Don’t forget, “Kill the guy and get the thing, then take that to this other guy.”

This was my experience using GLM 5.1 in Claude Code but it works far better in OpenCode, I’d really like to understand why. I think it’s a bit stronger than Sonnet 4.6.

I use the oh-my-openagent planning system and haven’t used vanilla OpenCode enough to know how much that is contributing.

The answer is easy, CC is bug for bug optimized for Anthropic models. They don't even test it with other models, let alone provide support for all small compatibility quirks of different provider implementations.

On the other hand, Opencode, Pi agent and other open source tool offer much better support for all models, including open source.

There was never an era in which #2 was a reasonable policy.

The canonical answer to any concerns with the AUR is always “just read the PKGBUILDs bro”

For every single update, for all your AUR packages, all the time.

You know that thing where if you make a security review feature obnoxious, after some time people will just accept everything without even looking? Yeah...

> For every single update, for all your AUR packages, all the time.

Yes, that's what I used to do when I ran Arch. It's usually easy. The PKGBUILD is usually small to begin with and the difference for a new version should normally be something like the URL and the version number and not much else, so you can just diff it against the old version.

paru presents all pkgbuild diffs to you before installing, that's what I use to read them.

I usually only use AUR to install trusted pre-compiled binary packages, the scripts are very simple and the only thing that should ever change is the url and the sha256

Yea, paru makes it really easy, i noticed the diffs are a little easier/different versus yay. Not sure though if it's a config setting, haven't figured out the details yet.

Also paru shows you coloured code syntax if you have `bat` installed, i think.

I do it too, but I can see why this can be a problem for users. There should be an "official" scan for potentially malicious changes. I use a third party AUR scanner to help me with this.

What third party scanner do you use?

You are thinking of the alarm fatigue[1], but it doesn't apply here -- there are no constant alerts warning that you are doing something dangerous to the point you get desensitized and start to ignore them. The correct analogy here are checklists -- things that you need to check if you are to do this "dangerous" activity (AUR usage), akin to pre-flight checklist.

[1] https://en.wikipedia.org/wiki/Alarm_fatigue

Oh yeah, that's the name of it. But I guess something similar happens with checklists, you do it so many times without anything bad ever appearing that you start to subconsciously assume nothing will ever happen. Why check the rotor of my helicopter when nothing ever happened to it for 5 years? This checklist is a waste of time!

> you start to subconsciously assume nothing will ever happen

True. But you still go through the fucking checklist.

That one's survivorship bias I think.

This is perfectly safe as long as you keep your username a secret.

I try to use stars as my username, but a lot of places won't allow it

Yeah it’s one of those words that gets snapped up early, like https://news.ycombinator.com/user?id=stars

That doesn't look like a well used account - how do I get it transferred to me?

Edit: never mind, I guessed the password as it was only five stars.

The person who "writes" code is also supposed to review their own work, and answer for that. If they won't do that - well - they should be fired. But if you have weak or uninvolved leadership, then the team's only rational recourse is to shun them.

Also - at least in GitHub if you squash with the PR merge action in the UI - the original commit history is still available in the ref maintained by the closed PR yet doesn’t clutter your branch or tags namespace.

What are the numbers these days? What percent of projects are running on github versus say <choking noise> Atlassian?

I'm guessing the number of people who care about commit logs is dropping and the number who don't is skyrocketing.

mise use -g ripgrep@latest

I recently found mise and have become a fan as well. I have used asdf for about a decade and it supports the same .tool-versions files so initially I used it for those exact same things.

But I use four different computers for development regularly and sometimes use Codespaces as well. While syncing dotfiles handles most of my setup, it doesn't handle binary dependencies of my dotfiles - my neovim setup wants fd & rg etc. So now those go in the mise global config. I also have a global node & python along with uv@latest which pretty much covers every tool I might want to install.

I have never cared for the fact that homebrew tries to maintain shared dependencies and several upgrades have broken stuff for me.

GLM 5.1 is stronger than Sonnet 4.6 in my opinion, but while they have a coding plan that is a good value MiMo beats it on price. I haven't used MiMo much yet but it felt pretty similar.

But you really don't have to switch. MiMo Code has the same provider support as OpenCode.

Even Claude Code you can use with any provider that exposes an anthropic API endpoint, which they all do.

Or by using a proxy, yeah. Personally I would still prefer a multi provider harness over CC when using it with another provider, if alone for the visible reasoning, model switcher, cost estimation and so on. So far I've only preferred CC when I needed to work with Jupyter Notebooks because it has built-in tools for that.

I agree that evolution could not produce a rational agent who would still reliably respond to lower level imperatives (such as pain, hunger, lust) without consciousness and feeling. The primitive parts of the brain have to be able to override the higher functions to ensure survival and reproduction. But an LLM isn't evolved in this way; its fitted to a functional output. It is entirely possible there will never be anyone home. I sure hope there isn't, because at the scale we're using them it would be a moral catastrophe.