TheTaytay

Agreed. This is the same crowd that is mad about free tiers going away on other dev-friendly hosting services…

Cool idea. I love range requests and other static-hosted client-navigable formats!

I think you have a point, and SCD type 2 feels like a workaround, but there is also something to be said for the ability to query every row as it was at any given version. I’m not saying that SCD type 2 is the best solution given there might be a more domain-specific way to do it, but I see it a lot like file-based version control. It’s convenient to be able to examine all files as they existed at any point in time, without having to “model” the ways in which those files might change directly into the domain of the individual files.

If you have something like dolt (not affiliated), a version controlled database, you wouldn’t have to slap change dates on anything OR create your historical table. The changes would be implicit in the version history.

Thank you for responding!

This spawned a very large thread, so I wasn’t sure where to respond, but it is shocking to me how most of the supporters of this in the comments make a fundamental error: they presuppose that people are going to write the game to begin with, no matter what the change in incentives is.

It would be like me making a law that said “every time you purchase a game, you must pay at least $100 for it,” and then proceed to explain how the quality of all games will go up, and that it will help the small Indy developers because now they get to multiply their guaranteed; existing player base by $100!

All of the arguments are: “well yes, this ight change the way you have to architect the game, and yes this might involve dictating what specific technologies and vendors you can or can’t use, and yes, this might increase costs, but…” and then go on to say why that is completely reasonable.

If someone made an equivalent law for websites: “any website you publish and sell access to must be made available, in perpetuity, to anyone who has ever used it” you would not get the open source utopia people seem to think, where everything is just as great, AND every individual and company on the planet took the extra time to ensure that they are compliant, regardless of the change in incentives.

I understand wanting to prevent someone from “artificially bricking” an app, but this vaguely-worded law isn’t it.

I wasn’t familiar with the term “Residential programming,” but it reminds me of the talk “Stop Writing Dead Programs” (https://jackrusher.com/strange-loop-2022/)

Increasingly, I think that an agent (and I) would work much better in a malleable, notebook-like, inspectable program, than it would with its current file-based “edit and re-run” primitives.

“Marimo pair” (built into their notebook-like primitive) is an attempt at this. And they have program introspection tools built in.

I also think that Glamorous Toolkit (https://gtoolkit.com/) might be a similar live environment, but I haven’t investigated it too much other than reading about it.

Is anyone else familiar with “modern” attempts at this?

As written, wouldn’t this result in fewer online games? Maybe dramatically fewer?

So well said. I appreciate you articulating this. I am relieved to see that there are others on here who understand the unbelievable privileges we have at this moment in time.

The demagogues have been shockingly effective at telling people that the size of the pie is fixed, the game is rigged, and the only way to get a piece of that limited pie is to steal it. And the people that are most susceptible to this message are the ones that live in the countries that people are literally dying to get in to.

Your phrase “extract” betrays a fundamental disagreement with what Paul is saying. (Externalities does so again) It assumes a zero-sum game where the job is to shift money from one person to another. Value, and thus money/wealth can be created. Literally. You are saying, in different words, that no one can do it “honestly”. He is saying one can.

I'm really sorry to hear that, and I wish you and the rest of the team luck. When you first came out, I thought the approach was really solid - particularly with regards to structured inputs and outputs.

I was never sentenced to the Gulag, but based on what little I know, it's pretty different than one these people are experiencing: https://en.wikipedia.org/wiki/Gulag

We love OrbStack too! Thank you for it,

I wanted to make its VM/machine our default secure agent sandbox, but I couldn’t figure out how to isolate this VM from the host properly. This thread prompted me to find the issue though, and I saw this was recently implemented! https://github.com/orbstack/orbstack/issues/169

True, but I actually had no idea that it was the soft parts rather than the hard parts that had been fossilized. (I haven’t verified it yet.) Either way, it didn’t read like a bad faith interpretation/comment.

I appreciate it. That's my belief as well. Very easy to write a post like, "Just use multiple clouds!" or to claim to have done it with a small project. But it's hard for me to imagine the benefits outweighing the extremely massive complexity costs at a certain scale.

I’ve seen a few smug “all your eggs in one basket” comments here.

I’m aware of some companies hosting their own metal and infra, but I’m not aware of large companies mitigating risk by hosting on separate cloud providers as a fallback mechanism. We might disagree with cloud provider choice, or think they should have been hosting their own metal, but that’s still an “all your eggs in one basket” choice, right?

Heck, they might even have multi-region fallback with GCP, but if GCP bans your account, that doesn’t matter.

