wnoise

I do. I want my hardware to do exactly what it gets told to do, and nothing else. A pity that that's basically impossible these days.

Heat pumps can work both ways.

Allies spy on each other all the time.

* https://www.npr.org/sections/parallels/2013/10/28/241384089/...

* https://www.vice.com/en/article/5d9bp8/us-spies-allies-south...

And good faith and intent is not a sufficient guard.

Intent makes a difference, but it's not magic. It only makes a meaningful difference if it changes what data gets collected and stored. The data gets collected either way. Data breeches happen, and intent isn't stable (or even fully coherent in companies. You might trust them now, but you probably shouldn't trust them a few years down the line. Look at how different google of today acts than the google of 2000.

For the case in point, that data includes "The categories of websites you visit, but not the URL itself, Includes universal plug and play devices and devices that broadcast information to your computer on a local area network: for example, smart TV model and vendor information, and video streaming devices.", so far more than "data about how my product is used in order to make it better". That's some other purpose. We already have evident of bad intent. Bad faith is not at all uncommon with large companies. You do remember the sony rootkit[0], right?

And even with good intent it's easy to overcollect data, because of the fear of missing out on something useful. It really could be useful be useful to see if crashes correlate with other running software, various registry settings, etc, but collecting that absolutely should be considered way beyond the line. (That rootkit was packed with telemetry too, collecting e-mail addresses and listening habits)

Telemetry we _know about_ tends to be documented, however there's an unknown amount of sampling bias. Further, secrecy is not binary, things can be poorly disclosed. I'd argue they often are, with documents that are neither obviously visible nor transparent about what is collected. This is in no way surprising. Disclosure doesn't directly help any bottom line, it just guards against possible reputation and legal damage if it is discovered.

As you say, users mostly don't care. We live in an age of mass surveillance and have raised generations who think it normal.

[0]: https://en.wikipedia.org/wiki/Sony_BMG_copy_protection_rootk...

Just because spying is useful for someone doesn't make it not spying.

True, but they don't recover themselves automatically.

> I'm not sure why this is the default behavior,

Um. You said the answer earlier:

> because the parser treats the literal as a double

BigFloats aren't built in to the syntax of the language (and probably should't be, so you need to escape the parser somehow -- either pass a string to the constructor, or use the @big_str macro to get a non-standard string parsed into a BigFloat.

No, it's really not. An index variable has very little details to it -- there's nothing to communicate.

Auxiliary functions are complex behavior that should be reflected in the name.

You're not doing what the parent definition promises though -- if you were you'd leave out the parent definition and the where, and just write the go definition with the true name.

Go is instead doing something similar to the parent that is easily transformed to the right thing (i.e. accumulated in reverse or something), or more general that does the right thing when called with specific arguments. Communicating how and why the function does what it does and works in conjunction with the top level wrapper actually matters.

Yes. Where is often lovely -- I want to delegate details, and not think about them yet, but keep that delegation scoped to the function that needs the relevant details.

But calling auxiliary functions "go" is almost always bad naming.

It's true nothing forces or forbids this, but some languages/toolings/communities/ecosystems encourage that more than others though.

Is it put into RAM? That really is the creation of a record.

That depends a good deal on your connotations for those words. Either work, so long as you restrict to the "live" allocations. If not, neither work.

That crashes too, because it still loads the at-fault javascript. But the actual site is now fixed.

Firefox desktop too.

Hence the joke that a vanilla soy latte is really a three-bean soup.

> exactly defining the specifics is not my goal

Exactly defining specifics is kind of important in having workable laws.

So, banning marketing lies, and fraud? No need to touch 230.

E-mail absolutely is user-generated content. Section 230 doesn't say anything about "posted publically". Now, there's not quite as much need for it to protect non-public content, because getting people to sue over that is a bit harder, but it really is rightfully protected too.

Extraordinary claims do. Ordinary claims only require ordinary evidence.

The signal theoretic framework depends on bandwidth limitations that often don't hold for signals we want to produce. Sharp lines and boundaries are reasonable things to want.

I'd say the sampling kernels aren't fully matters of taste, but matters of the display or sampling technology instead. Pixels on CRTs are indeed not little squares, but pixels on anything else pretty much are -- though they can be non-contiguous squares, with different patterns for different colors, even, which eliminates any simplicity that the little-square picture intuitively captures.

Fortunately, all this starts to matter less and less as pixel size and spacing gets smaller and smaller.

Tracked advertising is not all advertising.

Note that even for a spin-1/2 field, nothing weird-looking happens at a macro level. This is because the wave-function enters twice into the expectation values of observables, leaving a spin-0 incoherent portion, and spin-1 vector portion.

Symmetry of boundary conditions cashes out in symmetry of solutions in two ways. The obvious one is solutions that are themselves symmetrical. The less-obvious one is families of solutions, where the symmetry maps one solution to another.

All of the "s" states are spherically symmetric. The "p" (and higher) states aren't spherically symmetric, but are instead a basis for an entire family of states that are related to each other through rotations.

Have you tried to parse it? Even the Swiss army knife of document formats, pandoc, only supports it as an output format.

> nothing else is truly sentient on this planet besides us

This is definitionally wrong. Sentient merely means "having sensations" or even the weaker having "senses". Perhaps you mean "sapient", with reasoning, self-knowledge, and an "internal life".

This too would be wrong, but not as obviously so. When we actually try to look and measure other animals fairly, we generally find that in measures of cognitive capability, there are animals that are not that different from humans. Many animals pass the mirror test. There are birds with complex vocal languages. Both apes and birds have been recorded spreading adaptions in food preparation within a cohort and down generations (the two astounding cases are dealing with the toxic backs of cane-toads in Australia -- learning to flip them over and only eat through the stomach (corvids) and picking them up and washing them (ibises)).

You also have the opposite problem, where things we think we're good at, even ones some would say define us as a species, are actually pretty poorly done by many of us. GPT-2, which is clearly not sapient, and only capable of writing a terrible, meandering essay, still does better than many high-school students, or even college students. And we now have much better examples, that still clearly aren't sapient.

To elaborate, though the page does mention the distinction:

(Cross-)correlation is convolution with a reversed kernel (second signal) (or vice-versa, of course). (For discrete signals, reversing the kernel is just a swapping the indexes around; which makes absolutely no difference for deep-learning). Convolution is more "natural" because it's abelian, whereas swapping signal and kernel in cross-correlation time-reverses the result.

They're cousins. Joan's father, his uncle, was also a physicist.

"codebase already written in C" includes both "all the as yet unwrapped libraries" and "the OS interface".

This is where you really want to start using typedefs.

And any structure is O(1), not O(n), because C structures are not parameterized.