kibwen
Joined 15 years ago
Karma
55,665
Submissions
9,316
Member since
Dec 20
Recent submissions
Comment · 3 days ago
To be fair, the misery of software developers is often directly attributable to the decisions of hardware engineers.
Comment · 3 days ago
People do things based on motivation. People motivated by pride and passion know that their work reflects upon them, so they have an incentive to care about quality. People motivated by money know that their paychecks will keep arriving as long as they keep meeting a minimum bar, so they're not incentivized to care about quality, but they are incentivized to care about responsiveness and consistency. For critical software, adding in legal liability for one's work creates the quality incentive that's lacking in most professional roles, while retaining the accountability that we can't expect of volunteers.
Comment · 3 days ago
Those percentage of SWE working on critical infrastructure should be licensed and legally liable, yes. You shouldn't need a license to build a CRUD app, just like nobody needs a license to build a treehouse in their backyard. You should need a license to work on a car's braking system (https://safetyresearch.net/toyota-unintended-acceleration-an...) or radiation therapy machines (https://en.wikipedia.org/wiki/Therac-25), just like you need a license to engineer a highway bridge.
Comment · 3 days ago
Arenas aren't a free lunch either. You need a specific sort of memory access pattern, where large, conceptually related chunks of memory are being rapidly and predictably reclaimed, and if you don't have such a pattern then arenas are ceremony for no benefit. And the performance tradeoff is that you end up using more total memory than RAII because you're no longer eagerly reclaiming memory the moment it stops being used, but purposefully delaying your frees so that you can batch-reclaim it all at once. There are many cases where this makes sense, which, to reiterate, is why crates like bumpalo exist, which work gloriously well with Rust's lifetime system.
Comment · 3 days ago
Dropping vecs that contain strings does not make any Rust code in the real world slow. You would need to have a vec containing billions of strings, occupying several gigs of memory, before dropping such a structure took any perceptible amount of time. And for extreme cases like that, you can simply use arenas. The bumpalo crate is approximately as old as Zig is. The reason that people so rarely turn to arenas in Rust isn't because you can't do them, because you've always been able to, but rather because arenas are a very specific solution to a specific sort of memory access pattern, and going out of your way to strictly adhere to that pattern doesn't produce measurable differences for most software.
Comment · 3 days ago
That will require too much parking capacity. Cheaper to just instruct the cars to lock the doors and drive directly into the ocean.
Comment · 4 days ago
> Which is why most of the PC world is on cheap 4k monitors still.
According to the Steam Hardware Survey for last month, 47.91% of Steam users are on 1920x1080; adding the 2.56% on 1920x1200 means that a majority of users are still on 2K. The next biggest chunk of users are on 2560x{1440, 1600}, at about 32%. Only 5% of users are on 4K resolutions or greater.
And these are gamers, who are more likely than the average user to care about graphical fidelity. Most of the PC world just does not care about 4K, period.
Comment · 4 days ago
Assuming that the biggest bang-for-buck of JIT compilation is optimizing hot loops (which is also the biggest bang-for-buck of optimization in every other context as well), and assuming that, per the historical best practices, all hot loops in Python are already written in C (or Rust, these days), this outcome is somewhat what I would expect.
Comment · 4 days ago
Anyone credible? A nontrivial number of people also believe that humans used to ride dinosaurs around like in the Flintstones.
Comment · 4 days ago
Predictions markets are terrible mechanisms for this, because they require imposing an arbitrary time limit at which point the prediction will pay out. IOW, even if you religiously believed P=NP, but you think it would take at least two years to prove it, you would vote against the prediction if the deadline were one year.
Comment · 5 days ago
> It is impossible to quantify if they are now 13% less evil.
We could, but we would need access to Larry Ellison and a scale.
Comment · 5 days ago
The analogy doesn't check out. When a firearm company sells you a gun, they give up ownership of the gun. But Anthropic isn't giving up ownership of anything. They're a services company.
A better analogy would be a firearm company offering a service where one of their employees takes one of their guns and fires it off at whatever you ask them to. One can imagine that the liability implications of this service would be rather different from the perspective of the company.
Comment · 7 days ago
Microogletazon, here we come!
Comment · 7 days ago
I think it would be fair to allow a hypothetical 1920s CD player to occupy most of a room. I'd be interested to see exactly how hard it would be to produce such a thing using the available technology of the era, though I'm still somewhat skeptical regarding the complexity and precision required of the laser motor (which unlike a vinyl needle needs to be controlled manually, rather than just following a groove mechanically) and the spindle motor (which needs to accelerate precisely depending on the laser's distance from the center, unlike a record which revolves at a constant rate).
Comment · 7 days ago
In addition to the circuitry, I was also assuming that the laser would pose a difficulty.