newcomputerwhodis
No bio...
User ID: 3218
newcomputerwhodis
4mo ago
Quants care about latency, yes, but they're more than happy to throw a bit more hardware at their problems.
I can see I'll have to be more specific about what I take "performance" to mean. Performance is... efficiency. How much time, how many CPU clock cycles, how much memory, how many watts do you use while performing your task? Latency is one slice of it - a poorly written program will have poor performance on multiple dimensions, including latency - but low-latency alone is not the whole picture. A data center would likely not be happy to know you've reduced their latency at the cost of a large increase in power draw - power and cooling are a major factor in their operations! For game consoles, the hardware is fixed. If you take more compute than the console has to give you to get the next frame ready, your performance is poor. On any platform, if you use more memory than is available, everything suffers as you swap out to disk.
If your overriding concern is latency, to the exclusion of other performance concerns, I guess I can soften to say that GC may be workable.
newcomputerwhodis
4mo ago
Please update your prejudices to reject the current state of technology.
Gladly!
But more seriously, low latency isn't the whole picture. If I care about performance, why would I have so much spare CPU time laying around that I can essentially pin an entire core to be the GC manager?
newcomputerwhodis
4mo ago
Preemptively: garbage collection is a collection of garbage and we would do well to rid ourselves of it. I do not consider garbage collected languages a viable option for anything that even vaguely cares about performance, and they are objectively not a viable option for kernel or firmware spaces.
That said... yes, a safer-C would be useful, and it would be nice if Rust could be that, but I don't think it can. C has too much inertia and there are too many places Rust made seemingly-arbitrary-from-the-perspective-of-C-programmers decisions that grind against C-like intuitions for a comfortable swap, and so since the "pain" of C is actually pretty darn low on a per-developer basis (even if the occasional memory safety CVE is a big problem for society) nothing short of an official Software Engineer Licensing system is going to get them to move. Sort of a tragedy of the commons problem. Try again, but be more like C. Maybe then.
newcomputerwhodis
4mo ago
You can build anything with C if you're not a coward.
newcomputerwhodis
10mo ago
10-2 has excellent combat. If only I liked the story enough to push it higher... probably lands somewhere in high-mid. 7-remake was pretty fun at first, but eventually I got really frustrated with the way ATB fills for the active vs passive characters and the need to constantly ping-pong control around. Felt like I could never get a groove going on anybody. And then of course the plot gradually revealed that it's
newcomputerwhodis
10mo ago
I always hear people going on about how you're supposed to be eternally in love with your first, and 6 was not my first, so that can't be it.
...though I was about 14 when I played it, so maybe a point for your formulation there.
newcomputerwhodis
10mo ago
Don't mind me, just jumping off of you since you mentioned FFs and favorites and I had a pre-prepared ranking. Ahem. The FF series:
1 - one of a very rare few games where I will use the phrase "good for its time" seriously. A bit slow, rough around the edges, probably not a good starter for non JRPG folks. Raised to middle of the pack by stinkers at the bottom.
2 - a very deeply flawed game with few redeeming factors. A stinker at the bottom.
3 - a bit fresh, a bit new, a bit flawed. Eh. Middle?
4 - mostly good, a little great, a couple sticking points. High middle.
5 - yeah, good shit! Top of the middle or bottom of the top.
6 - chef kiss emoji.
7 - chef kiss emoji.
8 - some real question mark choices push it down from where it could've been.
9 - chef kiss emoji.
10 - chef kiss emoji.
11 - N/A, go home and play a single player game.
12 - Fuck you, Vaan. Do not pass Rabanastre, do not Collect Fran and Balthier. To the bottom with you.
13 - Just really weak. (Mostly) nothing to HATE, but so little to love that it can't overcome the flaws at all. Way low.
14 - N/A, pls stop putting MMOs in the numbers kthx.
15 - terrible combat, fun road trip. Low you go!
16 - a spark of greatness, wasted on samey filler and bad overall plot pacing and shaping. Get on down to low tier.
