HNHacker News
TopNewBestAskShowJobs

short_sells_poo

2,257 karma · joined July 19, 2013

submissionscomments
short_sells_poo··on Nuclear fusion: it's time for a reality check
I'm not an expert, but I actually feel like it's not terrestrial power generation where fusion would be useful. We have plenty of ways of generating power here, including a massive fusion reactor up in the sky which will continue operating for the foreseeable future :)

Instead, I think need to find some way of generating a lot of power over extended periods of time if we want to crack space travel beyond lobbing items into Earth orbit. Particularly for human space travel.

We need significantly more energy density than what chemical motors can provide, and nuclear is the easiest and most "within-reach" that we know of. Things like anti-matter and even more exotic forms of propulsion are pure sci-fi for the time being (in terms of practicality). I suppose fission could work in a very primitive way of basically detonating small nukes behind a pusher plate, but fusion feels so much more elegant (at least to a layman). Mostly in the sense of not having to use effectively the same nukes we stick in weapons. Something tells me that this will never fly (excuse the pun), for obvious reasons.

short_sells_poo··on IAC confirms existence of a Super-earth in the habitable zone of a Sun-like Star
It's murky concept though, so I wouldn't bet on mind uploading in our lifetime.

Some fairly fundamental things need to be answered first. Things like how does consciousness and the sense of self arise.

Naively, if you think that you are just a program that runs on a bunch of neurons in your brains, and that this program can be uploaded to a computer, you are still left with a very annoying problem: you upload a copy, and leave the original running in your squishy brain. So what then? Do you kill the original? But that involves killing a living and breathing human being. Do you wait until it dies naturally? That's still not a good answer, because you have to die so that a copy of you can continue existing.

So until we figure how to actually "teleport" our consciousness to some other host, we are in a pickle. And there's absolutely no evidence that we'll ever be able to do this teleportation. What if we never figure out the physics to do this?

Edit: I suppose you could sidestep this by generating fully digital consciousnesses that mimic what a human brain does. So a fully digital human. Assuming we can brute-force simulate a real human brain, this should be at least physically possible (as opposed to teleportation), but this still raises philosophical questions. What you are generating then aren't human beings, but conscious AI. You could argue that the human race would eventually be supplanted by immortal AI that are no longer bound by biology, but I'd argue that this isn't an evolution of the human race, rather a completely new life form (if you can call it that), which has nothing to do with humans except that we created it.

short_sells_poo··on IAC confirms existence of a Super-earth in the habitable zone of a Sun-like Star
They might have to be entirely aquatic at that point or perhaps have some very exotic skeletal structure that can support their weight. Perhaps ball shaped beings who roll around instead of any sort of walking? I'm unsure how would they deal with a small incline.

Weight lifters are able to lift 6x their body weights, but it's not a sort of load that we could profitably exist under long term. We'd need to have extremely thick limbs and at some point that won't help either because of the cross-section vs. weight scaling law. It's also a sort of weight where any kind of leverage against a joint will generate massive forces. E.g. try catching a 50kg falling weight, you are likely to dislocate your joints and/or break some bones. And yet 1/6 of that is entirely manageable.

short_sells_poo··on Black Swan's Taleb Says Nvidia Rout Is Hint of What's Coming
> crashes occur with much higher probability than a normal distribution would assume.

This is probably the most intuitive way to reason about it, but there's a subtlety there that relates to the magnitude of the crashes and their probability of happening. Intuitively, if the normal distribution was correct in estimating market behavior, you'd not expect to see 10 sigma events ever (we are looking at something like 1/10^20 of this happening). And yet, these events happen with some regularity when you try to use normal distributions for market returns.

So there are two aspects to this: one is that the normal distribution and people underestimate how often crashes can happen, and they also underestimate how big those crashes can be. It's the latter that is arguably more dangerous, because if you lose money more frequently than expected, you might end up earning less, but if you underestimate how bad a drawdown can be, you can be easily wiped out before you can react.

