user224

joined 2 years ago
[–] user224@lemmy.sdf.org 6 points 10 hours ago

Terminess nerd. From ttf-terminus-nerd, I think.

[–] user224@lemmy.sdf.org 12 points 10 hours ago (1 children)

Oh, that's not email, that's just sdf.org bulletin board: https://sdf.org/?tutorials/bboard-tutorial

Check the SDF FAQ: gopher://sdf.org/1/sdf/faq/
You will need some gopher-compatible browser. I recommend Lynx. I mean, there's also HTTP, but where's the fun with that?

[–] user224@lemmy.sdf.org 9 points 14 hours ago (1 children)

I didn't know what Lemmy was, but I've seen SDF say they now host an instance on Twitter, so I joined.

It is not very stable however. Sometimes it can go down for a week or more, sometimes the federation is broken. Historically, beehaw.org will always federate with us, even if somehow nothing else works.
Up until recently there was a 2 day delay with federation of some content inwards, however the Cloudflare outage probably gave us enough time to catch up, so it's working again for now.
Jerboa client often won't work with it because it times out after only 20 seconds. Sometimes you have to be more patient, at least 2 minutes. The MLMYM frontend seems to just load right away though.
The admins haven't been posting or commenting for 2 years, but they randomly changed banner and picture of the main community, so someone is still here.

At first they went big, with multiple instances. There is this US instance, but there were also ones in Germany, India, China and Japan from what I recall. But they're dead now.

My issue is they only take donations via PayPal which doesn't like my debit card.

[–] user224@lemmy.sdf.org 7 points 15 hours ago (3 children)

I mean, I do put in http://149.13.0.80/radio1hi.aac to listen to Radio 1. I also use it to test my network when DNS potentially screwed up.
I used to remember the domain name, then I did the IP for fun, now I only remember the IP. Perhaps I could do reverse lookup, but it's been working for quite a while now.

[–] user224@lemmy.sdf.org 7 points 15 hours ago* (last edited 14 hours ago) (1 children)

Our instance (in this context).

[–] user224@lemmy.sdf.org 35 points 15 hours ago (10 children)
[–] user224@lemmy.sdf.org 3 points 15 hours ago

Someone expecting LLMs with some sort of access to the system?

[–] user224@lemmy.sdf.org 9 points 15 hours ago

"Right, but what are you going to do? Use our competition? Which competition?"

[–] user224@lemmy.sdf.org 12 points 15 hours ago (1 children)

Somebody just walked in with this:

 

Just like a week ago it still took up to 2 days with some instances.

[–] user224@lemmy.sdf.org 4 points 22 hours ago

I am good, I am self hosting plenty of entertainment!

1 minute later

AWS outage took down Tailscale, I have nothing.

[–] user224@lemmy.sdf.org 5 points 22 hours ago

Wholesome to holesome.

 
 

Right, so Racknerd doesn't offer Arch image:

As for custom ISO installers, that requires opening a ticket with tech support, giving them a link to the ISO, and asking them to mount it.
Well, I am not doing all that.

So, there's also this outdated (will become important later) "rescue environment":

Linux Kernel 4.x is Debian 9 and 3.x is Debian 8. I don't know why they couldn't just say that.

So, the recovery environment has some RAM (but seems to be less than the VPS), and some storage (around 1GiB). The free storage is around 350MiB.
The recovery environment can be accessed over SSH. OpenSSH 1:6.7p1-5+deb8u4, on that older thing, if someone is curious. Modern OpenSSH client just complains about old key exchange (quantum-resistance), but connects.

Welp, Arch Linux bootstrap is 138MiB compressed, so let's go.
But not so quickly.
There's no wget, nor curl. So let's install them.
Well, apt no longer works. Old minimal environment without package installer. Cool.
I found some trick for HTTP on stackexchange using telnet. No telnet.
No lynx either.
So I downloaded it onto my PC. I first got the idea of unpacking it directly from different server, but yeah, right, no sshfs. That would have been useful for directly dd-ing images.
So I try to use rsync. Of course there's no rsync. scp saves the day.
Let's unpack the bootstrap now, shall we? We shall not, there's no zstd to decompress the archive.
The bootstrap won't fit uncompressed, and anyway, I am uploading over mobile data.
LET'S FUCKING GO! Gzip is installed.
I created a temporary 1.5GiB partition for the bootstrap, this later becomes swap space. And then I can more or less follow installation with Arch Wiki. There's also this wiki page, but it's mostly just regular Arch Install.


