It also seems like there's gravity coming from off screen assets (or maybe it's the bottom of the screen?) causing the projectile to curve in unexpected ways, and not be captured as strongly by the gravity of the visible objects as I'd expect.

So I run the agent in a VM (it's faster, which I find concerning), and run an MCP on the host that the guest can access, with the MCP also only containing commands that I'm okay with the agent deciding to run.

Despite my previous efforts with skills, I've found agents will still do things like call help on CLIs and find commands that it must never call. By the delights of the way the probabilities are influenced by prompts, explicitly telling it not to run specific commands increases the risk that it will (because any words in the context memory are more likely to be returned).

There's been a few ups and downs along the way but I've found it generally "just works" and gets out the way, which is exactly what I want in an email client.

I've tried almost every single email client I could find on Linux, and several on Windows (including Pegasus mail, if anyone remembers that), but always come back to Thunderbird.

I've been a regular donator to the project ever since they spun it out to MZLA Technologies Corporation.

Easiest alternative I guess is to pipe through head. It still grumbles, but it does work

    openssl enc -aes-256-ctr -pbkdf2 -pass pass:"$(date '+%s')" < /dev/zero | head -c 10M > foo

    openssl enc -aes-256-ctr -pbkdf2 -pass pass:"$(date '+%s')" < /dev/zero | dd of=/home/myrandomfile bs=1M count=1024

> Which for many engineers who got into this industry because they loved solving problems, it can be quite a shocking realization.

It's just another problem to solve, based on the same foundational skill set you develop as an engineer: Observation, interpretation, analysis, experimentation, and implementation.

All-hands meetings are boring as hell, but they'll give me all sorts of signal about various managers up the line. I'll also take any opportunity I can get to be "in the room where it happens" when decisions are made (or speak to people who were in the room) while I'm building up a mental picture of what motivates someone.

If they're glory hunters, I'll figure out how to pitch my thing as something they can brag about. If they're people oriented (rare, but it happens), I'll pitch the human impact angle. If they're money pinchers, it's all about that $/month savings figure, put it front and centre in the opening sentence.

Everyone has an angle, a bias of some description. If you watch what projects do and don't get approved, and what language was used in them, you'll be successful too.

In the US, one day you can have a job, the next none and you'd better hope you've got enough money saved up to cover rent/mortgage, food etc.

That was by far the biggest culture shock when I moved to the states, and really acted as the big "Oh shit, it really is a government by the businesses, for the businesses".

The only time investment on my side was the initial set-up, and that barely took half an hour.

[1] https://iocaine.madhouse-project.org/

The ZX Spectrum that followed, with its huge 48K of RAM was night and day. The programs were so much more complicated.

Even echo on linux these days takes 38K of disk space and a baseline of 13K of memory to execute, before whatever is required to hold the message you're repeating.

Like you say, the only way to stop it from doing something is to make it impossible for it to do so. Shove it in a container. Build LLM safe wrappers around the tools you want it to be able to run so that when it runs e.g. `git`, it can only do operations you've already decided are fine.

Electron really loves to claim absurd amounts of memory, e.g. slack has claimed just over 1TB of virtual memory, but is only using just north of 200MB.

They did trial run after trial run, made sure trying to make sure there were no bugs in the demonstration path. They nailed it, presentation went smoothly, live demonstration just worked. Provisioned a bare metal instance, had it running hosting something within minutes of launch. Larry was suitably impressed, but the thing that most impressed him was that he'd been presented with an end-to-end live demonstration. It had never occurred to any of the folks involved to do it any other way, but apparently all too often, all he ever saw was slide shows from product teams, particularly when things were several months away from public launch.

I reflect on that situation from time to time, wondering at which stage you sort of go from expecting to see live demonstrations, to slide shows. I assume it just slowly slips away from you, one at a time until you're stuck in the land of "make believe we have a good product".

edit: I'm pointing out the UK has apparently decided lootboxes are not gambling, because if they did classify it as gambling it'd be covered by existing gambling laws that restrict it to 18+.

Not that I personally hold that opinion, though I can see how I could have phrased my original message better.

It's a stupid decision by the government, they should be 18+ and recognised for being gambling.

I've never really understood S3's determination not to have a v2 API. Yes, the V1 would need to stick around for a long time, but there's ways to encourage a migration, such as having all future value-add on the V2, and maybe eventually doing marginal increases in v1 API costs to cover the dev work involved in maintaining the legacy API. Instead they've just let themselves, and their customers, deal with avoidable pain.

Comments URL: https://news.ycombinator.com/item?id=47327703

Points: 349

# Comments: 146