antihero

Our org has a really nicely set up review bot which has access to a repo of really strict standards to check new code against. It means that the reviews are much more useful and catch a lot of things that a plain bot wouldn’t. I think that’s a huge factor in being able to trust the reviews to catch more. You can also check the repo out locally.

I’ve genuinely found that context switching is a skill like any other. When I first started using LLMs and context switching doing 5-10 different work streams at once I was utterly, utterly exhausted at the end of each day.

Now many months in, it doesn’t feel nearly as taxing.

Maybe this is in no small part due to LLMs improving enough you don’t have to spend as much mental energy reviewing their output and them being better at managing worktrees and other tedium, but the idea of doing lots of things at once feels like a mental muscle that was previously weak and has gained in strength in ways I never though possible for me.

Ahh, I started my freelance career on a Portege R100 running Arch back in the day. The amount of times I missed meetings because of X11 failing to run or wpa_supplicant shitting the bed...

There's also Usenet.

Broot (br -w) can also do this (not with the big rectangles)

Ah so a little more serious than gang rape, I guess. https://www.bbc.com/news/live/cd0m38xndp3t

The margins/padding on stuff and the sizing is very haphazard.

Neat, could it leverage https://github.com/apple/container

Doesn’t Eastern Ukraine have more resources and more ethnic Russians?

Israel wants the Palestinian land. Many of them would be happy to raze Gaza to nothing and rebuild it.

I mean, all of it? Maybe someone born post rewrite will think wow AI made a cool thing but nobody is going whoa Postgres is made purely by AI?

You really think people should hide their honest opinions about others because they receive money from them?

What an awful corporate mindset.

But that's the thing, without the decades of work, it wouldn't BE trivial.

Everyone is standing on the shoulders of those which came before. If LLMs allow us to combine the incredible decades of effort and knowledge and experiences that's gone into building something as great as Postgres, and take that and combine the experience and philosophy that has led to the creation of a language that potentially provides tangible benefits, and for far less human time and effort that it would have otherwise taken...surely something that should be celebrated as absolutely incredible?

They aren’t purely for that, they also contribute to how an application feels to use in a creative manner.

Careful with that dangerous knowledge, you’ll end up in a list.

I asked it what the cheapest train fare would be for my partner to get somewhere and it hallucinated the two together railcard rules to the point it would have got us a fine. That said, British train fares are arguably more convoluted than even the most complex software application.

Ah, thanks for your write up - was just reading it when I saw the comment.

This has royally fucked a lot of people.

Doesn't Postgres' UUID type just do this for you anyway?

Why would you store it as as str column and not the inbuilt type for this?

https://www.postgresql.org/docs/current/datatype-uuid.html

If you are using SQLite well I guess that doesn't work.

Nope, the public repos are what the on-machine payload creates. Sorry, I worded that wrong, I meant it exfiltrates to.

The main attack is using compromised repo keys to:

* Create malicious actions to JSON dump and exfiltrate all GitHub org secrets.

* Commit the payload delivering hooks/scripts to any repo/PR it has access to.

* Mimics previous commits/timestamps, however you can see the key that did it by seeing the push in activity/audit logs.

We're trying to figure this out, we've nailed it to the developer machine but not sure how that machine got infected. Possibly through GHA.

Attack is called "Hades - The End for the Damned", it exfiltrates secrets including ALL ORG GITHUB ACTIONS SECRETS via creating compromised actions, through GitHub public repos with encrypted payloads.

Nah in competitive industries you need to build features and out compete people and getting AI to do that whilst architecting things well due to experience and having time to think more about the important stuff but have a lot of the more boilerplate and simple things ABs plumbing etc handled by agents is great.

When you try to replace your entire brain with AI things are going to go wrong.

Our company started monitoring our Claude usage so I've started coding personal stuff manually again and it's...really fun!

Yeah, this is pretty good devex from the hackers.

I would prefer my builds to break than the ecosystem to be compromised.

That said, once unpublished the version should be permanently unavailable to prevent publishing over known good versions.

Am I going to have to make it rewrite all the stuff 4.6 did?

It's free and simple and handles HTTPS termination and can be set up easily using terraform/pulumi.

Interestingly, in the early hours of this morning I switched from Cloudflare Tunnels to a rathole/traefik based solution (well, currently it's port forwarding and a low grade home-baked dyndns solution until I get paid and can afford a cheap hetzner box because I spent all of my money again).

I switched back because I didn't like the added complexity of having to manage the routes, what I'm using it for is technically against ToS, and I like the self-contained nature of my microk8s cluster.

I actually really like not having to worry if some licensing deal means my access to music I love gets shut off.

Can tailscale funnel do custom domains yet?

Personally I'm switching to rathole+traefik, weirdly something I was researching and experimenting with in the early hours of this morning (I have now not slept and have to go to work).

It is nice to download an app and then just point it at a public URL as opposed to having to rely on the device being in the same tailnet.