Ranked: 11/14 <<< uncrossable apples and oranges gap <<< 2 < 12 < 13 < 15 < 16 < 8 < 1 < 3 < 4 < 5 < 9 < 10 < 7 < 6
Honestly, though, most of what bothers me about 12 comes from a huge pile of factors almost everybody else considers trivial - they're all just big problems for me and when you stack them all up the experience is incredibly grating.
newcomputerwhodis
1yr ago
Yes, it is already happening, and it was even before AI. Entertainment media can be provided bespoke - that's exactly what artists working on commission do. For a whole lot of people and purposes, the quality/price curve is or very soon will be in AI's favor. I have a couple hours of music about wizards drift racing and I am eager for the moment I can poke at an AI for a bit and receive custom made retro game bubblegum tailored to my exact whims.
newcomputerwhodis
1yr ago
Yeah, I'll just endorse this. Reading the OP I kept being struck by a "huh? no, that's not what happens..." feeling - that the description of having difficulty believing the thing you're doing is stupid did not resonate at all with me personally nor many of the peers of my youth as I understood them. It seems to me that while maybe not the standard reaction of the majority, it's still quite common for performative pretense to have no effect or a negative effect.
newcomputerwhodis
1yr ago
An SBOM is a great thing... for projects that have dependencies. My employer has gone in too hard on trying to have as many stampable "we do X!" as they can to be safe and good and please other businesses, even in cases where it makes no sense. My little division works in a very special environment that has effectively zero available third party libraries. This has been true for over a decade - we are our own special corner. And yet! We must now have a SBOM (it's blank) and do a scan for known third party vulnerabilities (always zero) and pass an license compatibility check (no licenses but the one we put on our own stuff) and so on and so on. It's not that onerous, but it's extremely annoying to know that we are forced to waste some small slice of our time and effort keeping green flags for so many checkmarkables that we could not possibly fail. All this, and the cherry on top is that each component we make (none of which bring in any third party code) has to pass all these individually, and then the final product again has to pass them all - the final product made entirely and exclusively of the components we are already (pointlessly) checking.
Yes, compliance team, I understand the importance of validating third party code and the possibility of security issues! We just don't have any!
newcomputerwhodis
1yr ago
You're close to it - it isn't (necessarily) that the listener shares an identity marker with the artist. It's that the artist shares an identity marker with other artists the listener has previously enjoyed and they're taking that as a meaningful signal.
newcomputerwhodis
1yr ago
30% vs 90% (tilde number percent space tilde number percent) is showing non-strikethrough in my preview box, so it clearly can work and the real problem lies elsewhere. Now I'll post and it'll strikethrough and I'll look like an idiot.
newcomputerwhodis
1yr ago
I'm not endorsing it - I'm just using it as an example more people are likely to know by name. It is the type of protection you want to see - disc encryption that only unlocks under the right conditions.
newcomputerwhodis
1yr ago
You are, in principle, correct but that's exactly what dedicated cryptography hardware like a TPM is there to resolve. The BIOS stuffs some values not known ahead of time but measured/detected during the boot process (like a hash of the values in a bunch of different registers at point D during an ABCDEFG register sequence) into the hardware gizmo. Then the OS polls the gizmo for its current value and tries to decrypt its main boot volume using that as the key - wrong value, fail to boot. A compromised BIOS will now get different results from the measurements/hashses and can't reproduce that same state. If it had full control over the TPM, it could, but it doesn't - it does not respect ring 0. To be clear, there is still a way to beat this - you just have to monitor the values sent to the gizmo and then replay them in order, rather than trying to do the measurements yourself, but you can't accomplish that without physical access to the internals of the machine and some kind of sensor/probe to watch whichever bus the traffic goes over. You can also try to crack open the gizmo and read back its state, but that's also access-to-internals level.
newcomputerwhodis
1yr ago
It shows that people who "can be relied upon to turn out" would prefer a third party if they didn't feel like they "had to" vote against whoever they hate, like if their state ever went ranked choice. Without the vote swapping, they hold their nose and their third party signal vanishes. It does nothing for this particular election, but it maybe, just maybe, increases the odds of escaping the two party trap in the future.
newcomputerwhodis
1yr ago
This could be less serious than it seems. If the voting system is designed right, a BIOS password would be insufficient to cast a fraudulent vote - one possible example is in the vein of Windows' Bitlocker. A modified BIOS would cause the OS to reject the boot attempt, so you'd fail to get anywhere. If you booted something else, like a facsimile of the actual voting system but that swaps 25% of the votes for Trump to Harris or something, then you wouldn't have the credentials to submit the votes gathered and your pack of votes would be tossed.
