I as a "developer", I have probably used less than 10% of my time coding the last 15 or so years... I switched jobs to being a contractor in 2011, and after that it has been a downward slope towards less and less coding, and more of other things. Mostly validation and testing in various forms.

Meh, I still only buy the two-dial microwaves. They are both the cheapest and best at the same time!

Seriosly? They leak emissions if you OPEN THE DOOR WHILE ITS RUNNING?

I thought they were actually, like, certified? How can this not have been tested and fixed... shutting down the magnetron can not take long, right? Making it react fast enough doesnt feel like an intractable problem at all!

Having been trained to listen to the hum of the magnetron for several reasons (among them: it affects how popcorn pops and if you are in lunch room setting with a complete mixture of models you have to listen to know which sort of microwave you "lucked" into that day and adapt to its challenges to avoid burning popcorn) it takes a surprising amount of time for even a good one to spin up to full speed as much as a quarter second. As microwaves age or get cheaper some of them take a full wall clock second or two. Some of the cheap models even lie to you and don't start their own timers until after the magnetron hits full speed.

Something that becomes more apparent the more you listen (but also if you actually pay attention to diagrams of how a microwave works): the magnetron is a spinning thing with its inertia. Even if you immediately cut power to it, it still spins on its own for some amount of time. Given how much energy and wall clock time it takes to spin up to full speed, it shouldn't be surprised it needs similar wall clock time, if not energy to full stop.

But also, yeah the door pull sensor is a classic analog latch detector that has a slower sensing time than a button would by its very nature (and trying to avoid false positives from a loose/vibrating door). It's an easy thing to cut corners on and some sensors are worse than others.

(And also, safety certifications include a margin of error that it still "generally regarded as safe", what's a few extra microwaves escaping into your body among friends as long as it isn't full power?)

Er - you know a magnetron doesn’t actually spin, right?

Ackshully that's not strictly true. Some (very) old models did not rotate the food, but instead rotated the microwave emitter in the top of the cavity.

As a first approximation of referring to magnetic fields and their flux and inertia, "spins" is still a useful and common word for that. But yes, not necessarily the best technically correct word.

If I'm understanding the paper correctly, the bursts had a mean duration of 0.14 seconds, which for a 1000 W microwave would expose you to 140 joules, enough to heat about a shot's worth of water by 1°C. Seems plenty fast to me.

The magnetron itself has about about 65% efficiency, but the paper conjectures that the longer duration of the pulses is due to defects in the cavity that result in some emission at a lower frequency (1.4 rather than the normal 2.4 GHz), so the energy radiated must be a tiny fraction of the nominal power.

This assumes all the energy is leaked when you open the door, and that the power is constant rather than ramping down. I'm guessing a -lot less- leaks than this.

(And, of course, you don't absorb all of what leaks).

heywhatsthat?

I think that was it, thank you for finding it again!

Actually, I was thinking of https://caltopo.com/map.html but your site led me to it.

I mean this is coming to the same result as heywhatsthat, apparently using the same dataset. Sadly it is not really correct, in that I think it blends a lot of things, including TREEs into the height. Its very obvious many places that some height is just not true, unless you account for buildings and treetops.

I believe I _might_ have a 33km view FROM MY ROOF, from 2m above ground I have much less than 1 km.

I have to 1000% agree with this. In a large codebase they also miss stuff. Actually, even at 10kloc the problems beging, UNLESS youre code is perfectly designed.

But which codebase is perfect, really?

You can't possibly cram everything into AGENTS, also LLMs still do not perfectly give the same weight to all of its context, ie. it still ignores instructions.

Yeah, I tried various very sane-looking instrucions file when starting to use copilot 6 months ago. Turned out it was not really useful. It mostly follows the rules anyway, but it also often forgot to. So turns out, especially with the fast turnaround with models today, it was better to just forego these instructions files.

Ive been thinking about our company, one of big global conglomerates that went for copilot. Suddenly I was just enrolled.. together with at least 1500 others. I guess the amount of money for our business copilot plans x 1500 is not a huge amount of money, but I am at least pretty convinced that only a small part of users use even 10% of their quota. Even teams located around me, I only know of 1 person that seems to use it actively.

Sometimes when I was stressed, I have used several models to verify each others´ work. They usually find problems, too!

This is very useful for things that take time to verify, we have CI stuff that takes 2-3 hours to run and I hate when those fails because of a syntax error.

Syntax errors should be caught by type checking / compiling/ linting. That should not take 2-3 hours!