frisbee6152

For technical work, I’m inclined to agree. I’m a fan of Luna and Sol, and Astra has its place, but I’ve always found Terra an unhappy medium. If Luna is not good enough for a task, then the marginal improvement of Terra tends not to be worthwhile. If it is worthwhile, then the jump to Sol tends to also be worthwhile, so I tend not to use Terra.

While it does offer some meaningful improvements over 6-Luna for a few workflows, I suspect most of those benefits stem from 6.1-vs-6, as opposed to Terra-vs-Luna.

I’ve switched to a higher tier plan where I don’t hit limits and have found it less habit forming and more pleasant. I’m not going to hit the limit anyway, so there’s less reason to alter my behavior to match quotas.

The end result is that I’m considerably under quota every week, so I’m sure margins are comfortable for OpenAI. But it means my work doesn’t get interrupted, and I can use heavier models and fast mode liberally, because I’m not going to hit quota anyway, which means my sessions are shorter and more productive.

It’s hard to get anything “done done” when you get a quota limit, just as you’re getting into a good flow.

Nope. Just watching the volume and severity of CVEs coming through since it’s been running. It’s been a busy few months.

The article says in the second section that the author did not have access to Mythos. I think it’s dangerous to rely on claims made by others without even bothering to read them first, let alone check.

It found hundreds of vulnerabilities in Firefox, according to Mozilla: how does Mozilla benefit? It found a 27 year old vulnerability in OpenBSD. How do they benefit from that? Is that made up? Are the maintainers of those codebases lying for the benefit of Anthropic’s IPO? Is copy fail a fabrication by big AI? The 12 OpenSSL vulnerabilities found in January?

https://venturebeat.com/security/mythos-detection-ceiling-se... https://www.wired.com/story/mozilla-used-anthropics-mythos-t... https://cyberscoop.com/copy-fail-linux-vulnerability-artific... https://www.schneier.com/blog/archives/2026/02/ai-found-twel...

Im not sure whose claims you think I’m relying on. I trust Firefox that they’re not overstating the number of CVES they’ve found. Same for OpenSSL. The OpenBSD folks definitely don’t seem like the types. I’ve not known Linux to fabricate CVEs either. I think my sources are fine.

The most important thing I would point to is Mythos et al and the wave of vulnerabilities that have been discovered in the past couple months. It’s a completely unprecedented event, brought forth almost entirely by improvements in the models themselves. That said. keep in mind, I’m talking about over the past two years. With Claude code and the capabilities gained since December of last year, there have been incredible gains in the capabilities that are now available. Demand for inference is higher now than it was a year ago, because capability has improved. A specific criticism that I would hold is that claiming that progress with LLMs is slowing, prior to that point, is embarrassingly wrong in my view. One could argue that the model capability improvements are slowing, and all the improvements were in harnesses. I think that’s a stronger argument, but I have a few problems with it. 1. Utility is utility. Whether that comes from the model or the harness is irrelevant when making claims about utility. I don’t think that’s a useful distinction most of the time, but especially when talking about the technology as a whole. 2. Marginal intelligence gain is different than marginal utility gain. It’s estimated that intelligence grows logarithmically relative to investment. However, the utility of a marginally more intelligent model may grow exponentially, because once behavior crosses a reliability threshold, it unlocks new capabilities. 3. Even on those terms, it’s not clear to me that frontier capabilities are slowing down. With Mythos and its contemporaries, we have been seeing a vast change in the security industry as vulnerabilities are discovered at an unprecedented rate. OpenBSD vulnerabilities, more Firefox vulnerabilities found in a single month than the past two years, critical Linux vulnerabilities. It’s hard for me to look at the effects there, a radical new capabilities baked into the model itself, and see stagnation. A part of the reason it might feel like it’s slowing down is because we plebs don’t have access to the top models.

He’s been continuously predicting that the collapse was just around the corner, that progress was slowing, and that there was no market for inference, since 2024.

The fact he’s never reflected on the glaring failures in his analysis tells what we need to know about his intellectual integrity. There’s truth in some of his words about financial risk, but if you can’t acknowledge that there’s upside too, you can’t evaluate risk properly either.

I find it difficult to take him seriously.

What did you switch to, and what do you like about it?

Just use a slider, duh

A well-optimized program is often a consequence of a deep understanding of the problem domain, good scoping, and mindfulness.

It often feels to me like we’ve gone far down the framework road, and frameworks create leaky abstractions. I think frameworks are often understood as saving time, simplifying, and offloading complexity. But they come with a commitment to align your program to the framework’s abstractions. That is a complicated commitment to make, with deep implications, that is hard to unwind.

Many frameworks can be made to solve any problem, which makes things worse. It invites the “when all you’ve got is a hammer, everything looks like a nail” mentality. The quickest route to a solution is no longer the straight path, but to make the appropriate incantations to direct the framework toward that solution, which necessarily becomes more abstract, more complex, and less efficient.

I was on one of those robotics teams, and I recall this document being passed around as suggested reading on that team. It’s probably worth considering that this was written in 2000, when it was a more niche topic, and was probably itself a part of what made PID programming more accessible in the first place.

I second the “experimenting” thing. I would recommend going caffeine free for a couple weeks every once in a while, just to kind of… keep track of the addiction. I’ll quit coffee occasionally, historically usually for a tolerance break if it’s getting a bit too much, or to just kind of “reset” my brain chemistry.

I ultimately found that reducing was a good idea, but I personally like coffee a lot and feel it adds something for me. I also have adhd though, so I’m probably playing with a slightly different deck on the “how stimulants interact with the brain” front.

I have found that having controls and limits around caffeine intake is a pretty good practice. I found that to reliably sleep well, I need to never drink caffeine after noon. And reducing my intake a bit helps as well. But from there, I haven’t seen much further benefit from quitting entirely.