This seems like the perfect thing to implement as a browser addon. If you think you need it, install it. Otherwise, the code doesn't touch your install.
Much like the container tabs extension, in terms of requiring an addon for certain functionality. Or, if mozilla doesn't do it themselves, a (horrifying though warning) third-party-extension with an external helper application to connect to the serial device, and a API polyfill injected into every page.
When I read "wasn't entirely correct", and "probability of a crash was about one in a million", I wondered how you'd work that value out.
Predictable statistics screams RNG, to me. Especially if the code author knew it was an imperfect solution, and didn't know of one that wouldn't cause this crash. (I've done stuff like this before.)
Also, for the client to crash the server, it'd need to give it some form of bad/confusing data, but for some reason this doesn't happen with fewer users.
I think that was my logic at the time of writing that comment, though I've thought of another few guesses while typing this out/reading your follow-up.
The 'lower total cost, higher unit cost' problem is one of the reasons I love the way the EU mandates advertising prices.
i.e. anything sold needs to have the price you actually pay on it (including VAT/other charges), and the cost per unit. (€/l or €/kg).
This saves me an absurd amount of planning/thinking as I can compare an equivalent cost number whenever I see a substitutable product somewhere. Also saves you from being fooled by your own assumptions, like when the unit cost of the largest cereal box is higher than a smaller one.
I came here to reply to the parent comment's remark about "irreversible interaction like aerobreaking, but this dwarf planet has negligible atmosphere." by mentioning lithobraking because it's been consuming my thoughts for the past couple of months.
Like, imagine a collapsible rod about a kilometer long sticking off the end of a space probe, lined up so it hits the surface as close to perpendicular as possible, each segment made of appropriate material for its impact speed. (I think once you go past the speed of sound in a material, you can't transfer any more force)
With the far end of the rod, which impacts first and with the most force probably vaporizing/creating a crater on the surface (useful to align the rest of the rod), and later sections crumpling in on themselves predictably, like a highway crash barrier or car hood. With a certain max amount of Acceleration, Jerk, Snap, etc... that the probe can survive.
I would very much like someone to explain why decelerating a spacecraft like this is infeasible/inefficient so I can stop thinking about it.
Failing that, I wish to devote the next few years of my life to jamming a massive spear into the moon.
I've seen this turn out the other way on a road contract near my house. (rural Ireland)
Some Italian construction company underbid, and slightly later went bankrupt after finishing some of the work. They just about missed the contract milestones for payment too, and got nothing out of it.
Then, the contract went to someone else who bid, with expected costs adjusted. Think there was an 8-ish month delay.
> Heating water with electricity is the literally last thing we should do.
Would you consider electric showers an exception? They only heat whatever water you're using. I've lived in houses with no hot water in the taps for months, but could still take a hot shower whenever.