jackling

I'm not in academia, so I might be fully ignorant about how things operate, but if professors don't reaed the actual paper, can do they know if it's BS or not?

They're getting better, and are starting to be used in robotics: https://scenesmith.github.io/

Curious, when you do this, do you understand the math/reasoning of the paper and just have claude as do the coding? Not saying that matters if you just care about the end result, but I'm curious how much using an agent affects your understanding of what the papers are proving.

Never tried with that low of a timestep, wonder if that could start causing floating point issues which will lead to more instability.

Awesome work, what always prevents me from implementing more solvers is the amount of math required. While the implementation always seems simple, understanding the different optimization strategies for each solver gets confusing.

It's really impressive that the author was able to implement rendering papers and physics sim papers with such regularity. It really is a feat. Makes me curious to see what their background is.

Can you elaborate on what you mean? It could be a matter of perspective, For a stack of blocks, each 1 meters high, the stack can get quite high and your expectations on how it should look like IRL might not be correct, due to never experience a large tower of blocks being knocked over at that vantage point. Especially if the mass of the objects are strange (super light for their size or super heavy).

I know in older games, the recommendation was to keep gravity low (~6 m/s^2 iirc) to help with simulation stability and make things look better, that might contribute to your idea of things being floaty.

I don't find the examples in the git repo to be especially floaty, but I work with a lot of simulators so I might just be used to it.

Not sure how it is in America, but in Canada you can post a note inside your mailbox stating that you don't want unaddressed mail.

https://www.canadapost-postescanada.ca/cpc/en/personal/consu...

I was curious to see what someone elses perspective was on something I routinely engage with. I wasn't sure if it was someone trolling or genuinely upset.

What are the insane takes on the main page? It's typically average american-left takes. What makes you think people with these takes are mentally ill?

Both those statements need to be prove for her, and I don't see any strong evidence for either.

And if someone was going to to make false allegations of abuse, why include specifics about how interrupting his calculus and drums caused his anger? Why not just say he was abusive, or state a more common reason for abuse? To me, the specifics make her statement more credible. Combined with his predatory history regarding women[1], I view Feynman as a distrubed individual (but a genius nonetheless)

I find the allegation credible, as I don't see why someone in her position would lie, and especially give specific details on what sets Feynman off.

Also, unless I see some concrete data about the amount/percentage of women who lie in order to get a divorce, the comment you linked is pure conjecture. Nothing really to argue about since it's just the vague idea of what people think about that time.

[1] https://thebaffler.com/outbursts/surely-youre-a-creep-mr-fey...

I don't think people really make up domestic abuse charges with this much detail. His wife explains in the post specifically what causes him to get so angry that he hurts her.

I don't see her having much incentive to lie and make up these statements, and see no evidence that she did lie. Some women lie about domestic abuse, most don't.

I don't agree with the parent commenters characterization of Karpathy, but these projects are just simple toy projects. They're educational material, not production level software.

Sure, just a few examples off the top of my head:

- Abortion bans in the US constitute abuses of female reproductive rights and affect millions: https://www.amnesty.org/en/latest/news/2024/08/usa-abortion-...

- The Iraq War, in my opinion: https://www.ecchr.eu/en/publication/the-iraq-invasion-is-a-c...

- Guantánamo Bay: https://www.hrw.org/news/2025/08/29/us-migrants-face-abuse-i... & https://www.amnesty.org.uk/guantanamo-bay-human-rights

- Yemen, broadly: https://www.amnesty.org/en/latest/news/2025/10/yemen-us-air-...

- Depending on who you ask, you could also point to large-scale violations of the rights of migrants and asylum seekers, as well as the over-policing of minority populations.

The US also does not consistently uphold the same values that you say liberal democracies should. It does business with Israel, Saudi Arabia, and others.

As for your second point:

“It’s extremely relevant. If you believe in personal liberties and democracy, you should only do business with societies that uphold personal liberties and democracy. It’s that simple.”

I would ask, why? I believe in personal liberties and democracy for my people, my community, and my country. If another country’s population does not hold those beliefs as a majority, why is that my concern? If we truly restricted trade only to countries that share our beliefs, our list of trading partners would be very small. What would the benefit be?

Additionally, given your request for citations, I suspect we would disagree significantly on which countries actually reflect our values. I am not sure we could arrive at a consistent list of partners that share our values. For example, I do not believe the USA has a strong democracy. It has a rather weak one. Should it be excluded as well?