But you could also use that as the attack, if you deploy machines that look like they're doing the right thing but then have bad/missing credentials and their votes are not counted, you could poke holes in areas that lean heavily Trump. This would be detectable after the fact - there'd be machines that mysteriously "glitched out" but you can't trust the machine so those votes can't come back in.
Either way, that's assuming there's decent security design in place. And of course, if they are right that you need access to the machine to put in the password and there's no remote management gunk somebody forgot to disable and they're under 24-7 guard, then the leak isn't actionable in the first place.
So overall effect - if you trust them to be mostly competent, things are in fact fine (unless/until a bunch of machines' worth of votes are tossed).
newcomputerwhodis
1yr ago
Completely aligned. The current medical standards are hogwash, Skrmetti will likely prevail, and yet I am not satisfied by that outcome. Wrong tool for the job, second and third order effects worse than the cure. This is just How Things Are Done These Days, apparently. I want to get off Mr Bones' Wild Ride.
newcomputerwhodis
1yr ago
Thoroughly endorsed, subscribing to your newsletter. Any discipline that begins to sound like graduate philosophy has a major problem.
Friday afternoon culture war thread? No formal education outside STEM? Alright, let's solve philosophy by messily banging out a manifesto in under an hour and just paste it out there like I know what I'm talking about:
Searle's Chinese Room is no more interesting than p-zombies - both are empty questions. If you are definitionally not allowed to observe an empirical difference then the answer to the question is mu, as both answers yield exactly identical predictions about the future and so are the same answer.
Searle is assuming "understanding" means something functionally undetectable - he's smuggling in that there's "something more" to what we do, as all phenomenalists do. Even if we could open the brain and look inside to 99% accuracy, they'd continue to chase their mystery into the gaps. Their position is fundamentally reliant on there being an unknown element in play. If we had 100% certain explanation of exactly how the brain does what it does, there'd be no mysterious phenomenon left without explicitly postulating a non-physical ingredient.
Same story for Mary's room. If Mary has 100% understanding, then it's not possible for her to learn something new on seeing the apple, as she could just simulate the experience ahead of time. 100% means 0% remains, and anything else isn't part of the brain's physical system. The experiment's "insight" presupposes consciousness is not an operation of the brain.
I'll go one further. Every avenue that purports to explore the "hard problem" of consciousness must necessarily smuggle in dualism in just the same way. Either the mind is deterministically/probabilistically generated by the physical processes in play within the brain (or perhaps elsewhere in the body if your theories are exotic enough) or it is not. ANYTHING the mind "experiences" must come from these physical phenomena, unless there is some other thing not contained in the set of physics which is causing them.
To accept any theist view, one has to find some element of the world that cannot be explained by physics, else parsimony demands we not introduce the relevant deity. If one has such an element in mind, it belongs to a separate magisterium and so the dual layers of the universe themselves are quite an expensive answer to whatever question it was you couldn't answer. Further then, any specific description of or proscriptions from other magisteria cruelly desecrate poor parsimony's corpse. I simply can't see how any rigorous thinker can go this way.
A common objection might be that math or logic is not physical, but mathematics and logic can be instantiated in the physical - one can count apples, one can apply inputs to silicon logic gates. Let me clarify a bit. I am not saying that math and logic are physical. I am saying that despite the apparent ontological cost of introducing new categories, that cost is in reality dramatically reduced because as we can see by instantiating them physically they are not separate magisteria but manipulations of this one.
