MrEldritch

Confusingly, both articles do indeed mention each other for me.

Noticed it kept using the term 'resonator' or 'resonance', decided to navigate to a page for 'resonance cascade' as a joke, and discovered this fantastically broken article: https://halupedia.com/resonance-cascade

Hear, hear! I honestly think the obsession with cryptography and security has caused us to lose much of what is simply fun about technology. We have grown so used to the assumption that everyone involved is a corporate player and that fools must be kept insulated that we have left no room for play.

The concept is interesting, but I'm getting a lot of red flags from this - there's no experimental data or proof-of-concept work at all, which makes this feel more like a blue-sky "Look what we could do if we could arrange atoms however we wanted!" pipe dream in the Drexlerian mode. Something about the writing style's also pinging my LLM radar, which while not disqualifying in-and-of-itself is very discouraging in combination with the other funkiness. The chemistry and manufacturability strike me as questionable in particular, and I'm not convinced the physics of reading and writing are nearly as clean as the author seems to think.

(I'm also unclear how the bit is supposed to actually flip under the applied electric charge without the fluorine and carbon having to pass through each other.)

Write-Only Code is not a prediction about what we should want. It is a description of what happens when software production scales beyond human attention.

Have we considered whether it's even a good idea to produce software at scales beyond human attention? I'm beginning to suspect that, in terms of the net amount of economic effort and sheer quantity of software produced, we are already creating simply too much software relative to the amount of economic effort we put into hardware, construction, and human capital. Most human needs and desires can only be met through manipulation of atoms, and it seems as though we've largely refocused on those which can be met through manipulation of numbers and symbols - not because anyone really wants their life to revolve around them to the exclusion of everything else - but because they're the easiest markets to profitably scale for the least amount of capital input.

I primarily use it as a toy, but it's also come in genuinely handy for me as a scripting language and calculator (but a calculator that also works with strings). The terseness isn't really a code-golf thing; once you're familiar with the glyphs (which are really a fairly well-chosen set of powerful algorithmic primitives) it makes it more straightforward than any conventional language I've worked with to just Implement An Algorithm, with no boilerplate or fluff. Arrays with array broadcasting, combined with Uiua's stack combinators, are just a really flexible and general tool for expressing how data flows through a program, and the glyphs make it possible to (once you're used to them) translate those flows into code very naturally and smoothly.

It's difficult for me to express just how fun Uiua can be to use. It's rather like one of those Zachtronics games - both in that figuring out how to fit your task into the array-programming model can be a bit of a puzzle sometimes, and in that once you've done that it's an extremely quick and non-frustrating process to make it work. The pure essence of what makes programming enjoyable, for better and for worse. There's also just a lot less plumbing and documentation-reading involved; the flexibility and terseness of the glyphs means that a lot of things you might have to call out to a standard library for in other languages you can Just Write because the entire implementation could easily be of length comparable to the name. (There's also a lot less plumbing and documentation-reading involved because there's only, like, four Uiua libraries anyway. Less to plumb together. Like I said, what makes programming enjoyable, for better and for worse.)

(Why Uiua and not, like, APL? I actually find APL enormously more difficult to read, due to the syntax - APL glyphs have two context-dependent readings, depending on whether they're being used as an binary infix or a unary prefix, and figuring out how the parse tree breaks down when squinting at a sea of glyphs is painful. It's like a whole language of garden path sentences. Uiua glyphs have fixed arity and always mean exactly one thing; so there's twice as many glyphs, but parsing by sight-reading is way more straightforward.)

"Directed energy weapon" is essentially the more technical way of saying "death ray". Lasers, microwaves, particle beams, that sort of thing.

It seems very unlikely - the researchers estimate a lifetime for phosphine of just thousands of years in the more temperate parts of the atmosphere, and just thousands of seconds nearer to the surface.

the energy flux available in ambient radio waves passing through something the size of a handheld is microscopic, pretty much.

