oulipo2

It could still be useful if the spec is roughly as hard as a simple implementation, in case you have automated methods to find more efficient implementations, guided by the constraints of the spec

Okay, so it allows to run the same image, but is not CLI-compatible with docker that's what you mean? But is it more / less / equivalently efficient ?

How does it compare to something like OrbStack?

Problem is, usually describing the problem you want to solve *correctly* using formal tool is a task as hard (and often, equivalent to) the implementation. That said, having a formal description is useful

On the 1-element monoid it's trivially true

When I say "novel ideas" it means something groundbreaking, indeed, not rehashing common "this is the best practice for a simple CRUD backend"

Exactly, LLM is good at "code inpainting" : define clear structures and goals, and it will fill the boilerplate. But it doesn't work for reasoning and abstraction, so it fails to synthesise and propose novel views. But that's integral to the way it's designed and has been trained, to do a kind of "averaging" which limits it's capacity to explore novel designs

Yes. Because a few golf course are produced every 10 years / 20 years or so. But people generating videos is going to increase exponentially. So this is going to be a huge difference

You realize that "golfs are even worse so why people complain" is not a serious argument?

The fact that EVERYBODY here sees it but you, tells you that perhaps you should take a step back and try to understand... it might be YOU who don't get it at all...

Yes, even reworking a classic "infra diagram" with connection between a web browser and a server, but showing it as an animation could be interesting, but those little stories add very little

The fact that people in the tech-world don't realize that they are embodying deeply ingrained sexist / toxic behavior is the reason why women aren't more present, and why people like Musk or other sociopaths have been celebrated by this community, while the rest of the world hate them...

Same for Elon and Paypal

If you're using an iPhone, couldn't you automate this by extracting "Live images" which are kind of "mini-videos" around the photo you took?

Interesting, so you can bind a Clickhouse table as an extension to Grafana? Would you make a little Gist / post about it to show?

Interesting, have you also looked at other formal methods, like Abstract Interpretation?

How do you use it for visualization? Do you use ClickStack? or something else?

Meanwhile, the fragile-ego loser Musk is hoarding wealth and trying to destroy the planet

I guess they're getting bought because they had access to a lot of codebases from a lot of companies, and perhaps there's something to mine in those logs...

Okay, now try applying your thinking to abusive healthcare insurance companies and see how well this fares...

Did he... totally miss the point of AOC?

She said that no amount of "work" can "earn" billion dollars. So you earn billions through capitalism by rent, through owning companies, and having other people work for you. And ultimately, it should be obvious that societies have to cap this

We can TOTALLY say that there's a causal link, it's been proven by scientific stories

The frequency and magnitude of the event is directly related to the warming up of climate

Have you heard of "rebound effect"? Sure you can say, individually, one query is not that much... but then it becomes integrated in search engines, so suddenly when there was no queries at all, now there's 500 billions per day, and it gets included in your CICD at every commit, and soon enough in your OS, etc

The real point is not "one session", it's the fact that people now do that routinely, that CICD are using those to check every commit, and each search engine query now does that too, so it multiplies

Right = "let me do this thing for myself to get ahead, and then forbid everyone else to do the same"

That's a global problem everywhere. But when you're traveling locally, you go there for the landscapes / buildings / history / cuisine. Not necessarily to hang out with people on TikTok

"The principle is that the agent doesn't need code changes, including skills/MCPs - it just accesses systems."

That's why you're having safety issues.

The real (and boring, and tedious) way to do it IS to create a unique way (API, MCP, whatever) for the agent to access your data / infra in a secure way.

Think about it as "typing" in language. Sure it's boring to have to put all the type info (even though in many case it makes dev easier too, because it forces to construct stuff cleanly), but then once it typechecks, you're relatively sure that it's doing what it's supposed to.

Here it would be the same. You build basic building blocks that you know are safe for the agent to access, and you let it compose them

It would be able to spawn psql correctly, but wouldn't be able to connect to the database (if it's secured with user accounts). It would only be able to use the database through the MCP (which uses a read-only account to connect).

I understand the "centralized registry" thing, but it's also easy to "forget about one case", and agents are good at circumventing stuff ("oh, I cannot DROP table, let me just remove all rows", etc). So I'd rather trust the permissions of the original db (eg getting a read-only account) which I presume have been battle-tested for this

So, why not instead limit your agents to a few endpoints / MCP functions that you control, which give access to your db (or whatever) through read-only permissions?

It seems this is a bit like "reinventing permissions" no?