addaon

> What we have done and what comes next

"You've seen what we've done. The August 21st outage comes next. See you then!"

X86 asm as she is intended to be wrote. `repe cmpsb` for symbol table string comparison, etc.

Wow that site blows right past awful to baffling.

"The AI job-search Operating System" -- I'm on the first phrase and already so confused. "job-search" suggests we're not in title case -- and I don't know what "operating system" would mean in this context anyway. Maybe the reason it's capitalized is because I'm supposed to interpret it in German? But the hyphen in "job-search" suggests this is a noun compound and that "search" isn't a verb here; but then if we're supposed to be thinking in German why wouldn't that be capitalized? Just so we're not confused into thinking we're in title case and pick up the language hint? I'd move on to the rest of this sentence but I'm already exhausted. But at least there's a nice animation grabbing my attention that tells me that even if I continue the best I can hope for is 82. (Out of what? For what? If I could get an 85 or something maybe I'd read more, but if I can't even get an A-...)

"These are the exact commanded sRGB values shipped in every export." Tell me what that sentence means and why each of those words is there. Challenge: In the context of the first sentence of a section, with no preceding target for "these." If you manage to assign a meaning: Why do I care, at that point in the readme, about the information in that meaning?

"A warm filter is not a tinted overlay." That's lovely (I suppose), but only an AI would start a section by telling me what something is not, rather than what it is. It's also not really a calamari salad, but it didn't see fit to mention that. This time.

Strongest AI indicator: Words that may carry meaning, if you don't think so hard, but don't work towards any meaningful narrative (because there is none) -- they carry the LLMs need to generate text, not the reader's need to gain progressive understanding. All of this writing without a mind model of the reader shares the same flaws.

Which is what separates us from the animals.

It's a 30-pax (32-seat), Part 25.

Except the constitution is clear that issuing letters of marque is reserved for congress, not the executive branch.

Why are zero plural.

Unlike the Fortran statement which is pronounced `jo toe`.

> almost all of Javascript's good parts come from Smalltalk

Is this underselling the role of Self in JS’s prototype-based object system, or saying that a prototype-based object system is not one of the good parts?

Sure, but that's just lending you a computer at that point, unless they expect you to cover the depreciation cost or something when returning it.

Seven might disagree.

> They would also give you a PC in agreement for continued employment for N years.

Do you mean that in exchange for accepting a PC, he had to agree to a multiple-year enforced contract? Is the right to leave a company even a right that one can sign away in the US? Would people really do that for a few thousand dollars? That seems shockingly close to indentured servitude... "we'll pay two months of your rent up front, but you can't quit for five years no matter what we do to you."

This is excellent!

I am available to not build wind turbines at a substantial discount.

Very few of the MSP430s actually use FRAM, last I checked -- it's limited in both size and speed. (My understanding is that the FRAM speed is substantially better than Flash speed, but they treat it like SRAM for XIP, with I think two cycle waits up to 25 MHz or something, vs Flash where the expectations for XIP are much lower, so higher clock speeds are supported with more wait states.) The family is split between Flash (the majority) and the FRAM parts.

No, someone else has it. It's a zero-sum game.

Shouldn't a Fathom have a six foot bed, by definition?

It's not gone. It was just transferred in payment to someone who understood the rules of the game better.

NAND hasn't been scaling great lately. It seems like PCM or MRAM would both be better fits.

> I lose all of the fun in learning to be chef.

A great steak deserves the respect of preparing it the best, most consistent way you can. For me, that's a reverse sear.

Do something complicated with potatoes to stretch your skills and show off. Worst comes to worst the potatoes are sad.

Reverse sear over sous vide -- moisture is the enemy of crust.

Cheap knock off a Big Green Egg with hardwood charcoal can get to 650°F or so pretty quick, and does an amazing job of finishing up after a reverse sear.

Order placed.

I respect that you're doing a single variant up front, and I understand the business challenges and risks of going wide on a niche product like this... but as a single datapoint, I'll pay double for an ortholinear with similar features and style.

> "probably only 20% chance we all die"

Bad news for you -- there's a 100% chance we all die. Sorry to be the one to tell you.

I think this is kind of my point -- there's no way to develop that level of nuance in a movie. The options are to take way more time (see Silo), to do a crappy job (see... almost every 2 - 3 hour movie), or to instead change the goal: Don't have the viewer "develop… a deep understanding", but down-select to those viewers who already respect the work and have that deep understanding, and instead focus on showing them the society and its norms. Do what movies are great at -- make a story come to life -- and make the viewer/reader do the work of knowing the background, knowing what they're seeing, knowing what's important.

Define.

If you're asking would I pay $470 - $2400 for an (e-beam?) lithography machine capable of making a fully-custom digital IC for < $20 in consumables and no hazardous waste disposal issues using free software... yes.

But I somewhat doubt your ability to deliver.

The most recent Dune movies pull off the incredible feat of being a better telling of the tale than the book, but only if you already know the tale from the books. They’re not beginner-friendly, but they’re extremely respectful to the book reader, while still make conscious (and in my opinion, positive) changes to the story.

I run about 50 Philips Hue lights and I’ve replaced 3 in six years, so definitely not as reliable as the L-Prize lights, but not awful. The L-Prize bulbs are built like tanks… when you drop one you check the floor for damage before the bulb.

Every Philips L-Prize bulb I bought in 2012 is still working. Some have been on about 30% of the time since purchase. A few have been dropped multiple meters onto concrete; oops. Well made LED bulbs last. Cost-optimized ones may not.

> But when I need to write code in areas I'm less confident about, I'll let AI take the reins and the do a thorough review myself.

I see this mindset a bunch, and I just don’t get it. Reviewing code is so much harder than writing code! To write code, I need to find one path to achieve my goal, and convince myself, the compiler, and the reviewer that it does so. To review code, I need to consider all reasonable paths to achieve the goal, and confirm that a sufficiently-good one was selected. This is why we have junior engineers write code and seniors review it. If you’re beyond your comfort level writing it, how can you possibly review whether the author (AI or otherwise) made good choices? You only have exposure to one possible choice lane, and by problem statement you weren’t comfortable populating the other lanes…