HNHacker News
TopNewBestAskShowJobs

harrison_clarke

316 karma · joined March 23, 2013

submissionscomments
harrison_clarke··on String of recent killings linked to Bay Area 'Zizians'
not exactly a rationalist thing, but a lot of bay-area people will tell you that exponential growth exists, and it's everywhere

i can't think of any case where exponential growth actually happens, though. exponential decay and logistic curves are common enough, but not exponential growth

harrison_clarke··on Hacker News for Gamedev
i think a big part of it is that their audience isn't just other gamedevs, it includes gamers

those gamers often have strong emotional attachments to games/characters that very few people have for google sheets

studios get crazy backlash for nerfs and other changes, so i can see not wanting to attach your name and face to that. in the other direction, i wouldn't want a horde of gamers using my words as evidence that my employer is dogshit, unless that was the goal of my message

harrison_clarke··on Ozempic is causing trouble on Savile Row
that, and the contrasting of "obesity suffers" and "[...] fat people"

one of those groups is entirely contained within the other. contrasting them like that, and being judgmental about the latter, isn't going to be a popular opinion

the set of people that both agree with the class analysis approach, and are fine with the fat shaming, is very small

harrison_clarke··on Ozempic is causing trouble on Savile Row
it's currently expensive, until it's off patent and a generic comes out

so rich obesity sufferers are getting it before poorer fat people

harrison_clarke··on Jevons paradox
presumably there are limits to that. once every room is the preferred temperature, i'm going to stop running the heater

although maybe i'd get bored and run the heater and the AC at the same time

harrison_clarke··on A layoff fundamentally changed how I perceive work
turns out, i actually meant benefit corporation. (i've heard people refer to them as b-corporations, but didn't realize there was another thing called b-corporation)
harrison_clarke··on Show HN: Making AR experiences is still painful – had to make my own editor
is there a reason that this is a stand-alone tool, rather than a pile of Unity editor scripts?

i always liked the immediate-mode UI in unity. some people can't stand it, so i can see that being the reason

harrison_clarke··on A layoff fundamentally changed how I perceive work
that's true with publicly traded C-corporations

for private companies, it literally is the people you work with (and whatever legal enchantments they've decided on). some of those people will still fuck you over, but it's not a legally-conjured sentient pile of money the way a C-corp is

B-corps are an interesting attempt to avoid being a sentient pile of money. in theory, it's an egregore that is capable of valuing things other than money. (they haven't really been tested in court. and they might fuck you over in pursuit of some other value, even if they do work. or fucking you over for money might not conflict with its other values)

harrison_clarke··on Time to Take a Stand – Sam Altman (2017)
that line (generalized) has been the most useful piece of advice i've ever read

got me out of a gaslighting relationship, and allowed me to stay at a company during a rough patch (knowing i had a line i'd bail at)

harrison_clarke··on Is the world becoming uninsurable?
the world is always insurable

the more volatile it is (and the less you've mitigated the risks), the more expensive your insurance gets

harrison_clarke··on Life on a Closed Timelike Curve
futurama solved that: fry travelled back and became his grandfather's younger self, rather than his own younger self
harrison_clarke··on Can you complete the Oregon Trail if you wait at a river for 14272 years?
have you seen arthur whitney's code style?

https://www.jsoftware.com/ioj/iojATW.htm

i tried this style for a minute. there are some benefits, and i'll probably continue going for code density in some ways, but way less extreme

there's a tradeoff between how quickly you can ramp up on a project, and how efficiently you can think/communicate once you're loaded up.

(and, in the case of arthur whitney's style, probably some human diversity of skills/abilities. related: i've thought for a while that if i started getting peripheral blindness, i'd probably shorten my variable names; i've heard some blind people describe reading a book like they're reading through a straw)

harrison_clarke··on A laptop stand made from a single sheet of recycled paper
it also seems like a very small savings. the thing sitting on top of it is full of lithium, cobalt, etc. so why should i care if it's sitting on a bit of plastic/aluminum/wood?

that said, a tutorial to turn the shipping box for your laptop (or a flat of diet coke) into a stand would be good. useful in a pinch

edit: keyboard box might be the best box to print the fold lines on. you need that for a minimally ergo laptop setup anyway

harrison_clarke··on Writing a simple pool allocator in C
you can avoid resizing and moving the pools if you allocate massive pools up front. most OSs let you overcommit memory, and most programs using memory pools only have a handful of them

(on windows, you'd need to handle this with VirtualAlloc. osx and linux let you overcommit with malloc, and handle it with a copy-on-write page fault handler)

harrison_clarke··on Writing a simple pool allocator in C
why would you need to iterate on destruction? you free the whole pool at once, since you allocate it all at once
harrison_clarke··on John Carmack on inlined code (2014)
https://www.researchgate.net/publication/266655520_In_the_bl...

