ted_dunning

Because you're poor and need to hunt squirrels to get protein into your diet?

Or because you have a history of being in that position and feel like it is important to pass the knowledge on?

The Go community has not been very welcoming for embedded applications, for sure.

That might be changing (slightly) because of the Tamago work on NOOS.

Embedded systems benefit substantially from decreased GC, especially in the context of non-relocating collectors (as with Go) in machines without memory management units. The problem is not the overhead, it is the effect of memory fragmentation.

The OP article here is nice because it allows selective enforcement of allocation free code. If you have code that has some discipline about heap allocation such as having all small allocations then you are OK on the fragmentation front.

Don't forget the Boring Company. And Hyperloop. And Zip2.

Suck comes standard.

The image of a chain reaction of RRs popping up like popcorn when another appears to be about to land on it pretty hilarious, though.

Nitrogen, actually.

In business terms they did the first part and will fail on the second.

Best summary:

| ... cybertruck of sports cars ...

There are open references to usage by

ECMWF ASML Pumas JPMorgan Chase

and many other companies.

I use it routinely, but it isn't a company standard where I work.

This is a definite integral. The specific endpoints might allow substantial simplifications that obscure the anti-derivative in question.

This is similar to the way that $\int_{-\infty}^\infty exp(-x^2) dx = \sqrt \pi$. The answer is a constant that doesn't give you any information about the function erf(x).

Depends on the click-through license that linked to the terms of service that you agreed to as part of buying the machine.

Counter evidence:

https://arxiv.org/abs/2603.00042

It is true that block-headed quantization of everything doesn't work. But, as I am sure will read, if you remove the spiky parts of the parameter values, the residue can be dramatically quantized and compressed while retaining performance.

This is a multi-modal sort of compression where you use different techniques for different phenomena. Simply compression all of the weights, each in isolation, ignores the benefits of compressing them collectively.

A non-carbon containing fuel that is almost entirely created by burning fossil fuels is effectively the same (and often worse) than just using the fossil fuels directly.

Biodiesel or biomethane looks at first glance to escape this, but it has serious other costs.

After 1974 is when they decided to not award it posthumously. In 1962 that rule was still in the future.

That's nice and probably true. On average.

It is much less true when Russians invade your country planning to assassinate your chief executive.

Disorganized rabble aren't going to be able to stop that.

These new drones fly at 6000+m (20,000 ft). Machine gun turrets aren't going to touch that.

Generating the lean proof first is a viable approach as well followed by an explanatory pass.

Actually, they are questioning whether the natural language description of the proof is either not faithful to the formal proof, or simply wrong, or both.

Because it is really hard to read and the level of detail is so high that even lemmas that you can read may have such enormous levels of detail that makes real understanding difficult given that humans have limited working memory.

Natural language is ambiguous, but the Lean formalization is very well defined and unambiguous.

It's not the form language that is the real problem here. It's the ambiguity on the other side and the extreme difficulty of doing a useful and accurate translation.

This is the common experience in replicating a published paper by hand ... it is common to find "obvious" aspects that are anything but.

The scary thing is when AIs generate unreadable formal proofs and then effectively lie (or fabulate, to be polite-ish) about the natural language version of the steps. Since the natural language version is arguably the most important aspect of a solution to a flagship problem, this fabulation deflates the value of the solution while the existence of the solution discourages further work on the problem.

It actually doesn't do that.

Did you even read what was written?

They put him on the Beagle to have dinner with the captain so that the captain wouldn't go mad. The captain himself was the one who went looking for a dinner companion.

Also, Darwin had to pay his own way (with his father's money) as a civilian passenger and received no salary for the trip.

So "the British" didn't put Darwin on the Beagle.

If you think that the US is the same US as ten years ago when that was an insane delusion, then you are the one smoking too much weed.

This administration has gone to very great lengths to tear down the safety walls.

The boats in the Caribbean demonstrate this well. Service members who object to targeting civilians without any due process are forced out.

The difference between that is only a matter of scale. And even that scale factor has been breached in terms of the number casualties from the termination of USAID.

It may be delusional right now to imagine service members following a nuclear attack order on Iran, but it isn't clear it will stay that way.

But the rate for milligram sized particles hitting a 100m x 100m patch is 2-3 per month.

At 20km/s, that's plenty big enough to ruin your day.

In normal English usage, "with extreme prejudice" is commonly understood to mean that you advocate killing the person in question.

Either you don't know what you are saying and should improve your language skills

Or you do and should be dropped from this site for advocating violence.

It would be nice to be able to back up in the sequence. There were a few times where I realized I had not fully understood a previous step and would have like to be able to review some previous steps.

Otherwise this is completely excellent!

The average one way commute distance in the US is only 14.4 miles and the median is around 9 [1]. The one way average commute time is about 27 minutes.

As such, I don't think that "most" commutes in US cities are 30+ miles and 45 minutes or more.

[1] https://truedrivingcost.com/articles/average-commute-distanc... Data sourced from US census data.

Charge splitting being equal depends on capacitance which can be hard to make equal.

Fuzz tests with tailored random distributions and good logical property tests are excellent tests.

What would you do better?

Also, the author didn't say that there were no other tests. They just said that they use fuzz + property tests a fair bit.