HNHacker News
TopNewBestAskShowJobs

chefandy

4,947 karma · joined August 3, 2015

[ my public key: https://keybase.io/chefandy; my proof: https://keybase.io/chefandy/sigs/RqBSGYt7dxcO6Sx9c9KFAUARk4mojkuZ5V2l2mZ4BZY ]
submissionscomments
chefandy··on Turning a Bluetooth device into an Apple AirTag without root privileges
My gut says a lot of vulnerabilities disclosed like this had probably already been found by those people that won’t disclose them publicly. Programs that expose these vulnerabilities, funded by countries that can afford it, are probably the biggest stopgaps keeping it mostly in the realm of sketchy nation-state intelligence agency shenanigans rather than organized crime with ransomware-gang-level technical savvy and terrorist organizations. The frustrating thing is that there’s no way you can guarantee that some future problem would have been prevented by a program like this — it’s just good societal-level stewardship of our communications infrastructure. People that consider it reasonable for regular people to defend themselves against things like this without societal support and guidance are delusional. Beyond that, if we think we’re going to maintain dominance in the digital space, removing our collective investment in figuring out what that entails based on a dubious assumption that private industry will pick up the slack unprompted is as penny wise and pound foolish as you get.
chefandy··on Rediscovering Quaternions
Houdini is all quaternions under the hood— purportedly to avoid gimbal lock. Houdini’s thing generally is doing things the hard way if there’s any possibility it could lead to a better outcome. Unfortunately, without a (fortunately open source at the free-level) plugin, doing things like rotating a bunch of objects on their own local axes means wrangling the quaternions directly in code.

If you’re interested in seeing their approach, here’s the repo:

https://github.com/toadstorm/MOPS

chefandy··on Show HN: I got laid off from Meta and created a minor hit on Steam
Stock assets: the first “AI”. Honestly still a better option than AI for most anything in game dev with a few workflow-oriented exceptions.
chefandy··on Show HN: Breakout with a roguelite/vampire survivor twist
Nice work. Here’s some crit from someone with a game design degree if you’re interested in refining the visual design. If you’re happy with it as is as a your own little project, feel free to summarily disregard this comment!

Through the comments of the folks here, you’ve surely noticed that some people aren’t really ‘getting’ some of the stuff right off the bat. One of the toughest things about games is the visual communication aspect of it. Notice how little explanatory text most games have now — even the more low-bit style ones? Choosing coins as a mechanism is a deliberate communication strategy— it’s a game mechanic near everyone is familiar with so they don’t need explanation for how it works. A game’s (or any interface’s) visual components must be approached with the same mindset. A designer’s job is to take a step back and ask “what makes a [coin for example] look like a coin? What are the visual signals— from the most obvious, such as a metallic texture and being a round disk, that make people think it’s a coin? How do they tell it’s a coin and not a blank punch-out from a steel electrical junction box? If I take a bunch of pictures of coins into a photo editor and reduce the resolution to something extremely pixelated, what’s the lowest res I can go to while still realizing it’s a coin, and what about those coins are the few remaining pixels conveying?”

And deeper beyond that is figuring out how you can convey things like slowing the ball down without having a little icon. The cognitive load required to parse which icon you’re looking at and the implications of that is a bit much while trying to play a fast-paced reflex-oriented game. It doesn’t seem as hard as it is when you’re the one that chose the icons and configured their behavior, but especially in a deliberately casual game, people probably won’t take the time to become familiar enough with that system to push through the initial cognitive resistance. Identifying objects is a bit different than labelling a button on a control panel because it has context, physical form, and animation. A turtle might be a great way to label a slow-down button out-of-context, but maybe instead that brick could look gloopy and sticky? Conversely, maybe a speed up brick might be arcing or the shape of a lightning bolt, or vibrating like a revving engine block? Maybe you could have a little starburst where the ball teleported from and to when it teleports to visually orient the user better and on the trigger blocks include some of the graphical elements from that to identify it?