this is for a stylus, but people can detect input latency as low as 1ms (possibly lower)

with VR, they use the term "motion to photon latency", and if it's over ~20ms, people start getting dizzy. at 200ms, nobody is going to be keeping their lunch down

google noticed people making fewer searches if they delayed the result by 100ms

edit: if you want an easy demo, open up vim/nano over ssh, and type something. then try it locally

harrison_clarke··on Patent troll Sable pays up, dedicates all its patents to the public
the issue is that there's no knowledge transfer from the patent holder to the developer, in these cases. there's no causal link from the patent to the development

the (forced) licensing deal comes after the development, and hinders it. and it's not to protect development of a related idea, either

harrison_clarke··on Patent troll Sable pays up, dedicates all its patents to the public
if a company sells the patent to a "troll", but retains a license as part of the deal, i would consider that to be working as intended. it's basically a way to outsource the legal protection

if a company sells it to a broker, and it eventually gets traded or licensed to a company that develops it, i'd also consider that working as intended

if patents keep finding their way to companies that have no intention to either develop it themselves, or license it to others, and keep suing companies that do develop things, i'd consider it a failure of the system

pharma patents get traded to non-developers all the time, but pharma patents mostly do their job of incentivizing innovation. there's still flaws, but the troll problem isn't a big issue in that space

harrison_clarke··on Patent troll Sable pays up, dedicates all its patents to the public
it's what they don't do. a troll doesn't play any role in developing the tech they hold patents for, they just extract rent when someone stumbles into a similar solution

if a company doesn't develop products, but they actively license their patents to those that do, that's still patents working as intended, and not trolling. they're still helping to get the tech developed, rather than stifling it

(i think there are a lot of problems with software patents even when used as intended by real companies. mainly, they last too long)

harrison_clarke··on The Fastest Mutexes
biased locking can be faster than spinlocking, in the uncontended case

unbiasing/re-biasing a lock can be quite heavy, though. JVM did it by pausing threads (or waiting for GC to do so)

harrison_clarke··on How to Take Home "Garbage Bags Full of $20s"
if they don't like paying that much, they can just not do it

there are a bunch of cases where the simple economic argument is more thorny (like food). "pay to meet someone famous" is one of the least thorny cases i can think of

harrison_clarke··on How Olympics officials try to catch “motor doping”
wait until you hear about train doping
harrison_clarke··on CrowdStrike broke Debian and Rocky Linux months ago
this is one of larry wall's 3 virtues of great programmers:

hubris: the quality that makes you write (and maintain) programs that other people won't want to say bad things about.

harrison_clarke··on Big Ball of Mud (1999)
big ball of mud is a great pattern for games (non live-service ones, at least)

once you ship, you don't really need to maintain it the same way as other software. tech debt is free money if you ship before having to pay it back

and players often like complexity and interaction between systems, so there's less benefit to isolating them in code

harrison_clarke··on Lessons I Wish I Had Been Taught (1996) [pdf]
grace hopper would carry around nanosecond-long wires, for demonstrations

https://youtu.be/ZR0ujwlvbkQ?si=aAj2OkbS8cj_MeGo&t=2707

harrison_clarke··on Timeliness without datagrams using QUIC
i agree that "unreliable" isn't a good term for the transport. reliability is a property of the system, not of the transport. and you can make an unreliable system with TCP or a reliable one with UDP

but, "best-effort" implies that it's doing some effort to ensure delivery, when it's really dropping any packet that looks funny or is unlucky enough to hit a full buffer

i like "lossy", but this is definitely one of the two hard problems

harrison_clarke··on B-Trees Require Fewer Comparisons Than Balanced Binary Search Trees
red black trees can be thought of as 2-4 trees, and they're way less headache-inducing when you do

if you inline the red nodes into the black nodes, you get a 2-4 tree node. except that the 3-nodes have chirality. but then, you can just use a 2-4 tree

AVL trees seem not worth learning. they have slightly different performance characteristics than red-black trees. but, if i care about performance beyond big O, i'm probably using a hash table or a cache-aware tree of some kind instead of a CS101 data structure

harrison_clarke··on I learned Vulkan and wrote a small game engine with it
game engine architecture, by jason gregory

there's a section on memory allocation patterns

i believe casey muratori talks about allocation patterns in the handmade hero video series, too

edit: ryan fleury has a talk: https://www.rfleury.com/p/enter-the-arena-talk

harrison_clarke··on I learned Vulkan and wrote a small game engine with it
vulkan doesn't have global state, and the error handling is better

but it's not batteries-included, and that's often to be a deciding factor at small scales

i think if you're going to dabble in engine dev, you pick which one you want depending on which part of the engine you find interesting. if you want to make a game, you pick up unity or godot or something

harrison_clarke··on I learned Vulkan and wrote a small game engine with it
there's a common gamedev practice of allocating a big chunk of memory up front, and then using a bump allocator inside of it

in most games, there are about 3 "lifetimes": - permanent/startup - per-level - per-frame

and they're nested. so, you can use a single stack allocator for all of them. at the end of the frame/level, pop back to where it started

there are more complicated patterns, but this one will get you pretty far. you can use it on the CPU and the GPU

← PreviousPage 3 of 4Next →