jeffbee

If you substitute the word "Claims" for the word "Reveals" then the headline is honest.

No it's impossible. But it would be optimal!

Knuth's base-e proposal enters the chat.

They were right about everything 50+ years ago, but they didn't have the budget for the right hardware, had to write conference papers and books instead.

You may be thinking of the https://en.wikipedia.org/wiki/Clathrate_gun_hypothesis

It looks cool and I am building it r.n. but the submission title was a little misleading as it says "container ... runtime" right there, and I guess it is, but it is not a CRI implementation.

That's enough to turn a whole barrel of water detectably minty.

Whenever wires don't make sense, as is the case for aircraft, synthetic liquid fuels can make sense. In a world where the inputs are free, the inefficiency of the process is not relevant.

That's sort of weird though because unless the operating system supports your applications, the performance of the processor cannot be appreciated.

Is raw perf considered Apple's moat? I think it's their software. I would buy a MacOS laptop even if the performance is ghastly ... in fact, I did: the MacBook Neo.

What I want is the PMU in the loop.

Eh, not sure. Very few humans do software optimization productively, so I'm not optimistic about machines built by humans, either. Every encounter I've had with agentic optimization involved trying to apply a bunch of myths to an existing code base, in a way that seemed cool (unrolling loops, eliminating apparent branches, SIMD) but which was all pointless because the only credible optimizations were going to come from doing fewer loads and taking up fewer itlb slots and stuff like that.

I'd be willing to believe in factory overclocks, one magic speed bin that wasn't common enough to market to the public, but was common enough to yield a few here or there. I worked at a place that was able to secure factory overclocks from Intel, 100MHz faster than what everyone else was buying.

In terms of GB of DRAM it looks even worse.

Why would I ask you about this work? The root of my question is why is this type of output shared on github, rather than the inputs?

How are we supposed to take this stuff seriously if the author (sic) isn't even willing to write the readme? Claude exists! If I want some slop I can push the button myself.

The zoning for Mesa AZ is some of the craziest I've ever seen. The obviously unsustainable car-choked sprawl near NTT PH1 is listed in the growth plan as "sustain" while the industrial area is listed as "evolve". The industrial/commercial strip is between two sprawl neighborhoods. In the general plan the car-oriented cul-de-sac hellscape is noted to be "traditional residential" while the data center area is "regional employment". It appears online that the city of Mesa actively encourages data centers in their "technology corridor"

https://selectmesa.com/business-environment/incentives-progr...

It's in the middle of a large, disturbed area with big commercial buildings, roads, and parking lots, only some of which are data centers. This type of disturbance tends to get hot in the sun. But the larger question is how they are attributing the temperature at the end of the cul de sac to the data center when the somewhat more obvious reason for a hot spot there is that nothing shades it. That is, the measured cooler streets are shaded by the houses, but the edges of the neighborhood are shaded by nothing.

Anyway the main problem with this paper is it does not have an explanation for why the response isn't linear. If areal density of heat rejection was able to raise air temperatures this amount, then nobody would be able to go anywhere near a thermal power station without being vaporized.

> I'm sure other industrial uses can cause temperature increases too

If anything about this paper is accurate, which it clearly is not, then the >8000MW rejected to the atmosphere from the nearby Palo Verde power station should be raising the air temperature at least 250C, 1km downwind. Which it obviously does not.

"The attribution of this warming to data center waste heat is supported by the consistent alignment of the temperature signal with the prevailing wind direction across multiple sites, dates, and meteorological conditions. "

Well, no? That could also mean that the heat island effect of the (statutorily mandated) employee parking lot is causal.

There is no police agency in the Bay Area more deserving of dissolution than the Oakland Police. Oakland should take a page from the playbook of Camden, New Jersey, and completely delete the department, creating a new one under different leadership and with none of the former staff. In Camden, cops who wanted to stay on had to apply and take a psychiatric evaluation, which I think is a great idea for Oakland. I doubt even a bare majority of the current cops would survive a competitive evaluation.

Unfortunately Linux makes it difficult to fully get away from swapping because any file-backed page is eligible to be evicted, and even if you have "disabled swap" the kernel will cheerfully page-out your program, even if it needs to be immediately paged in again to continue. The only way to stop it from doing so is to remap all executable mappings into anonymous memory, a thing which is possible but which the loader and linker do not make easy.

If there was a public interest in these books then it was already attached, and it was an outrage for the sellers to exclusively possess these books, and it was an outrage to sell them to any other private party.

I remember when hackers believed in the doctrine of first sale. You can do whatever with the stuff you own.

"We reimplemented ICU" U+1F631 FACE SCREAMING IN FEAR

All benefits are temporary, my online friend.

I assume the first one is "High Performance HTTP Servers"

Yeah the comment is a bit of a cliffhanger. Wish they had given us their preferred translation.

Even in the 1990s, developers of the HID specification knew that the keyboard pictured at the bottom of the article was wrong. They wrote in the power management spec that interface devices should conditionally send wake events, and that a dedicated wake key was not necessary. Microsoft said:

"For example, a keyboard could include a special-purpose power management button (for example, “Power”) that, when pressed during a system sleeping state, generates a wake event. Alternatively, the button(s) on mice and other pointing devices could be used to trigger a wake event. "

I would say this is a fair question, and that slop can be laundered when there are multiple collaborators and at least one of them knows what they are doing. Whereas a monograph written entirely by Claude is always just slop.

It is extremely odd that this model has that crowbar, spending more for worse results at "high".