@VoxelVexillologist's banner p

VoxelVexillologist

Multidimensional Radical Centrist

1 follower   follows 0 users  
joined 2022 September 04 18:24:54 UTC

				

User ID: 64

VoxelVexillologist

Multidimensional Radical Centrist

1 follower   follows 0 users   joined 2022 September 04 18:24:54 UTC

					

No bio...


					

User ID: 64

We even have fancy new tools for this like Compiler Explorer, which is great for answering "does clang vectorize this like I want it to?".

Of course, the US has its own excuse (WP):

The US excuse also includes (has since 1996) that they only prohibit anti-personnel landmines "outside the Korean peninsula", which would otherwise AFAIK prohibit them from being a party to the treaty. It looks like the last few administrations have gone back and forth on expanding or contracting that condition, though.

It strikes me more as a B&D C++ alternative than a C alternative,

If you are really into this sort of thing, you should consider Ada/SPARK: Rust is cavalier enough to let the programmer engage in potential integer overflows (in default production mode) and doesn't support specifying custom valid ranges (type my_integer is range -3 to 11).

In actuality, I like what they're aiming for, but I expect most of the benefit I'll personally see will be from upping the safety game of C++ (and C, to a lesser extent) via language extensions, automated tooling, and general best practices. I reflexively write tests, use C++11 pointer types, check pointers for nullptr before dereferencing, and use the .at() bounds-checked methods for container elements unless performance is impacted. That said, I do occasionally cause segfaults, still.

The interesting question is if those jobs need to be so obviously unnecessary.

I think this may be one of those "exponential curves are self-similar" things: if you pulled up an administrator for Hammurabi and described the state of modern farming, I'm sure they'd look at you agog and ask what you do with the idle subsistence farmers. And the story there is that "division of labor" led to a centuries-long Renaissance in terms of pretty much every human endeavor that isn't "scratch out a living on a small plot by hand". For all the claims of "singularity", indefinite exponential growth seems an equally valid outcome.

So I guess I'm on team "we're pretty good at finding new ways to keep ourselves busy", with a look of part dismay at "consoom content slop" trends (as if alcoholism and other vices haven't been with us for ages too).

Part of it is just how you look at the economics: you can exchange money for goods or services, but when you buy goods that money isn't expended in the production and distribution of that good, and it at the end of the day it ends up in the accounts of one or more actual humans. Automation can reduce the number of humans in that chain, but the prospect of eliminating it completely seems pretty far off: "my car, fashioned from steel mined from land I own by my own robot army, fueled by gas my robots extracted and refined".

I agree there are reasonable concerns about the concentration of capital, but the free-market endpoint of "Scrooge McDuck holds all the dollars" is a self-defeating liquidity crisis where nobody can exchange goods or services (even spending from the gold pile ends the condition), and other than inflationary threats, the market is typically ambivalent about a huge hoard of unmoving currency. "Bertrand Russell's teapot, but it's a quadrillion US dollar bills that I own" is at best a way to start a religion, unless the astroid mining folks strike it rich, but even then it'd only directly hit goldbugs unless they have actual US dollars, which are a social construct.

I think you're probably right here about actual attacks here, but I suspect the threats start appearing at much smaller levels of celebrity. I have noticed that lots of small-ish YouTube creators have subtly started upping their opsec game as their channels have grown. I'm not trying to stalk anyone, but if you're watching, a public "I moved houses" followed by no longer showing the exterior is a fairly common arc (one runner I follow concurrently started driving to different places to start even short runs that previously left from home). Maybe some are just as paranoid as I am, but I'd bet it doesn't take much Internet fame on average for the crazies to start DMing you (or your own thread on KF).

Like the double edged sword of the Internet more broadly --- the crushing dichotomy of endless slop and almost the complete collective knowledge of mankind at your fingertips --- the tools are there for you and a couple friends to go produce, say, movies with effects that surpass Kubrick's with a much larger budget. Blender is free (and Academy Award winning!). Camera equipment is smaller and lighter and cheaper. LED lighting can run on batteries.

And yet, nobody that I've found is producing well-written, compelling movies on shoestring budgets that actually get eyes, while Netflix keeps churning out heaps of slop with the odd gem tossed in (KPop Demon Hunters was enjoyable). I'm really not sure what to make of it: maybe there is a stochastic element of movie magic that requires the stars to align for a good product and lots projects to produce a hit, or maybe it takes the collective will to power and collective experience of something like the Hollywood juggernauts to push to finish projects well. Or maybe it's happening, but not in genres I follow: are we in a low-budget horror Renaissance? Or it's a change in dynamic to creators of short-form videos?

And the same is true of other mainstream media. I suppose there are a few breakout hits on Substack or various podcasts, so maybe it's happening and we're just not noticing.

I'm not the biggest fan of Rand as a writer, but Atlas Shrugged and friends are generally considered right-wing and at least consider the idea of active hostility.