Like I said, if you’re happy with this as your personal project I’m not trying to say there’s anything wrong with it! But if you’re looking to refine the playability for broader audience, the visual design would be a great place to start.

chefandy··on It’s still worth blogging in the age of AI
Do you think you might be being a touch reductive there? Perhaps ignoring factors like context, medium, the economic realities of who makes this “data” and who ends up making money off of it at their expense, the difference between mechanically harvesting data because it’s there and internalizing things through the lens of your mind and existence and making it your own… you know… pretty much everything about it except the razor thin slice of the situation you’re choosing to take into consideration? Any complex and nuanced situation can become cut-and-dried if you ignore enough.
chefandy··on Tell HN: Generational Gap
I think the only really happy path with XML is starting with an extremely rigid spec for extremely consistent data where you have official dtd releases if something changes, etc. I think if you’re trying to figure something structural out when you’re processing the data, you’ve already lost.
chefandy··on Tell HN: Generational Gap
That very much depends on the job. The job I had to process all that handwritten XML in was a library job.
chefandy··on Embedding Python in Elixir, it's fine
Ah, projects. Ain’t it always the way.
chefandy··on Embedding Python in Elixir, it's fine
Been on the list for years — I’ll check it out.
chefandy··on Embedding Python in Elixir, it's fine
I’ve been mostly in Python, C# and C++ for the past decade or so but got into Elixir as my first functional language. Never got comfy with the syntax but dig how everything flows. Looking forward to digging into Gleam.
chefandy··on It’s still worth blogging in the age of AI
Sounds like that contemptible SJW needs to stop being so judgmental. And it’s definitely rude to imply motives that someone doesn’t directly state, isn’t it?
chefandy··on Embedding Python in Elixir, it's fine
I hadn’t heard of gleam. Looks cool! I like working with elixir in a lot of ways but never was a Ruby guy, and I think I’d prefer the C-style syntax.
chefandy··on DeepSeek open source DeepEP – library for MoE training and Inference
OpenAyyyyI swear babe I’m gonna open it up any day. Yeah for that grated good or whatever it is you keep yappin about.
chefandy··on AI-designed chips are so weird that 'humans cannot understand them'
It wouldn’t be cheap but I could see 3D modeling and physically based rendering (I’ve been working with Octane but others should do the job fine) being a really good use case for this. Having a bazillion years in 2D but getting into 3D at a professional level a few years ago, I don’t think I’d even suggest using a purely 2D approach if I was looking for optimal results. Match all the camera specs, simulate all sorts of lighting and weather patterns from all sorts of angles, etc.
chefandy··on It’s still worth blogging in the age of AI
No, that is not “my problem” with it. It is one facet of it, there are lots of sucky ways to take ideas that have nothing to do with scale, and there are lots of ways that fundamentally mechanically harvesting “ideas” as data is different than just learning quickly, but I’m not interested in re-arguing any of this for the 500th time on HN.
chefandy··on It’s still worth blogging in the age of AI
Straw man. I can’t imagine how culture could exist without some people being thrown in prison. That doesn’t mean that criticizing a law or law enforcement practice is equivalent to arguing for abolishing prisons.
chefandy··on AI-designed chips are so weird that 'humans cannot understand them'
What claim would that be? It’s a hilarious, factual example of unintended consequences in model training. Of course they fixed the freaking bug in about two seconds.
chefandy··on It’s still worth blogging in the age of AI
> People had been stealing ideas long before computers started doing it.

So it’s not ok to be bothered by it because it’s been happening a long time?

chefandy··on It’s still worth blogging in the age of AI
I don’t think there has to be some practical economic business justification to a) feel bummed out by your creative output getting munged up into something that, for all its better uses, will feed the great fire hoses spraying trillions of gallons of bullshit all over our information landscape, or b) reduce your creative output because of it. It’s weird how entitled people feel to other people’s creative work and get mad when people don’t freely create for and share with them, while simultaneously minimizing the value that work and its authors bring to our society. Despite what many say, the way people receive and interact with your work mentally/emotionally is really important, and all your work being sucked up into these models— often to create commercial products that are openly antagonistic to the people that created the work that made it possible— changes that. It’s sad that AI has devalued creative processes even to the creators themselves.
chefandy··on It’s still worth blogging in the age of AI
For anyone put off by the dubiousness of the left/right brain thing, rest assured that Drawing On the Right Side of The Brain is no less useful as an introduction to abstract visual creativity because of it.
chefandy··on AI-designed chips are so weird that 'humans cannot understand them'
My favorite was the ML learning how to optimally make the lowest-impact landing in a flight simulator— it discovered that it could wrap the impact float value if the impact was high enough so instead of figuring out the optimal landing, it started figuring out the optimal path to the highest-impact crashes.
chefandy··on Tell HN: Generational Gap
Right, because JSON makes a great ad-hoc data format. XML is a heavy duty format with very strict heavy duty tooling which can do things that JSON isn’t a great fit for (heavily nested, structurally complex, large, exactingly created, extremely formalized, etc.) Developers of a certain age were traumatized by having to work with XML that was the default choice for applications that JSON would have been a much better fit for: lightweight tasks with ad-hoc formats like simple configuration files, simple request data for async JavaScript, etc. They often had lax specifications with poorly considered nesting rules, improper namespace configuration, etc. Or, even worse, I had to regularly parse large amounts of XML handwritten in editors by nontechnical people following policies written by other nontechnical people that thought they were technical.

