lets your OS vendor read all your notifications.

Mobile XMPP clients usually implement notifications using XEP-0537 and it goes like this:

Your XMPP server -> Client vendor's notification server -> Client OS notification server -> User's device

It’s not end-to-end encrypted so servers will usually just send a dummy message through (You received a message from juliet@capulet.lit!) so you have to open the app to see the (hopefully) encrypted message.
It’s a similar flow on both iOS and Android and I assume Matrix clients work the same way.

@prologic@twtxt.net I know, right? It’s a very elegant solution to the problem using standard command line utilities. It was too hard to find. I went through 3 or 4 Stack Exchange threads from my Web search before I found somebody linking to this answer. People were misunderstanding the question and suggesting all kinds of crazy methods including weird, proprietary, GUI Windows software.

How To Efficiently Copy Files To Multiple Destinations: https://mckinley.cc/notes/20240508-copy-multiple-destinations.xhtml

@prologic@twtxt.net I can’t recommend it enough.

@movq@www.uninformativ.de

$ units -t '500 gigabytes per 9 hours' 'megabytes per second'
15.432099

That’s a very unfortunate speed in the year 2024.

@movq@www.uninformativ.de That’s no fun at all. I don’t like to throw away working hardware either, but I wouldn’t wait 7 hours (CPU-bound!) for my manual backup to complete if it could be done faster on a 10 year old laptop with AES-NI. How much data did you add?

Speaking of which @prologic@twtxt.net, have you heard from @ocdtrekkie@twtxt.net lately? He’s active on mastodon but I haven’t seen him around here in a long time.

@prologic@twtxt.net I agree with @movq@www.uninformativ.de. Good documentation is better than an interactive setup process. My difficulties (#isyb2aq) were because I was just doing it for testing and I wanted it running as quickly as possible. If I was running it in a production capacity, I would read through the documentation.

If you’re trying to make non-technical people set up their own Yarn pod, that’s probably (unfortunately) impossible. Management software like Sandstorm make it “as easy as installing apps on your phone” (direct quote from sandstorm.org) and most people still pay Google to store their photos.

I remember you were trying to do paid hosting for Yarn pods in the past. That could work, but as I’m sure you know it’s difficult to convince people to use this over X or Facebook, let alone host their own pod. I think it’s going to stay a small community of fairly technical people for the foreseeable future.

I did it again… #cm7e3ya #s4nbfta

I edited it because I started the line with 500., which the Markdown parser took as the start of an ordered list and made it number 1.

@movq@www.uninformativ.de I do wonder that sometimes, but I try to take notes if I’m doing something complicated. Just a few lines in a text file with some context plus the command I used. ffmpeg.txt comes in very handy.

It’s 500. I never changed it, so that’s the default of either Bash or my distro. It’s fine for me.

@bender@twtxt.net That’s what I suspected. I compared the text, including the alt text for the image. I guess I didn’t read it carefully enough.

No worries @aelaraji@aelaraji.com, it happens to the best of us.

Why are there two threads for the same post? #2hhvp2a #kz5qjza

@aelaraji@aelaraji.com I’m definitely putting that in the list. I like tmux but I just can’t wrap my head around the controls. This looks more like a tiling window manager.

@aelaraji@aelaraji.com Is that a terminal multiplexer? If so, which one? I suspect it says at the top but I can’t quite read the text.

@bender@twtxt.net Fair point… :)

@prologic@twtxt.net Planning it ahead of time is all well and good if you have the money to buy 6 or 8 hard drives at once. I really don’t, and I want to mirror the whole thing offsite anyway. Mergerfs will let me do it now, and I’ll buy a drive each for SnapRAID in short order.

QOTD: Have you ever suffered significant data loss? If so, what went wrong?

@bender@twtxt.net Ha, we both looked it up at once. You win.

@bender@twtxt.net Synology uses single-volume Btrfs on software RAID, which seems to be pretty solid in my research but that’s less flexible than ZFS. https://kb.synology.com/en-us/DSM/tutorial/What_was_the_RAID_implementation_for_Btrfs_File_System_on_SynologyNAS

@bender@twtxt.net Exactly. It’s just not an option with warnings like that all over the place. Some people have had success, but I’m not risking it. https://lore.kernel.org/linux-btrfs/20200627032414.GX10769@hungrycats.org/

@prologic@twtxt.net ZFS is fine but it’s out-of-tree and extremely inflexible. If Btrfs RAID5/6 was reliable it would be fantastic. Add and remove drives at will, mix different sizes. I hear it’s mostly okay as long as you mirror the metadata (RAID1), scrub frequently, and don’t hammer it with too many random reads and writes. However, there are serious performance penalties when running scrubs on the full array and random reads and writes are the entire purpose of a filesystem.

