Nick Guerrera

This too shall pass.

You keep using that word "config", but I don't think it means what you think it means.

14 Sept

In my personal projects, my new productivity is mostly consumed by fixing genuine bugs the AI finds that I never would have until much later. I wish we would spend more time talking about making things *better* than making more things or making things faster.

14 Sept

This is the opportunity we always wanted to do things right! Don't squander it.

Nick Guerrera reposted

P(aul) Frazee

@pfrazee.com

13 Sept

In 2027, mysterious hacks attributed to open source weights and definitely not connected to the companies that accidentally hacked a bunch of companies in 2026, which was okay

Quantian@quantian.bsky.social

lol two months ago OpenAI was signing a letter about how important open weights models were and now their house propagandist is saying “we will ban open weights and you will thank us for it”. You can’t hate this company enough

11 Sept

I will not do 2020 all over again. The internet never learns. Enjoy debating about how likely it is that the world is going to end, I will not be participating. Bye.

11 Sept

I've resisted tiling window managers after many attempts over 25 years. I have now discovered paperwm and it is just brilliant. One of the best demonstrations I've ever seen of thinking about an old problem in a novel way.

11 Sept

I guess I have to try niri (inspired by paperwm) now

It's more interesting to me that in 2026, there are people who still use (Free|Net|Open)BSD than 2026 being yet another possible Year of Linux on the Desktop. I'm happy because if we ever need to hide from Skynet, I think BSD is what we'll run.

I've never been a FOSS purist, but being able to see all the code that runs on my machine has never been more practical. 1. Agents can diagnose everything that goes wrong by reading code. 2. I can look into what an agent might do if I'm worried about giving it access to something.

I'm actively removing proprietary software where possible. A zero proprietary software standard is still out of reach for me, but also closer to my reach than ever.

Social media is a training environment for natural intelligence that rewards extremism.

I’m a naturalized American citizen born in Canada and happily living in Minnesota. If we don’t win big in November, I think it will be time to seriously discuss leaving the US and joining Canada. Our people are so compatible and we’ll never lose again in international hockey.

For the love of God, stop assuming everyone wants dark themes. Some of us have trouble reading dark themes. It's accessibility and inclusion fail to make things dark-only.

Also, "light themes blind me" -> try with the brightness at a normal setting instead of sell-monitor-in-retail-store mode.

And if they still blind you, I support your choice, just don't impose it on everyone.

devenv.shYour whole development environment, declared.Define packages, languages, services, tasks, secrets, and tooling once. Give every developer—and CI—the same fast, reproducible environment.

6 Sept

Omg, they ported Visual Studio to Bourne shell? devenv.sh 😉

Nick Guerrera reposted

Krzysztof Cieslak

@kcieslak.io

6 Sept

Computer use is architecturally terrifying and is disgusting way of using agents. But it just shows how bad were foundations we've been building software on for all those years - lack of compossibility, programmable APIs and walled gardens just came back to hunt us.

6 Sept

Seems I forgot I hate it online.

flyingpenguin.comDHH Nazism Funded by 1Password VP Who Wrote “Honest Security” | flyingpenguin

5 Sept

It fills me with anxiety to have to swap this particular piece of software, but well, shit. What's the best alternative to 1Password?

Miguel de Icaza@migueldeicaza.bsky.social

If the hubris that lead to security holes and poor engineering of doge/linux did not dissuade an enthusiastic friend, perhaps not supporting the Nazi ideology behind it will: www.flyingpenguin.com/dhh-nazism-f...

jakelazaroff.comDHH Is Way Worse Than I Thought | jakelazaroff.comDHH's politics are not normal. Maybe they used to be, I don't know, but as of right now the dude is_way outside of what most people would consider moral or acceptable.

5 Sept

Bitwarden appears to have everything I want. Please tell me their security is good and they don't give money to the scum of the earth.

Nick Guerrera reposted

Jake Lazaroff

@jakelazaroff.com

2 Oct 2025

show this to anyone else you know who also assumed DHH was just a normal conservative dude rather than a raging white supremacist lunatic

4 Sept

Now I’m 99% sure on this.

Nick Guerrera@nguerrera.dev

I am willing to bet there were people inside OpenAI who pushed for stronger containment of models under test and were shut down and told the plan was good enough.

4 Sept

This! A thousand times, this!

This post is unavailable.

I’m relieved that this got a lot of positive engagement and minimal trolling from haters. For a second there I thought I’d have to leave this place and I like this place so that would have been sad.

Nick Guerrera@nguerrera.dev

NGL, watching Bluesky bully a software developer for using AI makes a part of me question whether some of the worst people that I don’t agree with on much of anything are right about this place.

The easiest way to kill your AI productivity is to let one agent nerd snipe another with adversarial review.

Successfully avoid being made a manager for 23 years, retire and..

So I just discovered that Ubuntu 26.04 ships with `ls` version 0.8.0 rewritten in rust. My new distro will be called GETOFFMYLAWN/Linux Started working on new AGENTS.md for it: “You are old, you are tired…”

Fine, rewrite it in rust but I discovered this because --group-directories-first was silently broken. I don’t know man, a regression in, checks notes, ls?!?

Bliss would be a time machine that takes me to 2006 and I bring codex with me and don’t tell a soul about it, just use it to prevent all the dumbest software decisions of the last 20 years.

www.youtube.com/watch?v=0HHu...

youtube.comAI safety is doing cybersecurity's jobYouTube video by Zack Korman

Tough pill: Yesterday on twitter the algorithm served me thoughtful debate about cybersecurity, AI safety research, and the distinction between them. And the algorithm here drew me into a fight over whether one person can use AI to make a prototype for a difficult feature in software they give away.

Nick Guerrera reposted

Replying to @nguerrera.dev

You can chase some people back to Twitter and the staunchest allies into unemployment. Is that what you want?

Nick Guerrera reposted

Replying to @nguerrera.dev

Think really hard if you want all the software in the world to be written by the people who would never come to Bluesky. Because the bullying on here could only achieve that at most. It can only make everything worse.

NGL, watching Bluesky bully a software developer for using AI makes a part of me question whether some of the worst people that I don’t agree with on much of anything are right about this place.

Rick Brewster@rickbrew.bsky.social

Alright everyone, message received. I was not expecting this response, I honestly had no idea before I made the post Like I said, the code is experimental. A proof of concept showing that PDN-on-WINE is possible. It's not finished. I can literally just delete it. It won't be part of the next build.

Zero AI is not happening. It’s not possible. If we don’t shift our energy to responsible AI, then we’ll get irresponsible AI and it will be partly our fault. Don’t shoot the messenger, this is just the reality.

Both Arch and Nix are the wrong base OS for beautiful, fun, agentic Linux for everyone. No hate whatsoever to either.

While I’ve only been a user for a few days and I have 25 years (on and off) of Debian and Ubuntu in my fingers, I think atomic Fedora base is the winning move.