Having to parse a big pile of sloppy JSON data can inspire you to take an eeextra long lunch. Having to parse a big pile of sloppy XML using real XML tools inspires you take an extra long lunch browsing job openings that don’t contain the word XML.

chefandy··on Who wants to work in a factory?
You’re making stab-in-the-dark arguments against what you assume my worldview is and I’m making it hard because I don’t fit into the preconfigured worldview bins you’re used to. Well, I’m not interested in turning this into some meta discussion that you feel more comfortable with and I’m certainly not interested in playing 20 questions until you can figure out things you disagree with me on without having to come up with a single specific counterpoint to any of the specific things I said. So I’m going to leave you to wrap this conversation up on your own.
chefandy··on Who wants to work in a factory?
A) My wife and I both own small businesses. B) Being critical isn’t rage and dismissing criticism as blind rage is just an excuse to not have to grapple with difficult truths. C) Acknowledging the incentives to dehumanize workers is not leftism. It does not argue for workers to control the means of production or uncompromising democracy in the workplace. D) My political opinions don’t involve the government doing much of anything to anyone.

Any other opinions of mine you want to invent? Go nuts.

chefandy··on Who wants to work in a factory?
A lot of what I said was factual, and a lot of what I said was anecdotal, and none of what I said in either comment addressed employee owned businesses— what you think about what I wrote was obviously based on something other than I wrote, so great job writing a response that was even less useful and insightful than your first. Claiming victory exclusively over a position you assume someone holds while ignoring literally everything they actually said is just a sad, sad straw man argument.
chefandy··on Who wants to work in a factory?
What most people viscerally understand as the worst is the worst they’ve experienced, and even if you intellectually understand that worse things exist, it’s easy to assume you relate to other people’s ‘worst’ far more than you do. My username isn’t based on a hobby— I am a classically-trained chef experienced in and intimately familiar with food service operations and labor from the retail concessions level to Michelin Star fine dining. Working fast food jobs can be tough, greasy, hot, and sometimes dangerous work, but it’s physically, mentally, and emotionally the easiest work in food preparation— almost no other job has been workshopped, designed, redesigned, re-outfitted, optimized and re-optimized for human interaction — and it doesn’t even pay the worst. Especially as a casual teen employee, it doesn’t hold a candle to working as an a la carte line cook, let alone in heavy manufacturing, or nearly any other blue collar work as an adult with adult responsibilities and needs. Even in the food industry — The cog-in-the-machine aspect of being a fast food worker doesn’t hold a candle to being a fine dining dishwasher, which pays less and is far more physically demanding, dangerous, monotonous, micromanaged, etc etc etc. Fast food work is more cog-in-the-machine-like than being a fine dining prep cook, but being a fine dining prep cook is much more physically demanding, stressful and skill-intensive.

Lots of soul-crushing cog-in-the-machine work pays well because that’s the only way they can get anyone to work for them at all. Like working on the UPS truck loading line at the distribution point near where I grew up: the supervisors would walk around on catwalks above the docks with whistles and stop watches putting you on blast if they timed you loading fewer than X packages in a minute. Those jobs paid far better than other entry-level jobs in the area and made the chance of getting into the jobs people actually wanted— the delivery driver jobs— much easier. Those jobs were famously miserable, grueling, and dehumanizing and the people in charge couldn’t care less. They just kept inching up the pay until they could be filled because that’s the way business works. They didn’t think they had a problem with persistently understaffing the docs so it had to be run like a labor camp— they had recruitment and retention problems, which are a lot cheaper. And even delivery driver jobs these days are crap. People like trash truck drivers and corporate-employed Amazon delivery drivers interviewed often say they’re penalized for such minuscule lapses in productivity that they’re forced to piss in bottles because there’s no time for them to stop, even along their route. They’re not shooting for bonuses either— they’re just trying to not get fired. Does that sound like they’re happy? Do you think the bosses that created and enforce these requirements care if they’re miserable?

