I do a similar thing where the agent runs in a Docker container and I talk to it with Telegram. It has GitHub CLI access but only with a very restricted PAT. No bind mounts. Jira is pretty clever, though I'm not feeling enough pain with just Telegram to want to try switching at this point.

I have multiple relatively well-established Jira projects I've been able to add agents to, and also clone/use as a template for new agent-only projects which give me another kanban to manage, pretty comfortably ..

The big thing about my Jira use besides the fact that its a historical tool into which I've integrated agents, is that managing agents through Jira's custom workflows is really, really cool. You can actually do any of the old workflows with agents - they'll just do it. Finally, effective waterfall! ;) *Just kidding, I've always been able to do waterfall properly...

If "ledger on card" interests you, then you might enjoy Japan's FeliCa cards. They store balance locally on the card so you can pay very quickly, no network required.

Do these cards solve the electronic cash problem (in a completely different way than cryptocurrencies)? What I mean is that

- Are the card readers special/trusted issued by bank/govt in some way? Or you can transfer money from one card to another yourself fully offline?

- Is there any requirements that the transfers have to be eventually communicated to the bank by one of the parties to be fully resolved?

- Has someone managed to create fake cards with fake money in it, or this is impossible by design?

As I understand it, these cards work basically the same way as transit card systems in other countries, like the SF Bay Area's "Clipper" cards.

The overall model is similar to tap-to-pay debit cards. They're only used for consumer-to-business payments. When you tap the card, the card sends over an account number / signature / etc, which the merchant sends to a central clearinghouse to finalize the transactions.

The main difference is that the card itself keeps a running balance of how much money the customer has available to spend. In many cases, this gives the merchant enough confidence to e.g. let you through the train turnstile without actually waiting for the central clearinghouse to confirm the transaction. (I think in practice they usually send all the transactions in batches, daily or weekly or something.)

The readers do some trusted-computing/secure-enclave type stuff but are not especially hard to obtain; I think there have even been cases where companies like Nintendo have built them into consumer products, so that you could e.g. tap your card to your Nintendo 3DS to buy a video game.

I imagine there's a bit more security on the machines that let you load money on the cards, but it's probably not completely impossible to make a fake card. But the low value limit (usually a couple hundred dollars, depending on the card provider), the inability to get cash out of the system (often you can't even buy things like postage stamps), and the fact that you'll get caught relatively quickly (once the central clearinghouse notices the transactions don't match up) make it unattractive to do it in practice.

Thanks for the detailed answer.

For me, 2 kids and 2 jobs has done wonders for curbing my social media habits.

It won't get you days off because you (rather, your employer) will fall behind those who didn't opt to take days off.

You'd only get days off if it was only you who got a 10x increase. But it's everybody. So it's status quo: technology advances, and you have to keep up if you want to stay in the industry.

> It won't get you days off because you (rather, your employer) will fall behind those who didn't opt to take days off.

I don’t believe this is true. Or that more hours worked reliably translates to serious competitive advantage, in general.

But your point stands, because many (or most) employers think it does, and employees are usually incentivized to support that notion.

40 hours work week wasn’t always the case and yet here we are.

Yes I remember having a hard time finding other kids who wanted to actually play the pokemon card game. And even when I could find someone, they didn't care about the rules/energy costs. This was in elementary school though to be fair.

I would love to see the codebase once you reach the zero issues point.

I would advise against it, depending on the project.

My lone lisp project gets the most love. I spend weeks reading, reviewing, restructuring and rewriting everything. It's the project where I'm concentrating all my efforts. Everything I push to master is absolutely my own work and I do want everyone to read it.

I had no trouble letting Claude take over maintenance of my static site generator and virtual machine orchestration scripts though. I wanted to care but... I didn't. I did glance over the finished product just to ensure it wasn't going to nuke my laptop the second it ran, but that's pretty much the extent of it.

> Who claimed that?

Dario Amodei

If an infected computer gets disabled after deactivating one stolen credential, it might slow down the victim from deactivating their other stolen credentials.

Ugh. True.

I switched to Codex several weeks ago since the massive degradation of Claude Code's quality they recently apologized for. Since the apology and fix, I've considered switching back, but seeing this and other recent things, maybe I'm fine where I'm at.

Is it complex? I was somewhat taken aback by how simple it was. Still very confused as to how it could happen.

Only the weights and the RNG used to select tokens can answer that. You will understand much if you read up on the quality of code in the CC source leak, it's completely vibe coded and the printf fn is genuinely impossible for a human to comprehend.