Bcachefs has similar features (but not all of them, like sending/receiving) and it doesn’t have the giant scary warnings in the documentation. I hear it’s kind of slow and it was only merged into the kernel in version 6.7. I wouldn’t really trust it with my data.

I bought a couple more hard drives recently and I’m trying to figure out how I’m going to allocate them before badblocks completes. I have a few days to decide. :)

@bender@twtxt.net There’s stagit which generates static HTML files

@prologic@twtxt.net I remember running yarnd for testing on a couple of different occasions and both times I found all the required command line options to be annoying. If I remember correctly, running it with missing options would only tell you the first one that was missing and you’d have to keep running it and adding that option before it would work.

This was a couple of years ago, so I don’t know if anything’s changed since then. It’s really not a big problem, because it would be run with some kind of preset command line (systemd service, container entrypoint) in a production environment.

@bender@twtxt.net I avoid install scripts like the plague. This isn’t Windows and they’re usually poorly written. I think it’s better to prioritize native packages (or at least AUR, MPR, etc) and container images.

@prologic@twtxt.net That’s good advice. I don’t open any ports to the Internet if I can possibly avoid it. Everything is on Wireguard, even stuff that doesn’t really need to be. It’s super easy to set up on other people’s computers, too. Even on Windows.

@prologic@twtxt.net Both are very nice in my opinion. I don’t think you could make a mistake with either, at least when it comes to looks.

@prologic@twtxt.net I think this would be solved in the short to mid-term by fixing the mute function. Or, maybe, adding a “Hide this user from Discover” button.

@prologic@twtxt.net Picnic CSS is my favorite one on first glance.

@prologic@twtxt.net Are they changing unique IDs? I hate when people do that. If I ever do that with any of my feeds, feel free to mock me relentlessly.

@bender@twtxt.net Makes sense. We definitely need the ability to mute feeds from the Discover feed.

@movq@www.uninformativ.de I remember your solution. It’s very simple, I like it.

Yes, my backup target is my home server. I have a hard drive dedicated to Restic repositories. It’s still not a real backup as I don’t have anything offsite but it’s better than my previous solution. I had two very old hard drives I kept plugged in to my desktop PC and I would (on very rare occasion) plug in another hard drive and copy all the files over to it. Luckily, I’ve never suffered any significant data loss and I would rather not start now. Once I have automated backups on each of my machines, the next project is getting those backups offsite.

@prologic@twtxt.net I think one-way feeds are okay and we shouldn’t discourage them so strongly. On the other hand, I think it’s the duty of a poderator to filter out feeds that are just noise from the Discover feed. I definitely consider a truckload of one-way posts mostly in another language to be noise. Did you get rid of Gopher Chat too? I’d call that noise, for sure.

@bender@twtxt.net Standard twtxt is a microblog in its purest form. A blog, but smaller. It’s just a list of posts to read, and that’s an echochamber in the same way my regular blog is an echochamber. I don’t think there’s anything wrong with that.

@prologic@twtxt.net I support the delisting of ciberlandia.pt in the Discover feed due to the sheer volume of posts from there and the fact that most of them are in Portuguese with this being a predominantly English-language pod.

@prologic@twtxt.net Why do we need to avoid posting to the void? That’s pretty much what twtxt was made for. I don’t like the “Legacy feed” terminology, either. I support the delisting of ciberlandia.pt but I think this change is heading in a bad direction.

I like @sorenpeter@darch.dk ’s suggestion. It gives the users the information and lets them make their own decision instead of putting a big scary warning in their face. That’s what Microsoft does, and we shouldn’t be Microsoft.

@prologic@twtxt.net How do you manage multiple remotes? Do you just run restic backup for each one?

I wish there was a good GUI for Restic so I could have non-technical people using the same thing I do.

QOTD: How do you back up your files?

I asked this one almost a year ago and I started using Restic shortly after that. When I started, I was only backing up my home folder to the repository over NFS. Now, I’m backing up the entire root filesystem to a repository using the REST backend so I can run Restic as root without breaking the permissions.

I’m working on automating it now and I’m trying to come up with something using pinentry but my proof-of-concept is getting pretty obtuse. It will be spread out in a shell script, of course, but still.

systemd-inhibit --what=handle-lid-switch restic --password-command='su -c "printf '"'"'GETPIN\n\'"'"' | WAYLAND_DISPLAY=wayland-1 pinentry-qt5 | grep ^D | sed '"'"'s/^D //'"'"'" mckinley' --repository-file /root/restic-repo backup --exclude-file /root/restic-excludes --exclude-caches --one-file-system /

I’m curious to see how everyone’s backup solutions have changed since last year.