Most likely something on that channel is spamming multicast. That kills most consumer wifi routers (in default settings). Usually something like a sonos or Google home broadcast group.

I might (i.e. I definitely do) have a non-ideal setup at home that contributes to this, with the router/WPA, a RPi running HA with a Zigbee antenna just next to it, my server in the shelf next over and a Sonos above it. Worst of all, the server is running on WiFi and it is sat in immediate vicinity to my router. Why? Because I could not for the life of me make the ethernet transfer speeds be more than somewhere around 1-5 MiB/s, so I gave up. But considering these issues are so infrequent from this location I assume it is mostly due to outside interference.

And adding to that, …

Use an app like PingTools (Android) that can graph what is on each wifi channel. Check to find the cleanest channels in your area and configure your router to use that channel.

… this scan shows that all discovered networks are occupying the same channels (98-114 it looks like). I have still not figured out what OpenWRT option to go for (the OpenWRT One, which I had originally planned, quickly became very expensive with imports and tolls that I ended up not buying, despite having decided to go for it). So I am using the stock ISP router, which I assume everyone else in my building are as well. And I have been unable to locate an option to change channels in its interface.

If you’re curious and technically-minded, I highly recommend this write up: https://www.wiisfi.com/

That is a resource I didn’t know I needed! Thanks :)

Ah, that would make sense. I think my immediate neighbors are home, but about half of my building should be traveling for Easter (judging by the empty parking garage at least).

I use this:

rsync -rah --progress /path/on/source/ user@ip.of.local.server:/path/on/destination

I will try the ping next time I attempt this. There doesn’t seem to be a definite time when the issues start though. It tries to copy the file over, and when it is done it continues to the next. If the first one didn’t succeed, it will retry and if that also fails, it will say “ERROR verification failed - discarding update” (paraphrased) and continue to retry the next file if that also failed.

I do see some fluctuations in the transfer speeds during transfer, which could indicate the times the connection is struggling.

Makes sense that it would be able to, especially considering the flag that allows you to force an older version as the sender. Still find it strange that on equal version numbers they default to different protocol numbers. I would have at least thought there was an easy way to tell it which protocol version to use by default, but I have at least not been able to find out how.

I just joined an old-school phpBB forum dedicated to a specific topic, and it is glorious.

It is important. But I find the ones I have tried good, and would survive if I had to use either of them. I use KDE Plasma on my main personal laptop, I have Cinnamon running on a living room computer connected to my TV (not an ideal solution, but I’ve so far not taken the time to optimize the setup) and GNOME om my work laptop. I much prefer KDE Plasma out of them, but I like the others also.

I am not sure the shareholders will accept such a meager compensation. Did you include emotional damage in your estimate?

Correct me if I’m wrong, but Anna’s archive is not giving you song downloads, but rather metadata

Were they not going to release the songs as well? They just started with the metadata?

ETA: Yes, this is from their blog post about it:

The data will be released in different stages on our Torrents page:

[X] Metadata (Dec 2025)
[ ] Music files (releasing in order of popularity)
[ ] Additional file metadata (torrent paths and checksums)
[ ] Album art
[ ] .zstdpatch files (to reconstruct original files before we added embedded metadata)

Can we also take a moment to acknowledge how utterly unhinged this part is?

“This marks not just the next chapter, but the next book in SpaceX and xAI’s mission: scaling to make a sentient sun to understand the Universe and extend the light of consciousness to the stars!”

Their web installer makes degoogling accessible to everyone.

With the right hardware

otherwise they will be prone to warp and collapse in on each other. If your item will be as tall as you suggest, this is likely to happen before the print even finishes

Yeah, this is what happened in the original, failed print.

I ended up making the model with 3mm thick walls, using two perimeters and 10% adaptive cubic infill (I sliced with gyroid as well, but it looked weird). Turned out great. I made some that were not quite as tall as well, with 2mm thick walls and 3 perimeters, which worked fine as well. It might have worked for the main boss here as well, but I’m not quite sure. The difference between the models was about 55 mm in height. The difference in material usage between the two options was negligible (< 10 g), with the infill variant coming out slightly lower in consumption.

Good lord no. Don’t do that. That would be a waste of filament and also cause a host of other problems for you.

Just to be clear, it is the “hoping to use infill to get some support” that your response is aimed at, right? Out of curiosity, what kind of problems could I be looking at for this?

In the meantime, I’ve been printing some smaller bins of about half the height of my problem where I increased the wall thickness of the model from the default 0.95 mm to 2 mm, and used 3 perimeters, which resulted in fairly sturdy walls.

Looking at the same bin with 2 mm walls without infill, and 3 mm walls with infill, there is barely any difference in material usage.

I was writing up some additional questions as I didn’t quite get your suggestion, but while writing the response I think I understand. You suggest that I increase wall thickness in the slicer settings to match the model thickness, and not only in the model?

I was hoping to use infill to get some support between the outer walls and avoid having to use too much plastic and not having a single, free-standing wall.

If I understood you correctly, do you have a suggestion to what a suitable wall thickness would be to avoid the issue I described?

I do plain text accounting with ledger-cli. Nice and future-proof, and I can easily build stuff on top of it. Other examples are plain text accounting software is hledger and beancounter, the latter which is known to have a pretty good Python-ecosystem surrounding it.

We would need open source cars which will never happen.

:'(

When I initially scanned thtough the headlines, I read this one as “ASUS plans to produce RAM problems” and thought “Yeah, of course they are”. My expectations of companies seem to be very low in general these days…

That’s a weird thing to present as an absolute truth. As someone who has exstensively used both Windows (3.1, 95, 98, ME, XP, Vista, 7, 8, 10 and 11) and macOS (from 2011-2022), and now using KDE Plasma on my daily driver laptop, GNOME at work and Cinnamon for my living room machine: all three Linux DE are superior experiences.

Surely there are people who would prefer Windows and macOS over them, but it is highly subjective.

They got camera working on FP3 and FP4. Fairphone employs a guy who basically does work to get their hardware to run Linux.

Can’t wait to listen to a constant buzz of thousands of these in the air at all times.

How voluntary is it when these platforms have a monopolistic grasp on how consumers access music these days? And the more people believe that the artists are actually fairly compensated from this model, the firmer this grasp becomes. What choice do they have of being there if they want to have any kind of reach?

A Spotify Premium subscriptions will cost someone 156€ a year. If that person instead spent that entire music budget on purchasing albums from select musicians according to the enjoyment they derive from their works, or buy concert tickets or merch, and decides to pirate the rest of their music listening, what changes? For the consumer, they are now left with actual, irrevocable access (legal and illegal) to the same music you had rented access to before, and have spent the same amount of money. For the musicians, the ones who received the purchases are left with much more of your dedicated music spend, and the rest will have marginally less (their share based on total streams of your monthly subscription x12). For Spotify and Taylor Swift, they receive marginally less money (but more than the artists you actually listen to) of which they should probably not have received to begin with.