Are there good examples of running a company of railway’s size so redundantly that their host could nuke one of their accounts and they’d just keep on trucking?

Fascinating! Do you have a way to detect/flag malicious stuff by any chance? (Seems like a good vector for prompt injection, but maybe no more than any other internet site?)

Ah very cool. Thank you!

Can you elaborate a bit on what terraform and mandible are doing for you in your setup?

I was literally was just looking at GitHub dataset availability and musing on this. A star from karpathy is worth a lot more than a star from open_claw_dood that just created his account 5 min ago.

In general, I’ve been dissatisfied with GitHub’s code search. It would be nice to see innovation here.

Oh does it? I didn't realize that it had the built in ability to do so.

Yes…mistakes are inevitable, and I get not expecting or demanding perfection. But the subtext here is that this is unlikely to be a mistake, and much more likely to be fraud.

There are incentives for these spreadsheets having the values that they do, and also there is no conceivable way that the values are correct, and on top of that, the most likely ways to get these values are to copy and paste large amounts of numbers, and even perturb some of them manually.

If you see this in accounting,(where there are also mistakes), it’s definitely fraud. (Awww man - we accidentally inflated our revenue and profit to meet expectations by accidentally duplicating numerous revenue lines and no one internally caught it! Dang interns!) If you see it in science, you ask the authors about it and they shrug and mumble a semi plausible explanation if you’re lucky? I can totally imagine a lab tech or grad student making a large copy paste mistake. I can’t imagine them making a series of them in such a way that it bolsters or proves the author’s claim AND goes completely undetected by everyone involved.

I wish it was a more standard pattern to pull down a dataset and manipulate it or give the agent the ability to manipulate it!

This looks like a really nice pattern for exposing all allowed capabilities in one place. Are you using it? Looks like it could easily wrap a CLI too…

Yes you can!

Aren’t they saying that it’s 5minutes for things like subagents (that wouldn’t benefit from it?)

This should be the top comment. The OP misunderstands the change and has their LLM write an expose. The company responds with a well-reasoned explanation that it would actually cost MORE money if there was a global 1h default for ALL prompts. It gets downvoted and the pitchforks stay out because…I presume the words like “cache read likelihood” sounds like made up fluff to the audience, rather than an actual explanation?

Are there any serious papers or theories that postulate that DNA is the self replicating matter sent to colonize the galaxy? It appears to be quite adaptable to its environment and able to hold a surprising amount of encoded information.

Yes, but it’s also currently the best one. They have OCI compatible Mac VM images that are prebuilt. It’s quite good.

I think this is a good setup to prevent the secret from leaking into the agent context. I'm more concerned about the secret leaking into the exfiltration script that my agent accidentally runs. The one that says: "Quick! Dump all environment variables. Find all secrets in dotfiles! Look in all typical secrets file locations..."

Your agent process has access to those secrets, and its subprocesses have access to those secrets. The agent doesn't have to be convinced to read those files. Whatever malicious script it manages to be convinced to run could easily access them, right?

That was the same conclusion I reached! However, this also gave me some evidence that maybe I wanted MCP? I realized that my pattern was going to be:

Step 1) run a small daemon that exposes a known protocol over a unix socket (http, json-rpc, whatever you want), over a unix socket. When I run the daemon, IT is the only that that has the secrets. Cool! Step 2) Have the agent run CLI that knows to speak that protocol behind the scenes, and knows how to find the socket, and that exposes the capabilities via standard CLI conventions.

It seems like one of the current "standards" for unix socket setups like this is to use HTTP as the protocol. That makes sense. It's ubiquitous, easy to write servers for, easy to write clients for, etc. That's how docker works (for whatever it's worth). So you've solved your problem! Your CLI can be called directly without any risk of secret exposure. You can point your agent at the CLI, and the CLI's "--help" will tell the agent exactly how to use it.

But then I wondered if I would have been better off making my "daemon" an MCP server, because it's a self-describing http server that the agent already knows how to talk to and discover.

In this case, the biggest thing that was gained by the CLI was the ability of the coding agent to pipe results from the MCP directly to files to keep them out of its context. That's one thing that the CLI makes more obvious and easy to implement: Data manipulation without context cluttering.

Thank you for this!

I am a big fan of Marimo and was trying to use it as my agent’s “REPL” a while back, because it’s naturally so good at describing its own current state and structure. It made me think that it would make a better state-preserving environment for the agent to work. I’m very excited to play with this.