vinnymac

I disagree with the premise that you need to know what you’re doing to solve problems or create something new with LLMs (unfortunately).

I wouldn’t say that Dans recent work in maths was necessarily that of an expert. More of a fan of Lean than anything and he did a breakthrough in a relatively short time period.

https://overreacted.io/how-i-vibed-a-proof-of-conways-conjec...

It seems to me it’s about enough money, knowledge, and effort to want to achieve a thing. Which is good, but can and is being weaponized in terrible ways against us.

Oddly not much to trip it. I once ran “touch AKAMAI.md” and then had Claude run git status and it fell back to Opus as if I had just broken into the pentagon itself.

I tried to use this the other day and found out they still don’t support arm64 :(

Somehow ECS does though… tradeoffs everywhere I look have made me dizzy.

I find this recommendation funny, because I completely agree from a sysadmin perspective.

I used to use Gitlab at work, and small teams would run into so many footguns with CI that we had to throw up guardrails to prevent mistakes. Far too many links to gitlab issues that were not fixed even after >8+ years of being open ended up biting us. With GHA I haven't had that experience, and same for all of my self hosted Forgejo instances. I used to hate using GHA from about ~2018 to 2021, but they've fixed a lot of things I disliked since then.

If you're actively considering self hosting, I recommend giving forgejo a try. The experience is much more similar to what GitHub offers, actions API is nearly identical to name one similarity of which there are many. Moving essentially becomes one prompt and a coffee later for a small team.

Just to be clear, I am on Github Enterprise, and am also experiencing this disruption both privately and publicly on every org and project I have access to.

For me this started happening when I type "stat" too.

Which surprised me because of the many hundreds of services I rely on regularly that also have status pages.

As someone who notices a lot of things, I was talking recently with someone who notices almost nothing. When I spoke about my desire to seek out the truth, they simply replied, "I just want to live a happy life."

There’s a perception that those who seek knowledge and truth are somehow less happy. Ignorance, and the willingness to take the path of least resistance, can be seen as an advantage for those who would rather not look too closely. Fundamentally, perhaps the difference comes down to red pill versus blue pill, and the illusions we’re willing to accept.

Thanks to you I’ve now upgraded to v16, and can confirm the logs API works beautifully!

Nice! I was hoping they would add this in v16. Unfortunately I am still on v15 of all my instances of Forgejo

If you know GitHub actions then you’ll immediately understand Forgejo actions. It was designed that way intentionally. There are some differences, but at least for me not enough to warrant any pitchforks.

If you have advanced use cases you might be more frustrated, but I’m not aware of any off the top of my head. I think my biggest complaint is that they haven’t exposed action logs over the API, so I can’t build tooling around them at the CLI level, feed them to an LLM, or more quickly diagnose problems that arise without using the website.

I’m hosting my own https://tangled.org knot and spindle, and also hosting and building my own Frontend browser for https://forgejo.org looking forward to open sourcing it in the future for others to enjoy.

For people who want a similar experience today, I’d recommend trying Codeberg (An official instance of Forgejo) though. However, keep in mind they voted in July to ban mostly AI generated repositories. You can read about that here: https://blog.codeberg.org/protecting-our-floss-commons-from-...

Same, I have gotten good results out of Opus 4.6,4.7,4.8 for my security work though. So I continue to use them for this.

Curious if you’ve find yourself enjoying DS or Kimi more than Opus 4?

I once described Mythos and then Fable as escaping the box we put them in. This would certainly fit that narrative.

> For a long time they would have global outages where the status page would list everything as green and the only reason you'd know something was wrong was the forum threads. Their response was that they were too busy fixing the issues to update the status board. Right.

Are you talking about AWS or Fly here? I seriously cannot tell.

I created a web-of-trust forum built on ATProto. https://caterpilla.rs

Probably one of the most fun projects I’ve ever worked on, and I largely attribute that to the protocol.

I’m particularly interested in the permissioned data spec, but don’t want it to hold me back, so some functionality is on ice, and I went live without it.

Yesterday I gave Claude Fable a difficult task. It then proceeded to spawn 415 agents. It got it done, but damn was it expensive.

I’m working on an open source Forgejo Frontend that exposes additional features like this, and offers a better user experience, faster large diffs, and basically fixes every little thing I don’t like about Forgejo. Would be interested in hearing more of your complaints so I can continue to improve.

No doubt in my mind, a future Apple model will be the best to use for this purpose. They likely have more swift to train on than anyone else, and would benefit directly from more quality apps, rather than the slop flowing into the App Store (>1k app submissions per hour; they claim)

Well said. In creative spaces they talk about “Dirty” vs “Clean”. Dirty they say lets you move fast. Clean is slow.

Happen to be a startup that isn’t mission critical to someone’s health and well being? Great, now you can use AI and be as dirty as you would like.

Are you working with dangerous chemicals that are ingested by others, or systems that control hunks of metal flying through the sky with hundreds on board? Maybe we should stay clean in those environments until we make AI itself clean.

Curious, why do you say that as if the "I'm Feeling Lucky" button isn't still on the homepage in 2026?

Same here, I now self host Forgejo and Tangled Knot + Spindle.

Only use GitHub for employers at the moment, but don’t intend to put new code there going forward.

I’m actively working on a free and open source alternative Frontend for Forgejo that I self host called Joui

I’m in the process of developing an alternative Frontend for Forgejo that’s incredibly fast, and works perfectly well on Safari and Firefox.

Here is a screencap of the wip mobile UI on old safari: https://files.catbox.moe/bo7pxn.jpeg

I’ve always felt this automation shouldn’t exist at all, but should rather be selectively controlled via a hook. The hooks yarn offers out of the box for example can be used to run any code you need to after install. Putting the project owner in control instead of the dependency.

In case others are unaware, you just have to set https://yarnpkg.com/configuration/yarnrc#npmMinimalAgeGate to the value you want. It defaults to 1 day.

It’s on by default in yarn 4 too now, but pnpm was the first to market that default minimum gate.

https://github.com/yarnpkg/berry/pull/7135

I tried it but see a lot of false positives.

One funny thing I see it doing is deleting seemingly random comments lines, for example if a file has a comment that spans multiple lines but doesn’t use a multi line comment syntax. It just chooses one at random transforming the once useful comment into slop.

The problem is that we have an ever growing and large number of constraints, and not following even a single one means the result is sloppy.

I don’t see them fixing this any time soon, and thus human in the loop is a requirement to use these tools effectively. That is unless you love your slot machine dopamine rush enough to ignore quality gates and respect for your peers time.