So why do we use the normal distribution so often in finance? Because it is convenient. It works fine for 99.99% of the cases, and it is easier to deal with the tails as a special beast rather than always having a complicated model to look at. There's also an element of lottery. Significant crashes happen once every 10 years roughly, so you don't need a great deal of luck to make a lot of money without seeing one :D

> managing options is hard. when do you close the hedge after it gives you a profit? when do you put it back on. It is definitely quite hard. Options are nonlinear instruments with a significant complexity to their behavior. This is why people pay Universa and other, lesser known tail risk funds to handle this complexity. You also have instruments like Variance Swaps, which allow one to easily lock in convexity pnl - considerably removing the timing aspect. You can construct these instruments synthetically using options, but this is not for the faint of heart.

It is doable on your own, but not without understanding options in depth and having tools that allow you to manage an option portfolio. These days, a person who is decent at freshman math and knows how to use python+pandas can easily manage this entirely on their own with a handful of scripts, but they'll likely need to spend a few months learning the theory and then it can take a year to build the neccessary intuition.

short_sells_poo··on Black Swan's Taleb Says Nvidia Rout Is Hint of What's Coming
I'm all for some Taleb bashing, particularly in the midst of a massive tech bull market case around AI, but I wouldn't read this as him predicting a market crash.

The gist of his advice is generally that for the classic equity investments (ie buy the Nasdaq), your return distribution is heavily negatively skewed. You make money most of the time, and occasionally you suffer a fairly significant drawdown - ie a market crash. Nobody knows when the drawdowns will happen - that's exactly his thesis. Don't try to time the market, that's a fool's game. Structure your portfolio such that when the crash happens, you cut this convex downside.

Another way to think about this is that 90% of your returns are going to be a normal distribution, but the tails become something like a pareto. The simplest thing to do is to hold convex payoff assets - ie buy puts. The devil is in the detail of knowing enough about options to determine what to buy to avoid bleeding all your profits away. This is the business model of Universa - they know what to buy, how much of it, and to a lesser degree when to buy it.

short_sells_poo··on Black Swan's Taleb Says Nvidia Rout Is Hint of What's Coming
I wouldn't read this as Taleb predicting a crash. The gist of his advice is generally that for the classic equity investments (ie buy the Nasdaq), your return distribution is heavily negatively skewed. You make money most of the time, and occasionally you suffer a fairly significant drawdown - ie a market crash. Nobody knows when the drawdowns will happen. His thesis is that by their nature, they cannot be predicted.

Another way to think about this is that returns are linear, and drawdowns are convex. 90% of your returns are going to be a normal distribution, but the tails become something like a pareto. The simplest thing to do is to hold convex payoff assets - ie buy puts. The devil is in the detail of knowing enough about options to determine what to buy to avoid bleeding all your profits away. This is the business model of Universa - they know what to buy, how much of it, and to a lesser degree when to buy it.

short_sells_poo··on Boom Supersonic to break sound barrier during historic test flight today
That's a fair statement. I'm not well versed enough in commercial aviation to be able to judge how much flights would cost with externalities priced in.
short_sells_poo··on Boom Supersonic to break sound barrier during historic test flight today
This. If time didn't matter, people would use trains instead of short distance flights. And many do! Many European cities that are a 2 hour flight apart also have an ~8-9 hour train connection, which is much less hassle and stress than the flight, but people who have to travel a lot will care about the time savings of even a few hours because it adds up over time.

Lots of people have to regularly travel between Europe and US, and if they can save 50% of the travel time, it ends up being a significant amount of time. I believe the market for these flights aren't families going on a holiday, rather people who have to travel on business 3-4 times a month.

short_sells_poo··on Nvidia, ASML Plunge as DeepSeek Triggers Tech Stock Selloff
That's basically what I wrote: "it simply means that if DeepSeek is legit, you need much less NV hardware to run the same amount of inference as before."

So I still don't understand what it is that you are so strongly disagreeing with, and I also don't understand how having owned NVidia stock somehow lends credence to your argument.

We are in agreement that this won't threaten NVidia's immediate bottom line, they'll still sell everything they build, because demand will likely rise to the supply cap even with lower compute requirements. There are probably a multitude of reasons why the very large number of people who own NVidia stock have decided to de-lever on the news, and a lot of it is simple uneducated herding.