I know one guy that got life-altering injuries as a tech school welding student. He was repairing a metal ladder above large underground storage tanks that the plant manager said had been emptied, triple-rinsed and dried. Well, he lied. They were full of miscellaneous waste chemicals. The sparks from the tig welder somehow ignited the gases in the tanks and he was blown through the roof of the plant and briefly woke up in the parking lot before waking up months later in the hospital wearing the silicone compression gloves save neck wrap he’d have to wear through his twenties to keep his skin grafts from falling off. In court, the plant manager said emptying and cleaning the tanks whenever people were going to be doing welding was just too expensive. He wasn’t fired. The medical expenses and facility damage were handled by insurance. Do you think he cared how miserable that boy’s life became? How about the private detectives their insurance company paid follow my friend around just in case they could snap a picture of us bowling or something they could use to prove non-compliance and deny his claims? Plant is still going strong today! They have had a few more explosions since, regrettably.

Do think those types of stories are rare? Do you think you understand what working in a place like that is like based on your fast food experience?

Of course business executives are making logical decisions based on the economics— that’s the entire fucking point. They’re doing that because that’s the way the system works. They didn’t create the system, it’s not their fault that the system works like that, but that doesn’t change that it is the way it is, or the effect it has on workers.

> You say “few executives care if their workers are miserable” and that is the dumbest, most misinformed, wrong take on earth. Even if the answer is only one of pure self-interest in profit, having all of your workers hate their lives and the company itself is beyond stupid, and is not the plan.

Got me! Always good to throw together a quick insult to make yourself feel better about not being able to put together a counterpoint, right? Saying I’m accusing executives of sitting in the c suite twisting their mustaches plotting ways to intentionally make their employees miserable is the “dumbest, most misinformed wrong” interpretation of what I said. Would you feel better had I said “few executives think twice about favoring all but the most trivial bits of profit over workers not being miserable?” Do you think manufacturing without union protections or worker safety laws would be better than in Victorian era mills or early 20th century mines? Why? Because executives care about their employees happiness beyond what it takes to keep them from quitting? If so, why isn’t that the case in China? If worker protections didn’t exist and the job market allowed it, do you really think that basic worker comfort and safety wouldn’t be traded for shareholder profits in a heartbeat? If you think that, you’re going to need some solid empirical evidence because that’s not the way humanity has operated historically.

The only places in our country that support anything but high-skill speciality manufacturing are the places where people don’t have a choice. Look up a dozen run-of-the-mill production plants on Glassdoor and skip the reviews obviously made by HR— and then tell me that the people running those places are prioritizing worker happiness. The difficulty of profitably running a manufacturing business in the United States doesn’t change the fact that workers often bear the brunt of the efficiency measures implemented to make it possible.

The author of this nice well-meaning article made some totally valid observations about reasons people don’t actively choose manufacturing jobs in this country when they have a choice — what’s not realistic is thinking we can make it all better just by changing the way we look at it.

chefandy··on Who wants to work in a factory?
> Finally, we must ask a new question: not “Why doesn’t anyone want to work in a factory?” but “How do we make building things exciting, inspiring, and financially viable for millions of Americans?”

This would require viewing workers as something more than fungible labor units whose success or failure is almost entirely a matter of their own performance. Few executives care if their workers are miserable— maybe they’ve got ‘retention’ problems, or ‘recruiting’ problems, both of which usually result in some HR-based initiative.

I don’t see a executives taking that query seriously any time soon.

chefandy··on Paxlovid: You'd have expected more
I’ve had Covid 5 times— I’m pretty familiar with how it progresses. The most recent time was the most severe— I went to the ER with a temp of 103 with ibuprofen. Got Paxlovid— the major symptoms nearly switched off like flipping a light switch. Of course there’s a possibility it would have done that anyway, but I’m sure as hell going to take it again if necessary.
chefandy··on Reviewing the cryptography used by Signal
I think they’re mostly referring to bypassing countrywide web filters.
chefandy··on Firing programmers for AI is a mistake
Counterpoint: https://medium.com/@sweaty.phd/yet-another-ai-will-take-your...
← PreviousPage 2 of 34Next →