AlotOfReading

A lot of companies have Bloomberg terminals for e.g. futures hedging.

Order independence/monotonicity is strong EC rather than regular EC.

Taking a look at their LogQL language, I don't see anything that would be particularly hard on the planner. You can't get the fan-out that makes fully relational query planning so difficult with the kind of boolean filters they seem to use. Planning should mostly be a matter of sorting by column cardinality and query optimization so you aren't doing unnecessary operations.

Does tinygrad not count for some reason?

FPChecker and Herbie are complementary. FPChecker runs the dynamic analysis to figure what your code is actually doing under realistic workloads, which gives you the information to make those code improvements via other tools like Herbie.

Part of the issue with saying what you mean in numerical code is that approximately no-one outside a small cadre of experts bothers to do numerical analysis. And unless you're in HPC, very few of those experts are optimizing for the kind of constraints your program has. Gamedevs often care more about reproducibility and performance than characterizing accuracy or stability, for example [0].

This tool is neat because you can take whatever garbage code you're given and dynamically analyze it in-situ as a quick first step that might suffice instead of a much more difficult numerical analysis.

[0] https://jrouwe.github.io/JoltPhysics/#deterministic-simulati...

FPChecker looks very cool. I wish the exponent graph had the option to show base 2 though, so it's slightly easier to map it back to the underlying float representation.

I added a shebang to a readme once (written in literate style) so the poor engineers on the other side wouldn't have to deal with the multi-step monstrosity within.

It doesn't seem like the value of human labor is going away. If you look at luxury goods, they're still "handmade". Telecoms still advertise human representatives. Nursing homes still charge massive amounts for personal service.

What's changing is how much of that surplus value is captured by the workers doing the labor.

I've been pretty disappointed with how horrifically memory-bound the spark is. By all rights it feels like it should blow strix halo and apple hardware out of the water, but it's completely hobbled by the low memory bandwidth.

And when the bug is in hardware instead, good luck getting them to admit it. Instead they go "We can't confirm that there's an issue, but it'll be gone in the next revision and meanwhile don't use that instruction sequence".

The parent's point is that possession of a physical good is a bright line separation. For digital files, there's a huge difference between [Files you own] on a hard disk, and files [on a hard disk you own]. There are files you can put on a hard drive that you don't own and will ultimately kill themselves when specified criteria are met, like DRM'd ebooks.

Worth noting that the term "domestic" in auto has included Mexico and Canada since NAFTA. The legal meaning was recently changed to "USA-only**" by the current administration, but that's for tariff folks. In the common sense of the term, for manufacturing, absolutely.

I still think it's a useful distinction though. There's also a whole lot of people involved that aren't on the production line (like me). The corporate culture is also very different between the American and Japanese conglomerates I've worked for.

I think that these protectionist measures are ultimately going to hurt the competitiveness of the US auto industry. Exports and international sales are a big, but not overwhelming part of the revenue for major manufacturers. It's something like 20% of GM's revenue, and maybe 35% of Ford's. When they can't compete internationally, it's small enough that executives will convince themselves to focus on their core market in the US. That will in turn lead to production volumes (and hence economies of scale) slowly dropping. Vehicles from American brands are going to become even more unaffordable than they already are,

At least we can hope that newer manufacturers like Slate and Rivian will keep things reasonable, and major foreign brands like Kia and Toyota might bring the fruits of their knowledge from competing internationally to their US models/factories. I'm not hopeful for the long term future of big three though, especially Stellantis and GM.

I don't program zig, so it's not clear to me if you can use zig's bitsets arithmetically.

Sometimes it's just more clear to work with integers than other representations. Most situations with a state space of N bits have meaningful integer representations, where arithmetic functions on those representations are also meaningful.

For example, CRCs can be written as the remainder from long division of the message by the polynomial. Defining nontrivial cyclic permutations is also much more straightforward as functions on integers than on bitsets.