But we are fundamentally dealing with a power law here - the forward value expectations for NVidia have exponential growth baked in to the hilt, combined with some good old fashioned tulip mania, and when that exponential growth becomes just slightly less exponential, that results in fairly significant price oscillations today - even though the basic value proposition is still there. This was the gist of my comment - you disagree with this?

short_sells_poo··on Operation Leg: When the RAF airdropped a prosthetic leg into a German POW castle
I'm not a military thinker, so I might well be off on this, but I'd say because that would usually results in a draw, whereas the commanding officers generally want to win so that they can achieve the overall plan set out by strategic command - and want to achieve a crushing victory from which the opponent cannot recover from.

I guess an army on the retreat who would favor an instant draw would favor the fraternizing, but generally by that point the winning army - including the soldiers - are not amenable to stop their advance and play a friendly game of football.

short_sells_poo··on Microplastics block blood flow in the brain, mouse study reveals
Fair points, but perhaps let's look at it closer. Perhaps education has to be commensurate to people's access to information and the ability of power blocks to project influence. I agree that we are surrounded by knowledge and education, but we are also surrounded by disinformation, propaganda, marketing, adtech, etc...

I agree that people have never been as educated as today, but it was also much more difficult to project power over distances. Today, a fool in power can travel to the other side of the world in half a day and bully countries there into submission. More to the point, the fool doesn't even need to travel, they can do the bullying over a zoom call instantaneously. An extremely connected world with a vast number of feedback loops would need a proportionally more educated populace with much more robust foresight, no?

For what it's worth, I agree with your fundamental thesis that a lot of these issues are basic human nature, and unless that changes, anything we do will be highly ineffective. But we have to start somewhere, right? And it's much easier to sell a populist agenda to an uneducated populace who can only think forward 8 hours.

Consider this: people have been known to vote for increased taxes to pay for various public good projects. It can be done. Perhaps not necessarily on very large scales, because it's difficult to feel unity with people a 1000 miles away who are completely foreign to local concerns, but perhaps that's an argument for direct democracy. But direct democracy requires a population who is educated - at least in civil matters - and can put the long term goals in front of the short term pain.

short_sells_poo··on Operation Leg: When the RAF airdropped a prosthetic leg into a German POW castle
I'm not condoning it at all, but I think it's easy to understand why commanding officers didn't like their troops fraternizing with the enemy. If GI John and Soldat Hans share a coffee and a cigarette together, and perhaps even a pleasant chat, they will quickly come to realize that both of them are human beings, with human feelings, a family worrying about them, children possibly, hopes and dreams, and so on. And once you are acutely aware that your enemy is just like you, only perhaps speaks a different language, you are much-much less inclined to shoot them and stop being an efficient soldier.

I've never been in a war, and hope I never live to see myself in one, but the only way I can see one human viciously trying to kill another is by not thinking of their enemy as a human being, but rather just a thing.

short_sells_poo··on Microplastics block blood flow in the brain, mouse study reveals
That's because the at best, the government can be a penalty condition in the objective function, and recently it's effect has been almost completely eclipsed by the near term gains. I believe what we'd need is education. The more educated the masses are, the more likely it is that they vote for governments that have the long term interests in mind, and the more likely it is that the government itself will then prevent swings to short term thinking.

There are many feedback loops in the system, and we are so far down the short term maximization route that it'll be mighty difficult to swing it back.

short_sells_poo··on Nvidia, ASML Plunge as DeepSeek Triggers Tech Stock Selloff
That's literally what I wrote in my post, which the parent disagreed with. You could disagree with the part that it is because inference is now cheaper - but again I'd argue that's just a different way of saying there's no moat.
short_sells_poo··on Nvidia, ASML Plunge as DeepSeek Triggers Tech Stock Selloff
Loving the absolute 100% confidence there and the clear view into all the traders' minds that are trading it this morning.