I largely agree, but there are conflicting goals which makes it hard to evaluate if this really is a bad long-term play. Canada has environmental commitments, and giving the population access to cheap EVs will help meet those goals. I don't think this decision is just a short-term political win, there is potential for it to help with the longer term vision of Canada. But I do agree, this is bad for the local automotive industry in Canada.

I mean Canada's largest trading partner is the US, which also has many examples of large scale human right abuses.

As a Canadian, it's not really relevant to me that a country we trade with isn't liberal, and I don't agree with the premise that China is inheriently anti-west. Anti-western values, yes, but China does not threaten west violently in anyway that I can see. They mostly threaten western dominance economically.

IMO, Canada should just do what's best for its citizen, which is get good trade deals, and ensure that our values don't morph into something unrecognizable. What other countries do in their own borders is largely irrelevant.

Really, I found 5.2 to be a rather weak model. It constantly gives me code that doesn't work and gets simple APIs wrong. Maybe it's just weak on the domain I'm working in.

He saying that using a single metric like GDP isn't sufficient for claiming that the economy isn't tanking. The economy != GDP. For many regular people, it's terrible right now.

Ah I honestly don't know about STB aside from his header libraries and his tech talks, what makes you think he's a lefty nutjob? Briefly looking over his website and X profile, he seems like he's on the left side of the political spectrum, but what inparticular gives you the impression he's a nutjob?

Sean has been on Jon's stream, and there's a good video where briefly Sean states that he have Jon some syntax advice for Jon's language Jai.

I assumed they were friends as there are several videos of them conversing. The parent comment pointed out that Sean agree's with the negatives about Jon, which could not mean much, but the fact that Jon's negative as described in the Dreknek are really bad indicates to me that Sean likely doesn't view Jon as a good friend anymore. This is surprising to me because I really did enjoy one of their videos where they try and solve a problem together.

The fact that Sean agrees with this critical take of Jon is further evidence of how much Jon has changed since the pandemic.

STB Is the intial's for Sean T Barrett, who also created a software library with the same name.

Damn I thought STB and Jon were friends for a while. I find this indicative on how Jon has changed these past 5 years.

Is it typical for the marketing for a game to reference those who worked on the game? Those designers were employees of Thekla as far as I can tell, why would they get a shout out?

Jai is designed for games, it aims to do a few things that can help game developers, as well as developers in general.

- Lower compilation times for debug builds. - Better debug messages. - A standard library that comes with a production ready graphics API, so gamedevs don't need to worry about the current state of graphics API and can just dig in. - Standard input API for cross OS development. - AOS to SOA automatic conversion to simplify code that needs to be performant, while retaining a clean syntax. - A context system, which should help with simplifying functions definitions while keeping things strongly typed. - The ability to rewrite ASTs, to do compile time programming. Ideally simplifying code, while keeping runtime speed performant, and keeping compilation speed fast.

This is just to name a few off the top of my head. The performance and API stuff is directly going to help game devs. I view it similar to Odin, something that is in production software right now, where you can have a clean langugae, with a strong standard library and primitives to help you develop quickly.

Yea I largely agree with you on that point. I think when discussing Jon, Casey (and to add another, Mike Acton), there's actually a series of advice that they give that get lumped into a whole, and people don't really see the parts of what they're saying and instead focus on the part that sounds most critical to their work.

I do agree that if you take from their "teachings" that every dev needs to optimize every thing, and never use any other language than system languages, that advice is myopic for most devs. However, I don't really see them arguing for that, at least not entirely.

From following their teaching for a while, they mostly preech about the following things which I agree with, even when talking about higher-level languages, technologies.

- Get the clowns out of the car: Don't make things needlessly expensive. Write simple procedural code that maps cleanly to what the hardware is doing. This is essentially stating OOP, large message passing, and other paradigms that abstract the problem away from the simple computations that are happening on your computer is actually adding complexity that isn't needed. This isn't about tuning your program to get the highest amount of performance, but rather, just write basic code, that is easy to follow and debug, that acts as a data-pipeline as much as possible. Using simple constructs to do the things you want, e.g. an if-statement versus inheritence for dynamic dispatch.

