> N900 and the Palm Treo

Performance-wise, a modern iPhone is more closely comparable to room-sized (building-sized?) supercomputer of that era.

But yes, lack of an USB bios recovery path is a pretty terrible design fault, and I’m surprised that they haven’t improved this by now.

If I understand correctly, the problem was that the board was the exact opposite of that. The post title uses the phrase “bricked”.

I agree that FWs handling was pretty bad, but then again, what exactly is the proposed ideal-but-realistic solution they should have offered here?

Manufacturer lets you replace each part separately: it's a conspiracy to make money

How so? Most FW parts seem quite resellable, did you just throw them away after upgrading or something?

It looks like the pop up is part of the browser UI in Chrome, and displayed as part of the website if using non-Chromium browsers. I block ads and don't use Chrome, so I don't see any version of this often.

Thank you for this beautiful insight. Just to clarify, you consider yourself an intellectually curious, HN-worthy individual?

For the record, I measure acceptable latency in milliseconds. I think it has something to do with my brain.

> which is literally admitting to being reflexively contrarian.

They’re very clearly stating that they’re expressing their opinion because it is underrepresented.

> Did you forget to make a point here?

No.

> that can run their tool calls in a separate process/host

That indeed seems like a simple and elegant setup. I can’t recall a single instance of using a harness when I didn’t want the tool calls to be local, though. Why would I be SSHing into a SBC just to run the agent back on my laptop?

> You can have a single instance running that handles your sessions in parallel.

Sounds even more elegant! Presumably, they communicate via HTTP?

> implicitly asserts that you need that many instances running at once.

I don’t need to, but I can. That’s the entire point, in case you’re still missing it.

> Your arguments against Javascript appear to be based on fallacies, lack of understanding of technology, and strong emotions against JS rather than actual fact.

How beautifully ironic.

- I regularly have 10+ instances of my harness running or idling on my laptop.

- My laptop is often at near-100% CPU and memory usage due to the (irreducible) cost of the development tools and environment.

- I often (50% of the time) downclock my laptop’s CPU to 1-2 GHz to save battery and/or reduce noise and thermals while charging.

- I sometimes run my harness on machines that are low-power, low-RAM single-board computers.

If my harness was written in JS, this would be both annoying in terms of responsiveness and limiting in terms of how much work I can do in parallel. This is why using JS for serious software is a... subpar decision.

> but then it also means you can't use Claude Code because it also runs on top of JS.

Correct. I don’t use Claude Code.

Why would you say that? Is it good form to post that someone is being contrarian for the sake of it? OP’s comment looks like constructive criticism with plenty of good points.

> It took the same amount of time (3.5 seconds) as cursor-agent and codex on my machine.

So what? Nobody is pretending that these are minimalist. Codex is 1.2M of Rust LOC.

> which necessarily means that you have some powerful and expressive extension language, ideally the one the harness was written in.

Don’t be silly, JS is a joke. I was going to check Pi out, but now that I know it’s based on a shitty stack, I’m staying the fuck away from it. I don’t have the RAM to spare.

> If these are your complaints, then this is one of the strongest endorsements of Pi that I've ever seen. I think I'm bookmarking this comment.

> good form

And never let anyone tell you otherwise.