If they'll sell everything they make and it's all about the moat of their clients, why is NVDA still down 15% premarket? You could quote correlation effects and momentum spillover, but that is still just the higher order effects I mentioned about people's expectations being compounded and thus reactions to adverse news being convex.

short_sells_poo··on Nvidia, ASML Plunge as DeepSeek Triggers Tech Stock Selloff
It's not just the usual herding behavior though. There's a convex response to news like this because people look at higher order effects like the growth of growth for stocks. Basically the DeepSeek story is about needing 40x fewer compute resources to run inference if their benchmarks are true. The dip doesn't mean that NVidia is now doomed, it simply means that if DeepSeek is legit, you need much less NV hardware to run the same amount of inference as before. Will the demand rise to still use up all the built hardware? Probably, but we went from a very stratospheric supply constraint to a slightly less stratospheric one, and this is reflected in the prices. Generally these moves are exaggerated initially, and it takes a bit of time for people to digest the information and the price to settle. It's an oscillating system with many feedback loops.
short_sells_poo··on Impact of Low Temperatures on the 5nm SRAM Array Size and Performance
Right - very good point! But this is really only relevant for HFT algos, almost everything else is much less sensitive to speed and is also more parallelizable.

For HFT to work, it needs to be colocated I believe, and I haven't heard of anyone trucking in liquid N2 or producing it on premises though. Not saying it isn't happening, I'm involved in mid freq trading so I only have circumstantial knowledge.

short_sells_poo··on Impact of Low Temperatures on the 5nm SRAM Array Size and Performance
Ah thank you! I'm surprised they'd pick an odd temp like 10k then. I had a vague memory of this being close to the He boiling point but couldn't remember how tight the margins were.
short_sells_poo··on Impact of Low Temperatures on the 5nm SRAM Array Size and Performance
77k is basically the boiling point of liquid nitrogen, and 10k is probably the same for liquid helium. Liquid nitrogen is in ample supply and is not difficult to manufacture, I suppose one could have a facility on site to produce it and use it immediately. It is going to be very energy intensive though... to answer your question, I struggle to think of a scenario where it would be better than buying more compute power. I suppose for stubbornly serial workloads... but I'm not sure what that could be? Running Crysis at 20k resolution?
short_sells_poo··on Open Socrates by Agnes Callard review – a design for life
If we set up a ladder of abstraction, going from the most practical to the most abstract discipline, I think we'll arrive to something like: Engineering -> Physics -> Math -> Philosophy

We progressively move from the How to the Why, Philosophy being the ultimate Why?

Then again, I'm but a humble engineer so why would my opinion matter :D

short_sells_poo··on Don't use cosine similarity carelessly
If you imagine a cartesian coordinate space where your samples are clustered around the origin, then a zero vector will tend to be close to everything because it is the center of the cluster. Which is a different way of saying that there's a bit of nothing in everything I guess :)
short_sells_poo··on LLM based agents as Dungeon Masters
I find this idea fascinating, and I think a lot of very interesting emergent behavior can be added to video games by having something like an LLM act as a Dungeon Master. Rather than having to pre-program a bunch of choices that the player is allowed to pick between, have an AI model evolve the story and just move the pieces on the story board based on what the player actually does. Story driven games these days are still very much railroaded. They are like interactive movies. The player has very limited agency. They can't really pick any choice which the designers haven't thought to implement upfront.

At the same time, I'm personally not super enthusiastic about using LLMs for a pen and paper game. Let's assume for a moment that we have free will, then the most fun aspect of PnP dungeon mastering is the fact that the DM has agency. It adds an (IMO) very fun mini-game where the players try to get away with as much shenanigans as they can, and the DMs job is to try to thwart these shenanigans in the most entertaining ways possible. The keyword is on "try" - the GM has to put entertaining obstacles in front of the players to keep the game fun, but ultimately the players should have an avenue of success. Even if they don't always succeed, they need to succeed enough and when they fail they need to feel that it was fair.

I feel that with an AI DM, this goes out the window. At least with the current crop of AI, I would never feel that I'm facing an agent with it's own free will, rather just a machine. As such, it cannot replace a human DM for the PnP purposes.

short_sells_poo··on Backdooring Your Backdoors – Another $20 Domain, More Governments
Food is tricky. The food supply is one of the highest national security concerns IMO. Free market proponents love to go about saying that growing food should be left to countries and regions who do it well (due to climate and infrastructure), but if your country cannot grow enough food to supply it's own citizens' basic calorie needs, you are literally living on borrowed time. If the food supply is cut off for any reason, things go down very-very rapidly and the government has days, if not hours to sort things out before things descend into chaos.