- Understand your problem domain, including the hardware, so you can reason about it. Don't abstract away the hardware your code is actually running on too much where you lose vital information on how to make it work well. I've seen this many times in my professional career, where devs don't know what hardware the code will be running on, and this inevitably makes their code slower, less responsive to the user and often drives up cost. There are many times in my early career (backend engineering), that just simplifying the code, designing the code so it works well for the hardware we expect, greatly lowered cost. The hardware is the platform and it shouldn't be ignored. Similarly, limitations that are imposed by your solution should be documented and understood. If you don't expect a TPS greater than some value, write that down, check for it, profile and make sure you know what your specturm of hardware can handle, and how much software utilization of that hardware you're getting.

- Focus on writing code, and don't get bogged down my fad methodologies (TDD, OPP, etc). Writing simple code, understanding the problem more deeply as you write, and not placing artifical constraints on yourself.

Now each of these points can be debated, but their harder to argue against IMO then the strawmany idea of them proposing that you must optimize as much as possible. And they argue that you will actually be more productive this way, and produce better software.

FWIW, you may have some datapoints showing that they do propose what I called a strawmany version of their ideas, but I have seen them advocating for the above points more so than anything else.

---

I do want to add, for Jon Blow, I don't think he has a problem with people using engines. From what I've seen he's played, and loved games that used engines in the past, and had no problem with their output in terms of gameplay or performance. From his talk about civilization ending relating to game dev, he's more concern that if no one tries to develop without an engine, we as a civilization will lose that ability.

More like, shipped 2 hit games, which were both technological and artistic feats for their time. And developed a blazingly fast compiler. Casey also was a developer in RAD game tools developing animation tools. Their output is probably better than most industry developers. I understand if you don't like their attitudes and the way they attempt to teach/preach to other engineers, but IMO their work speaks for itself. I take their advice and try to apply it to my own work, because it seems to have work for them.

Why are they being criticized from the arbitrary metric of the last 10 years, when both had careers far longer than that? Jon's advice for software is the same advice he used when developing Braid and the Witness, which are both great games and for their time, technological feats, especially from an indie.

Jon's production from the last 10 year isn't even due to bad software methodology from what I observe, it's mainly seems to be because his company is creating a new programming language tailored to games. This doesn't seem to be done to make money, but rather, to try and fundamentally fixed issues that he perceives in game development. It's a lofty goal, and the compiler itself uses the same software methodolgy that he argues for, and it's quite good.

So I don't think this critism is fair. We should look at the arguments they present, and their multi-decade long careers as a measure of thir authority on this subject.

Sources show, the source commenter I was discussing with in this thread agreed, why are you challenging this that were established in the thread? Why are you insisting that we don't use the context in the thread to continue discussion?

So, we established that there were injuries among people surrounding those with the pagers. Therefore, the parent comment’s claim was false — the explosions could hurt people nearby and weren’t small enough to affect only the combatant.

My other points still stand, but it’s strange to me that the argument seems to go (not necessarily from you, but from other commenters above):

The explosions were too small to hurt others, so the reported number of civilians injured must be false.

We see that the explosions did hurt civilians.

Well, only a small fraction — the numbers must still be false.

Can you see how this is moving the goalposts? The argument shifted from “the explosives were so precise that Israel must have known exactly who was targeted, and those injured were combatants,” to, in the grandparent comment:

How do you know they were civilians?

Now we see that civilians were present and injured. Perhaps you're correct that the videos show only a small number, but the videos still confirm the core point: civilians were harmed.

@tptacek, I don’t have a problem discussing this with you, but each thread you respond to splits off into new points I have to address. It feels like arguing with two people making contradictory claims.

I’ll leave you with this: the videos show only a minority of the pager detonations. Civilian injuries are most reliably known by Lebanese hospitals and government sources. The idea of detonating explosives in civilian-populated areas without knowing who is immediately around those devices is deeply problematic. And there is no way Israel could have known who would be harmed with any reasonable certainty; the reported numbers only reinforce that fact.

There are videos where the surrounding people were hurt by the pagers, so, what's the explaination for that?

How do YOU know they were terrorist? What would you call people who were around the individual with that pagers?

Cmon man, there are sources pasted all over this thread from my discussion with OP. I'm not going to post the same source that was already discussed with him, why would I waste my time to do so?

OP did split this chain, but a sibling comment has the sources you want.

EDIT: Getting downvoted because I didn't want to paste the same source N times. Nice.