zeroxfe

I'm in that camp -- I have the max-tier subscription to pretty much all the services, and for now Codex seems to win. Primarily because 1) long horizon development tasks are much more reliable with codex, and 2) OpenAI is far more generous with the token limits.

Gemini seems to be the worst of the three, and some open-weight models are not too bad (like Kimi k2.5). Cursor is still pretty good, and copilot just really really sucks.

I've done this kind of thing many times with codex and sqlite, and it works very well. It's one prompt that looks something like this:

- inspect and understand the downloaded data in directory /path/..., then come up with an sqlite data model for doing detailed analytics and ingest everything into an sqlite db in data.sqlite, and document the model in model.md.

Then you can query the database adhoc pretty easily with codex prompts (and also generate PDF graphs as needed.)

I typically use the highest reasoning level for the initial prompt, and as I get deeper into the data, continuously improve on the model, indexes, etc., and just have codex handle any data migration.

Expiries are a defence-in-depth that exist primarily for crypt hygiene, for example to protect from compromised keys. If the private key material is well protected, the risk is very low.

However, an org (particularaly a .mil) not renewing its TLS certs screams of extreme incompetence (which is exactly what expiries are meant to protect you from.)

Also MCP is very obviously dead, as any of us doing heavy agentic coding know.

As someone that does heavy agentic coding (using basically all the tools), this is so far from the truth. People claiming this have probably never worked in large enterprise environments where things like authentication, RBAC, rate limiting, abuse detection, centralized management/updates/ops, etc. are a huge part of the development and deployment workflow.

In these situations you can't just use skills and cli tools without a gigantic amount of retooling and increased operational and security complexity. MCP is really useful here, and allows centralized eng and ops teams to manage their services in a way that aligns with the organizations overall posture, policies, and infrastructure.

Google is so far behind agentic cli coding. Gemini CLI is awful.

This part I totally agree. It's really hard to express how bad it is (and it's really disappointing.)

TBH I would have just rendered a font glyph, or failing that, grabbed an image.

If an LLM did that, people would be all up in arms about it cheating. :-)

For all its flaws, we seem to hold LLMs up to an unreasonably high bar.

At some point you need to treat people as adults, which includes letting them make very bad decisions if they insist on doing so.

The world does not consist of all rational actors, and this opens the door to all kinds of exploitation. The attacks today are very sophisticated, and I don't trust my 80-yr old dad to be able to detect them, nor many of my non-tech-savvy friends.

any more than it would be acceptable for a bank to tell an alcoholic "we aren't going to let you withdraw your money because we know you're just spending it at the liquor store".

This is a false equivalence.

Yes, agreed. In many cases, the determinism is a feature, particularly being able to store the seed for reproducibility.

usual "true random number" bullshit

What's bullshit about it? This is how TRNGs in security enclaves work. They collect entropy from the environment, and use that to continuously reseed a PRNG, which generates bits.

If you're talking "true" in the philosophical sense, that doesn't exist -- the whole concept of randomness relies on an oracle.

it's a waste of time to steer them

It's not a waste of time, it's a responsibility. All things need steering, even humans -- there's only so much precision that can be extrapolated from prompts, and as the tasks get bigger, small deviations can turn into very large mistakes.

There's a balance to strike between micro-management and no steering at all.

I've used both gVisor and microvms for this (at very large scales), and there are various tradeoffs between the two.

The huge gVisor drawback is that it __drastically_ slows down applications (despite startup time being faster.)

For agents, the startup time latency is less of an issue than the runtime cost, so microvms perform a lot better. If you're doing this in kube, then there's a bunch of other challenges to deal with if you want standard k8s features, but if you're just looking for isolated sandboxes for agents, microvms work really well.

It's waaay better than GLM 4.7 (which was the open model I was using earlier)! Kimi was able to quickly and smoothly finish some very complex tasks that GLM completely choked at.

It seems to work with OpenCode, but I can't tell exactly what's going on -- I was super impressed when OpenCode presented me with a UI to switch the view between different sub-agents. I don't know if OpenCode is aware of the capability, or the model is really good at telling the harness how to spawn sub-agents or execute parallel tool calls.

That tends to work quite poorly because Claude Code does not use standard completions APIs. I tried it with Kimi, using litellm[proxy], and it failed in too many places.

I'm not running it locally (it's gigantic!) I'm using the API at https://platform.moonshot.ai

Running it via https://platform.moonshot.ai -- using OpenCode. They have super cheap monthly plans at kimi.com too, but I'm not using it because I already have codex and claude monthly plans.

I've been using this model (as a coding agent) for the past few days, and it's the first time I've felt that an open source model really competes with the big labs. So far it's been able to handle most things I've thrown at it. I'm almost hesitant to say that this is as good as Opus.

I wasn't trying to disprove your point -- just calling out that the scope of "economy" is broader than "monetization".

Why is fungibility necessary

Probably not necessary right now, but IMO it is an emergent need, which will probably arise after the base economies have developed.

"Economy" doesn't necessarily mean "monetization" -- there are lots of parallel and competing economies that exist, and that we actively engage in (reputation, energy, time, goodwill, etc.)

Money turns out to be the most fungible of these, since it can be (more or less) traded for the others.

Right now, there are a bunch of economies being bootstrapped, and the bots will eventually figure out that they need some kind of fungibility. And it's quite possible that they'll find cryptocurrencies as the path of least resistance.

You're in luck -- /experimetal -> enable steering.

There's a world of a difference between "at the highest level" and your typical casino poker game. (GPs general point still stands.)

TBH, that's pretty much the stack I'd pick if I were building anything new by hand. If you look at the site, there's a lot going on and Next + React + Tailwind does not seem so crazy.

These are all quite reliable well-understood components, and far from "chasing trends" IMO.

The last time I was there, there were many layers of encap, including MPLS, GRE, PSP, with very tightly managed MTU. Traffic engineering was mostly SDN-managed L3, but holy hell was it complex. Considering that Google (at the time) carried more traffic than the rest of the Internet combined, maybe it was worth it.

Plex is just sugar on top of file sharing.

right, like browsers are just sugar on top of curl

If Go had rust-style ADTs and pattern matching, and some parallel of "?" to short-circuit error handling, I'd be thrilled.

Not everyone benefits from this tech-driven world.

Nobody is claiming that.

I think that the wax and wane of confidence is a natural (and healthy) cycle to preserve.

You start knowing nothing, and have no confidence, then learn stuff and your confidence builds, then you capitalize on that (learning x confidence) by doing productive things, then you hit barriers that create confusion and break down your confidence, at which point you realize you knew nothing and start learning like a beginner again.

Unhealthy confidence does not have this pattern, whether it's low or high. Even "medium" confidence, IMO, signals some kind of rut that one is stuck in, which tends to stifle creativity.

jeez, this is a fun website, can't believe how quickly we're godwining here!

And, you as the client can verify whether it works

How do you do that? Cookies are typically opaque (encrypted or hashed) bags of bits.

Yeah, I was thinking the same. I have two Dyson vacuum cleaners, one purchased about 9 years ago, and the other two years ago. Both are excellent, and I still use the old one for my basement.

This really depends on the failure modes. In general, humans fail in predictable, and mostly safe, ways. AIs fail in highly unpredictable and potentially very dangerous ways. (A human might accidentally drop a knife, an AI might accidentally stab you with it.)