It’s been years since I was experimenting, trying to set things up properly, but at the time I could swear that I’d have the following set:

PubkeyAuthentication yes
PasswordAuthentication no
UsePam yes

and attempts to SSH into the system would still result in a password prompt, even though PasswordAuthentication was set to no. After some experimentation, I found that if I set UsePam to no, you wouldn’t get a password prompt, so that’s what I’ve been doing on every SSH server I’ve had since like, 2000.

Tinkering with it now though, I see there’s also another directive, KbdInteractiveAuthentication, which when set to no, will ensure I don’t see a password prompt even when UsePam is set to yes. I can only assume that 26ish years ago there was no KbdInteractiveAuthentication, or I just didn’t know about / notice it!

Well no harm done in my case anyway, as I don’t generally expose SSH servers to the outside world anyway. But it’s good to know that I don’t have to turn off PAM to disable password logins anymore.

"Didn’t you wear that like, five years ago? "

I’ve been setting UsePam = no for decades because leaving it on meant that it’d still present a password login of keys failed. Is this still the case, was I wrong from the beginning, or is this somehow ok?

I don’t know where you saw this “No”, but all of their code has been up on their (self hosted) GitLab for years: https://git.jami.net/savoirfairelinux

He compared an LLM to a compiler suggesting that the concerns about using them are the same as suggesting that everyone should write in assembly.

The comparison is invalid, since one mode is deterministic, while the other is not.

LLM-driven development leads to programming via the abstraction of the model, disconnecting you from the source to the point where you don’t understand what it’s done or why. He’s suggesting that this isn’t a problem because an LLM is just like a compiler, which suggests that he’s arguing in bad faith or doesn’t understand the “tools” he’s using.

Does he… think LLMs are deterministic?

Your work lets you use Linux?

Honestly? Gentoo. It’s got a steep learning curve, but the docs are great and after your first install, you’ll know more about Linux than most users.

I don’t see a link to the source. What’s the licence?

Back in the days when WordPress started out, 1GB of RAM was a lot! You’d be surprised what you can do when you’re just running a server.

[LLM executes it to test]

I understand why you’d want to avoid MIT, but as someone who generally favours the AGPL for everything I do, I’m curious why you opted for the EUPL instead.

Just get one of these and set everything up in a few browser tabs. Either that or use an OS with workspace support and put the different apps you want on separate workspaces.

In our livingroom, our old laptop runs Arch Linux with GNOME. Jellyfin is on desktop 1, and Firefox is on desktop 2. Firefox runs Netflix, Spotify, and YouTube in separate tabs and switching between all of these is a keyboard shortcut on the above remote. Frankly, it’s simpler than most “smart” TVs I’ve had to work with.

Or just a cheap laptop.

Given that sort of description I’d bet that it is DNS actually. If you run a DNS lookup from your LAN for your domain, do you get the external IP or the internal one? You can try this with dig in Linux or nslookup in Windows (I think).

If I’m right and you’re getting the external IP even when querying from your LAN, then what you need is either a separate DNS for your LAN or what they call a “split DNS” whe are a your domain server returns a different value depending on the originating IP.

You can check for this by editing your hosts file. If you add your domain and local IP there and then try to ping that domain on that machine, it should work. If you’ve got a small network, you might even just copy the hosts lines between machines (this is what I do for example).

TIL! This is very cool looking. Thank you!

This wasn’t nearly as interesting as the headline made it sound.

It’s a physical box you purchase to effectively break the law, so some of the architecture is going to be shady because it has to be.

• it polls multiple domain servers to get around blocks (and probably updates that list regularly)
• it obfuscates the endpoints it hits because they’re trying to secure a box they’re handing to someone else from tinkering that would allow people to access their (pirate) services without buying a box.
• it obfuscates its user agent when connecting to a weather service to avoid being blocked.

There’s nothing special or nefarious here. Indeed I’ve worked on projects that had to take the same considerations into account.

The “someone else’s Hulu account” claim sounds like bullshit to me because it doesn’t make sense from a business perspective. They’d want to control the accounts in question and rotate the passwords regularly, again to avoid freeloaders. More likely they’ve paid for 25ish Hulu subscriptions in every region (hence the initial call for IP geolocation) and are then relaying credentials to a box based on this info.

The only really sketchy thing in this whole video was the disabling of TLS checking, which was likely done to get around sketchy pirate websites with bad or nonexistent certs. It wasn’t clear though which part of the OS this applied to. If it’s only doing this for the pirate streaming, that’s sort of a bullet you have to take if you want the service. So long as the only data being sourced unencrypted/unverified is some audio and video, you’re fine (assuming you’ve already accounted for streaming such data in your jurisdiction, ie. you’ve got a VPN). If you’re pulling down software updates though, you’re gonna have a bad time.

At my first-ever tech job back in 2000, I was given a little Celeron desktop computer to do my work: mostly writing some ColdFusion and cutting up images in Photoshop. For the most part, the machine worked fine, except whenever I scrolled down on a webpage in Netscape, the box’s PC speaker would start screaming. I tried to ignore it, but I was in a small room with two other nerds and it was getting really annoying.

One day, my frustrated colleague decided it was time to investigate. As part of the process, he noticed that there was very little air coming out of the power supply fan port. “Must be a busted fan and that’s causing an overheating alarm” he declared. We unplugged the box, popped it onto my desk, and opened it up. The internals were pristine. No dust, nothing. The computer was practically brand new after all.

We were just about to crack open the power supply (not recommended, but we were getting desperate) when my colleague noticed that something was wedged inside the fan… it was the biggest (dead, thankfully) cockroach I have ever seen, at least 10cm long. For context, this was in Vancouver, Canada. Those just don’t exist there.

My colleague jammed his screwdriver in there to grind up the cockroach carcass, then plugged the box back in. After a shittone of dessicated cockroach guts spewed out the back, my little computer was operating normally.

As best we could guess, the monster crawled into the fan in Malaysia when as was assembled and somehow got pinned in there, died of starvation, and then stowed away to Canada.

There’s a nifty project on Mastodon where someone is running an instance entirely on solar (and I think a battery?). They’ve been tinkering with some of the code to do things like optimise images for reduced weight or CPU cost.

The project is called @solaradmin@solarcene.community and the account for the person running it is on a separate instance, presumably so people can message them about problems: @smallsolar@techhub.social.

The right time to look elsewhere was when Microsoft bought it.

The next best time was when viable alternatives like GitLab and later, Codeberg appeared.

The next best time was when it became clear that they were stealing your code to feed into their sparkly autocomplete and were going to sell it back to you.

The next best time was when they dumped a bunch of vibecoded garbage into the codebase and killed the uptime.

The next best time is now.

You don’t even have to migrate all your stuff. Just start all your new projects on Codeberg, or GitLab, or something self-hosted. Once you get used to the new place, you can migrate your old stuff when you’re ready.