"Free will" is a popular card in the theist deck beyond the necessary, saying that God has granted us this. Agency is a useful fiction, and as we cannot map the causal web anywhere near deep enough to fully apply determinism to the actions of conscious beings, we are (for now?) free to let ignorance be bliss. But how could it be any other way? For matter to "choose" to behave differently than physics requires it to would be going right back to dualism again, once again importing that very same separate magisterium - and this time not only in the creative capacity, but in a 'has observable physical consequences' way.
Philosophy's mostly hokum. Essentially everything comes down to empiricism and consequentialism, but remembering that unknown and unknowable are distinct classes and keep in mind that Chesterton's Fence works everywhere. That is, assume an external reality exists (because without one everything falls apart and you can't get anywhere), find out what you can, be humble about what you can't or haven't yet, and make decisions based on the known consequences and not-known-to-be-impossible possibilities for which those Fences help you choose in the absence of your own data. To those who cry out that virtue ethics or deontology or any other framework are needed, hogwash! Prioritizing a virtue above and beyond its apparent consequences is really just going up a level and looking at second/third/fourth/etc order effects - sure, in this instance a bad thing happens, but because Virtue is preserved later more goodness happens with higher total value. It's all just fancy window dressing over consequentialist reasoning. Categorical imperatives are just nth order effects with very high n. Being the kind of person who does/doesn't do the thing reinforces other practices of doing/not doing the thing and sets the example that people should/shouldn't do the thing and etc. You're free to use these heuristics, because you can't fully map the causal web, but don't pretend they're some fundamental truths.
Justice (and many of its brethren concepts) are n-th order effect based feedback mechanisms that society instantiates to adjust the behavior of its constituents.
"What is good" is a category error and the values that congnitive systems overlay onto the world are simply chosen axioms (which consequentialism helps pursue the satisfaction of).
This is Physical System Realism.
To leverage PSR and eliminate even more persistent questions: the "self" is the shared boundary of several cooperating systems - a mind, a body, a genetic sequence, perhaps a few more - where they all align in roughly the same place: where their direct and immediate physical instantiation and control end. There are quite a few known pathologies of confused identity that map precisely to these boundaries falling out of sync. In some cases, when a person is particularly invested in the fate of a social organism they are a part of and very strongly feel "part of a community" their identity model may well include that (and this may again explain some pathologies).
Art fulfills axioms related to happiness and wellbeing through satisfying aesthetic preference or providing new heuristics (subtextual messages). Ideas are potential memes or infiltrators or viruses of the cognitive system, but upon examination most are benign. The true threat category is those that change axioms, but then we must allow for the possibility that if the axioms are ranked, a meme may "beneficially" change lower axioms in service of optimizing the higher.
Put very plainly, "believe what is true, act on what is helpful" - which just sounds like common sense. You only have to take it seriously.
Since everything non-quantum is fully clockwork without free will, can we clean up quantum mechanics? Superdeterminism sounds pretty cheap. What extra cost does it impose on us, besides needing to assume the expansion of the universe (which we already accept) began at a single point rather than beginning from some non-single-point state?
None. So accept it. Quantum randomness is just what the current state looks like from within our light cone. With a (much) longer cone, we'd see the causality. It's all just frames of reference. From within our light cone quantum results are indistinguishable from the probabilistic models, and so since we can't escape our light cone there's no reason to worry about predetermination. Universally predetermined, locally random.
One last stroke. Surprise is your heuristic for detecting that you need to update your model. If you can see the fixed future, you cannot be surprised. With omniscience's inability to be surprised and the fixed future, the very idea of a deity "touching" the universe becomes impossible. If any deity even could exist, it would be solely one that set the initial condition of the universe and hit go - an entity elsewhere running a simulation that is our universe. Theism is now isomorphic to the simulation hypothesis. Because this generates infinite regress, parsimony demands we remove it. There can be no god.
Philosphy's pretty easy - you just can't give up when something feels cold. Friend, the universe is on average quite cold. Axiomatically choose warmth, then go find it.
More options
Context Copy link