That's a very healthy memory usage. RAM nearly full when something else is running, swap typically above half. But their RAID-10 SSD setup seems to be doing well for that.
Speedtest, or really anything is mostly limited by that single virtual core.
I don't know what their shutdown, reboot, change root password, and reconfigure networking would do or screw up in this case. I haven't tried them yet.
The VNC cuts out with Cloudflare captcha every so often, by the way.

 

The previous one: https://www.the-sun.com/news/15337505/two-trains-crash-slovakia/


Ex 620 crashed into the rear of REX 1814.
Based on what I could gather from local news, REX 1814 "drove onto a rail where it wasn't supposed to be" and "possibly one of the trains crossed a red light".
In both cases was cited lack of ETCS.


Anyway, procrastination is useful sometimes. If I wasn't 1 and a half hour late, I could have been on either of these 2 trains (most likely the REX).
Like this, I just ended up on a later one with 170 minute delay.

 

Cyber criminals purchase advertisements that appear within internet search results using a domain that is similar to an actual business or service. When a user searches for that business or service, these advertisements appear at the very top of search results with minimum distinction between an advertisement and an actual search result. These advertisements link to a webpage that looks identical to the impersonated business’s official webpage.

In tips section:

Use an ad blocking extension when performing internet searches. Most internet browsers allow a user to add extensions, including extensions that block advertisements. These ad blockers can be turned on and off within a browser to permit advertisements on certain websites while blocking advertisements on others.

216
Aliasing (i.imgflip.com)
 
 

Based on other similar recordings I made, I estimate it at 4GiB.

It was a baseband recording of APT+DSB from the NOAA-15 satellite from when it had AVHRR scan motor issues. Not that rare for NOAA-15 (xD), but now that the satellite has been decommissioned, I'll never record it again.
https://www.ospo.noaa.gov/data/messages/2025/08/MSG_20250820_1410.html

I do have baseband recordings from good NOAA-15 and 18 passes, but still, this one would have been special.
I've posted about it when the issue was occuring: https://lemmy.sdf.org/post/3035683
Alternative link: https://lemmy.world/post/4162384 (preferred - our instance is slow - trying without cache a few hours ago, loading main page took 2 minutes and 12 seconds excluding loading of thumbnails)

I've just been looking into this again yesterday, and remembered that at some point I had a recording of this partial failure, but it seems I permanently deleted it. The last place it could have been, a HDD from my old laptop, I wiped 2 months ago (incl. full overwrite).

At least I still have the demodulated audio of the APT signal from that partial failure - keep in mind this was still analog - NOAA-15 launched in 1998.


Perhaps not the usual file with sentimental value, like picture or video, but I am a bit weird. I can never record it again. Fuck, I need to start archiving everything.
Now I feel like BBC, erasing TV shows to re-use the tapes.
Or perhaps more aptly (pun intended), NASA re-using Apollo 11 landing imagery tapes.

Oh, guess where I had the 2 remaining recordings. On the cheapest unbranded DVDs I bought on sale in Kaufland at 10 cents / disc, which seem to corrupt after 4 years and can split apart easily with fingers.

228
submitted 3 weeks ago* (last edited 3 weeks ago) by user224@lemmy.sdf.org to c/memes@lemmy.world
 
 

OK, first of all, I am not a programmer. (yes, I heard the "thank god") Perhaps I could make the top example simpler.

But anyway, I kind of like goto too much. I find it more intuitive to just jump around and re-use parts rather than think about how to do loops without too much nesting.

In high school, we only did Python. I really wanted to do goto in Python as well, but all I found was this April fools' goto module.

Now in college we're starting with C, and I am starting with Bad Habits^TM^.

Anyway, tagging every line was BASICally just for the joke, but it is useful to just jump to any random line.

 
 

I am going to try not to write yet another wall of text

  • I was supposed to get 3 months 50% off after adding my student card to transport system that I've already been using

  • An employee told me it indeed doesn't really work and just gave me 3M 50% off code from some spreadsheet

  • I tried to activate it nearly a month later, at which point it had "expired"

  • I contacted tech support, they told me they don't know about such offer, nor there being any spreadsheets with discounts, and instead offered me 50% for 1 month (possibly from the friend referral program), or that they could try opening a case which may take 30 days (basically urging me to get that 1 month)

  • I went out again to find that one single very specific employee who knows what I am talking about. He gave me another code from the spreadsheet. I warned him that it's for 6M, not 3M. "Whatever, so you get 3 months more", he said smiling.

There goes unlimited data for EUR 10.25 / month.

view more: next ›