At the same time, governments do not have a good track record of running the food/ags industry. I guess a system where the government heavily subsidizes it and incentivises domestic production, but lets farmers do their thing is probably as good as we can do?

short_sells_poo··on Backdooring Your Backdoors – Another $20 Domain, More Governments
Certainly, and to be clear I'm not arguing for communism as a realistic system. It would be ideal in an ideal world without greed and selfishness. As long as those exist, we need to have a system that functions when the individual actors place their own interests far above the interests of others.
short_sells_poo··on Backdooring Your Backdoors – Another $20 Domain, More Governments
Exactly. Democratic and highly civilized countries still enforce property rights with pointy sticks. They maintain their claim on their territory against outside invaders with the army, and internally they enforce the laws of ownership using the police.
short_sells_poo··on If we had the best product engineering organization, what would it look like?
I wholeheartedly agree with your thoughts, with perhaps the exception of the CEO. Or rather, the CEO needs to be someone who will be followed by the rest of the organization. Their decision making could be replaced by an AI, probably more readily than the specialist engineers in fact, but it comes back to humans being tribal. Would the middle management follow an AI overlord? Would the engineers buy into the AIs decisions?

The CEO's job is to decide on the future direction of the company, and then convince both the owners and the employees that this is a good direction. The first part is easy to replace with AI, the second part isn't. At least not for now.

I guess taken to the limit, the CEO will become replacable the moment that the employees have been replaced with AIs, because that that point there's nobody left to lead really. One could actually argue that this is tantamount to the CEO cutting off the branch they are sitting on. After all, once all the employees are AIs, what's to stop the shareholders from saying: "Hold it right there Steve/Jeff/Mark/etc, why are we paying you big bucks? You can be replaced with an AI that will make much better decisions, and there are no employees left to lead anyway."

short_sells_poo··on Backdooring Your Backdoors – Another $20 Domain, More Governments
Ultimately it comes down to force. The person with the pointiest sticks will likely be able to enforce their view about ownership over others.

Taken quite literally, property is armed theft from the commons I guess. Unfortunately, it's tricky to do otherwise in a loosely organized swarm of barely tribal actors, because any peaceful society based on shared ownership will be prone to exploitation by malicious actors. It's basically a very large prisoner's dilemma: the global optimum would be to abolish private property, but as long as there are (enough) people around to exploit the situation for their own benefit (and to the massive detriment of everyone else), we have to stick to a sub-optimal system where everyone is worse off than the optimum.

short_sells_poo··on Backdooring Your Backdoors – Another $20 Domain, More Governments
I tend to think is neither of those, but meant very literally. For that reason I like it and I think it's an interesting subject.

What is ownership after all? The universe does not seem to have any form of ownership embedded in it's fundamental laws. If ownership is a human construct, then it is only meaningful insofar as a group of humans agrees on it.

I can stroll up to the White House and declare that I own it, but I'll struggle to convince a sufficient number of other people that this is true. If I can't assert my ownership, then I don't really own it, do I? It doesn't matter whether it is just, or fair (again - purely human constructs), ownership only matters if it can be enforced.

Being a human construct, it is also by definition temporary. It is only valid as long as humans are around to enforce it, and humans are fleeting. Humanity might endure, but there's no reason to think we are going to be around for eternity.

So it looks like ownership is not only temporary, but it is also fickle. People routinely disagree on ownership and are willing to kill- or be-killed for asserting their claims.

It looks like neither the communists, nor the liberatarians are in the right. Things will be owned by whoever has more pointy sticks :D

short_sells_poo··on I deleted my social media accounts
I don't think the parent poster was arguing to exclude themselves from social life or do coding instead of talking with people. They merely argued that it's better to have fewer but meaningful and deep connections with people you genuinely care about (and they care about you), rather than having a 1000 meaningless connections with people who are basically strangers on facebook.

The role social media plays is in encouraging large numbers of superficial relationships, rather than a small handful of deep ones. It stands to reason: I don't need facebook to keep in touch with a dozen close family and friends. I can do that perfectly well in person, or over phone calls/messages. What the various social media apps did was kill the close circle of friends in favor of having 1000s of followers and turn everyone into a one-way broadcaster.

short_sells_poo··on Getting silly with C, part (void*)2
Sooo, assembly :)
← PreviousPage 7 of 29Next →