dmos62

Are they elected?

Kudos. The proposals are actually interesting to browse through. Good paper name by the way.

What is COM?

How do you reuse traces as benchmarks (eval datasets)? Is it about being able to synthesise an agent session from the trace, and having another agent watch the first agent's behavior? Having trouble wrapping my mind around all the little details.

Edit: I initially asked also about general tips for observing agents, but later, noticing that there were no interactions with my comment yet, edited my comment to remove that question, because I felt that it was too broad and would detract from my main question. To my surprise, both were answered by op after my edit; kudos!

I wonder if it's the kind of thing where, if it's good at self-sustenance, then you don't know about it. And, if you do, then it wasn't.

That's very insightful. This hadn't occurred to me.

I've seen a graphical designer struggle to use an analogous product. They tried prompts like "remove background from this photo of dancers, don't change their face": the model couldn't do it: it always changed the faces at least a bit. In other words, the model was not removing the background, instead it was generating the whole picture anew, but without the background, which not only seemed inefficient, but was also way too lossy.

I wonder if this model could finish that task.

The memory invalidation based on tying memories to pieces of code (e.g. functions) is the big thing here, I think, though I see why it wouldn't be front and center.

I've always found memory systems icky, because it's difficult to tell whether a memory is still valid. Also, discoverability is always a concern (when is a memory relevant). This being both a code-graph and a memory system (seemingly unrelated things), actually addresses this in a novel way.

Great video!

True. The kind of guarantees you can get when staying within the database can often solve whole categories of problems. But, DX of having non-trivial logic inside Postgres is not great either: I feel that I'm making a significant trade-off. I've also seen interesting languages that compile to SQL, would consider those in some cases.

I presume you're not talking about autonomous agents, right? Because then you could give it a set of tasks and check its success rate. But, even so, are you not giving it tasks? Are you principally interacting with it through discussions that are more difficult to quantify? Even question answering has benchmarks. I'm having trouble imagining how you use them (or "how people use them" in your words).

Why? Sounds like you don't have a way to support your viewpoint.

Care to back that up with a benchmark? I've not seen that. OpenCode is on here, not exactly dominating though: https://artificialanalysis.ai/agents/coding-agents

What you're saying is sus. If you have a harness that's a tier above frontier labs' offerings, link to it.

I'm asking if there's something like Sky Map, where you point your phone up and it tells you what sattelite you're looking at.

Tangential, how do you identify what is a satellite in the night sky? They're not shown on Sky Map, the app I use.

That's fair criticism.

Are you comparing with `curl | lynx` or with Chrome?

> how does someone write software that needs to run for ~50 years where the hardware will need to be replaced with non-equivalent, newer hardware?

Make it open-source, or don't buy without source.

Was that because there wasn't much fentanyl use or because fentanyl metabolites are somehow harder to detect?

Expand? This is part of plot to The Godfather. Not what you'd call exotic, I'd think?

The implication is that holding up middle-fingers is illegal?

Snowden and other whistleblowers, of whom there have been many (many of which paid arguably a dearer price than Snowden did), made it plain obvious to everyone that US intelligence services are doing illegal things abroad and at home. Before that, only tinfoil-hat-types thought so: now everyone knows it. They moved the Overton window a tremendous amount.

In my opinion, you underestimate the power of an individual. I'm not saying you have to be Gandhi, but he was one person. A problem with your assessment, is that if you're right, or, if everyone believes you're right, you'll have to wait for some external force to come in and change things, like a calamity or desperate revolt. By the time USSR broke apart, bread riots were happening, and, tragically, 30 years later we still have the same authoritarian, corrupt regime (in all but name) wrecking havoc at home and abroad. That's what a people get for believing in the story of disempowerment. That's my take, at least.

Yes, that's what I'm suggesting. If you feel that the problem is higher up or lower down in the institutional hierarchy, then go higher up or lower down. That's the civil approach, as I understand it.

The answer is not to avoid engaging with a corrupt institution. US is a democracy: vote, be active politically, go into organizations that need change.

I disagree. Parallelization, coordination shouldn't be model-level functionality. Further, how do you know if or when we'll get orders of magnitude increases in context sizes. Or, do you mean that the labs will just nail the perfect plug-and-play harness? That's fair, but why wait.

Yes. I want reproducability, open-sourcing, accessibility, correctness, and most of all: usefulness. I don't care how it was written or reviewed, as long as some assurances regarding above things can be made, and I don't see why LLMs would get in the way of that.

What can't be gotten rid of fast enough is the notion that having written something is meaningful on its own. Making something that looks right was a level above total novice: now it's the floor.

That's legal?