Temperature ratings are the allowed ambient temperature. The actual silicon will inevitably operate somewhat higher, because coolers are just moving heat down a temperature gradient.

Nvidia's automotive and aerospace variants get ratings up to 85C, for comparison.

Commercial usage is almost always significantly more expensive than retail. Software is the same way. I don't think it's common for personal use fonts to charge that much though.

The AI price inflation is unreal. Used to be that you could get the grad students doing all the actual work for the price of a pizza party and alcohol.

Or you take the alternative approach of flattening and minimizing your dependency graph. Having so many dependencies you can't reasonably field bug reports in them is a chosen tradeoff, even if it doesn't feel that way.

The EU has a €300B program directly subsidizing farmers [0]. To quote the wiki page:

     The [EAGF] consumes a large part of the general budget of the European Union. 

Apparently the overarching CAP program consumes something like 30+% of the budget, so I'd love to know what you consider "heavily subsidized" if this doesn't merit.

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

If you think Canada of all places doesn't have immigrants driving around, boy do I have news for you.

It's not being pedantic. With different assumptions you can make a system where 2 + 2 = 0 and it turns out to be extremely useful. You can also build a system where 2 + 2 = 22 like the other commenter lampoons and the free monoid that corresponds to is again useful.

If we had a radically different perspective (like Borges' Funes the memorious), you can imagine how adding wholly distinct objects might seem ridiculous and derive some other wacky system of arithmetic instead.

Of course, you could alternatively derive it from set theory, but you might also end up with something fundamentally different than what the grandparent intended like presburger or skolem arithmetic.

How much did your clever approach save over a naive approach + file compression?

It's the norm in the US (and pretty much everywhere else with well-implemented developed financial infrastructure) for banks to apply extra scrutiny and roadblocks to large withdrawals. There's a patio11 article going over some of the reasons [0], but it notably generates paperwork for the bank reporting the withdrawal to the government and enables a lot of fraud to allow immediate access.

[0] https://www.bitsaboutmoney.com/archive/two-americas-one-bank...

Yes, that's exactly the point I've been making. It's undefined today because C has a type system and that type system has this arbitrary rule, not because there are implementation constraints necessitating it (where it could just be implementation defined instead).

The majority of browsers are derived from KDE's KHTML/KJS, webkit included. Apple made very significant contributions (like the webkit project itself), but that's not where the codebase originated.

Unaligned pointers are undefined behavior even when the hardware fully supports unaligned access, because you're violating the type's rules.

To be honest, I've never seen much indication that the C and C++ committees are particularly fond of each other. They sometimes coordinate, but they're mostly content letting each other evolve in different directions. C is the way it is only after a long process of evolution away from the bits and bytes of BCPL into the strictly typed language we got from ASNI.

C actually does have a type system and it's one of the bigger issues with the language. If it didn't, unaligned pointers and signed overflow would be totally fine.

    See humans have the tendency to romanticize the past, and if this is allowed to compound they elevate really quite dismal people to the realm of literal godhood in some cases. If you asked someone a thousand years ago what they though life was like thousands of years in the past and what it will be like thousands of years in the future most would have said the past was better in all regards including health, strength, morals even technology; while the future would be viewed as the continual circling of the drain.

Historical futurism is actually a pretty interesting subject rather than the tediously gritty monoculture you're imagining here. Early roman writers often imagined the empire would continue without end, for example. There was no circling of the drain. Many renaissance thinkers thought that the "dark ages" of medieval Europe were a temporary measure, hence the name "renaissance" or revival. Others were much much more cynical, and believed in an imminent apocalypse (e.g. the forecasted deluge of 1524). By the 17th century, people were writing utopian science fiction [0, 1], though some of these are really only utopian in the sense that Fukuyama's "end of history" is.

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

[1] https://en.wikipedia.org/wiki/The_Blazing_World

USB-IF does in fact do this and has for years. Certified devices are allowed to have a "Certified USB" logo somewhere on them, usually placed next to the regulatory compliance marks.