While on the subject of FORTH for the 6502, I can't resist mentioning Eloraam's now-defunct (but huge in its heyday) Minecraft mod Redpower 2! It included fully-functional computer blocks that could interface with redstone machinery ... running a custom FORTH on an emulated 6502 processor :)

I think what they're doing with those sculptures is that they're taking three sine waves of different frequencies A, B, and C, and that the sculpture is the path that the point (cos(A t), cos(B t), cos(C t)) takes over the combined period of all three frequencies. That's why it's cube-shaped; x, y, and z each vary from -1 to 1 over the respective periods of each component. Like a 3D Lissajous figure; see this (https://gfycat.com/angelicdismalamazonparrot) for a 2D example.

But I also agree that I shouldn't have to be guessing like this, and that it'd be better if they had more explanation.

Does Hacker News have a "Report" button?

Perhaps I'm missing something, but ... this malware's initial infection vector is via email. If your computer is connected to email, it cannot possibly be air-gapped, unless I'm severely misunderstanding what "air-gapped" means.

I assume that it's meant that the malware infects an internet-connected computer, jumps to removable storage, and then hopefully that storage is plugged into the target computer, possibly through multiple intermediate infections? But the fact that viruses can spread via thumbdrives is hardly novel either.

And the kicker, of how you get the files out of the air-gap, is also not mentioned; "ESET says that during its research, it was not able to identify any Ramsay exfiltration module just yet." I'm certainly aware of a number of sexy proof-of-concept side-channel attacks that modulate things like fan noise or graphics card activity or infrasound to try and exfiltrate data in a way that an external agent could pick up, but there's no evidence that this malware uses any of them; perhaps the hope is that another infected flash drive gets plugged in with an exfiltration module, slurps up the data, and then transmits it out when it's plugged back into a network-connected machine.

Obviously, like everyone, I'm a little concerned about "Embrace, Extend, Extinguish" and I don't really believe that Microsoft are suddenly "the good guys".

But I don't, ultimately, think that this will lead to the demise of Linux. Desktop Linux as a serious competitor for a general-use operating system was already not happening and not really showing any signs of growth beyond a tiny percentage of geeks; if it was going to happen then I think this would be a major barrier to it (why bother abandoning Windows or learning to dual-boot for Linux functionality when you can just use WSL?), but since it wasn't going to happen anyway, another reason it's not happening won't really matter.

(and on the other hand, by providing a less-scary way to get familiar with Linux from inside the operating system you're already used to, it might lower the barrier by an equal amount)

And Windows is resource-heavy and full of enough overhead (not to mention license fees) that it's never going to replace Linux in the server, high-performance, or embedded space, and this certainly won't effect that.

I suppose it may end up leading to fewer devs directly running Linux in the workplace, since the corp can issue them a more corp-comfy Windows machine instead and trust that they'll still be able to get work done, and all the odd troubles of getting Linux to work comfortably on a laptop just sort of vanish if you can just use Windows anyway, which could erase a good chunk of the tiny desktop/laptop marketshare that Linux already has. Or it may not. Who knows!

Given that that story, with that specific wording, has been a running joke circulating around the Internet for years, I suspect your co-worker's kid has never said any such thing.

These are very interesting! The Legible Mathematics one also made me think of APL, which is very hard to read - but specifically it's hard to read for the same reason that traditional linear programming math is hard to read, the lack of visual grouping[1]. I wonder how it might be possible to redesign something like APL to take advantage of rich text and more complex formatting?

[1] The weird symbols might seem like an initial barrier to this, but they're only hard to read if you're unfamiliar with APL; they're actually very easy to learn and remember, as there's not that many of them and they all do very basic operations. However, APL loves operator overloading and likes to give operators different functions based on whether they're used in prefix ("monadic") or infix ("dyadic") forms, and there are also higher-order operators that consume the operators immediately adjacent to them and then operate on the expressions after that; all of this makes the nominally right-to-left parsing require a fair bit of mental effort instead of being able to rely on immediate visual recognition.

With all due respect (which is not much), this is complete nonsense.

See also APL, which was designed as a concise algorithmic notation first and only adopted as an actual language afterwards. (Hence the use of custom symbols)

Although unlike this notation, APL is more typewriter-optimized; many symbols are made by double-striking two other more basic symbols, and the typesetting is simple and linear - it clearly doesn't take advantage of the full expressive power of handwriting. (Plus, its array-based programming flow, while extremely powerful, is very different from the way most people write algorithms these days)

I don't think this is fair. I have ADHD. It is not an exaggeration to say that I can't function without drugs - I have trouble doing literally any task that requires switching focus or making long-term plans (including complex, multi-step tasks like "put on pants, go outside, find a place to eat, and eat some food so you don't starve".)

Specifically, I can't function without prescription amphetamines, in dosages substantially greater than Erdos was taking.

The point of doing an operating system UI in a game engine is so that you can use that operating system UI within games written in that engine!

There's several examples - Simulacra, Emily was Away, Hypnospace Outlaw - that use the interface of an (imaginary) computer as their primary framing device.

Or it could be useful in an immersive-sim type game where there happen to be computers you can interact with; you could immediately see the applications in some kind of hacking-themed game.

It's not supposed to be a a real utility, or run real applications! It's intended for use in computer games; there's several such games that use a virtual "OS" as their framing device and core interface.

I'm not sure if you realize, but this isn't intended as a "real OS" for practical use.

It's intended for use inside of computer games; whether for simulation of using computers inside the game-world, or for games where the framing device is a (simulated) desktop interface.

In which case, the "limited real world use" is hardly a flaw!

I wonder if one can arrange such conduits (6-way connected pipes and chambers) to replicate a feed forward neural network. If it can be done with light...

It actually can't be done with light, not without active switching components or very high intensities. (The "optical neural network" papers published so far are fully linear networks, lacking the nonlinear activation functions between layers that are critical for deep neural networks to have any more power than a single matrix multiply).

However, I think nonlinear acoustic properties are actually more accessible than nonlinear optical properties, and so an acoustic neural network may be more plausible.

This is a large part of why (at least in the deep learning field, which I'm more familiar with) scientific publishing takes place more or less exclusively through ArXiv, because the field has been moving fast enough that while peer-reviewed papers are also published, by the time they come out they're long obsolete.

(Largely forgoing peer review as a scientific field is, however, not without problems. To say the least.)

This is true! I was trying to simplify things a bit for a basic explanation, but I fear I oversimplified. I just meant that the generally used math breaks down; if you're aware of the problem, you can correct for it, but very often people don't.

This is true, and a valid point. The way they phrased it does make me feel more than a little suspicious, nonetheless.

(Besides, there's some other oddity there, like that apparently the alignment only matters when the magnetic field is calm)

basically, because once you start trying multiple hypotheses on the same dataset, the math used to determine "is this conclusion real, or am I just fooling myself" begins to break down.

The statistical significance threshold usually used is p<0.05, meaning that something is (generally, this is beginning to change since the replication crisis) considered to be a real discovery if it has less than a 1/20 chance of being a false positive under the chosen model.

As soon as you start trying multiple hypotheses, then that 1/20 chance of being a false positive begins to become meaningless. If you can just keep rolling d20s until one of them comes up with a critical hit, then you can easily generate false positives that still look very robust.

This is exactly the sort of bad science - p-hacking, fishing expeditions, and the garden of forking paths - that led to the replication crisis. (And that makes sense, as this paper is from 2013, and predates the widespread discovery of the crisis)

hm. upon reading the paper, this is sounding very suspicious.

The study was truly blind. Although the observers were acquainted with our previous studies on magnetic alignment in animals and could have consciously or unconsciously biased the results, no one, not even the coordinators of the study, hypothesized that expression of alignment could have been affected by the geomagnetic situation, and particularly by such subtle changes of the magnetic declination. The idea leading to the discovery of the correlation emerged after sampling was closed and the first statistical analyses (with rather negative results, cf.Figure 1) had been performed.

Like, am I reading this wrong, or are they straight-up saying "we couldn't achieve statistical significance on our original hypothesis, so we just went fishing for correlations until one of them came up significant, and it turned out to be magnetic declination"?

You don't really want one, honestly. New Scientist is the Daily Mail of science publications. For anything they might publish, if you can find another source, then that other source is likely to be more accurate and less sensationalized and anything you might have learned in the NS article would be more profitably learned at the other source; and if you can't find another source, it's almost certainly not true.

I think the Web just kind of stopped being full of searchable information.