@self_made_human's banner p

self_made_human

C'est la vie

16 followers   follows 0 users  
joined 2022 September 05 05:31:00 UTC

I'm a transhumanist doctor. In a better world, I wouldn't need to add that as a qualifier to plain old "doctor". It would be taken as granted for someone in the profession of saving lives.

At any rate, I intend to live forever or die trying. See you at Heat Death!

Friends:

I tried stuffing my friends into this textbox and it really didn't work out.


				

User ID: 454

self_made_human

C'est la vie

16 followers   follows 0 users   joined 2022 September 05 05:31:00 UTC

					

I'm a transhumanist doctor. In a better world, I wouldn't need to add that as a qualifier to plain old "doctor". It would be taken as granted for someone in the profession of saving lives.

At any rate, I intend to live forever or die trying. See you at Heat Death!

Friends:

I tried stuffing my friends into this textbox and it really didn't work out.


					

User ID: 454

I didn't expect a banana to give me trypophobia today. Out of a desire to upgrade my diet from becoming 100% junk food to merely 90%, I bought a bunch of them.

They arrived at a non-ideal level of ripeness, and then I let them sit for a few days. Now they're nice and yellow, but have a pattern of spots on them makes my skin crawl. Just about the only image on earth that otherwise does that is a photoshopped pic of someone's tits with holes added on, purportedly from worms.

In this case it seems particularly evident that the issue with drugs that trick you into not feeling hunger at your normal rate is that it becomes that much harder to operate normally without them.

You simply regain your appetite, and around half the weight back in a year if you stop Ozempic. I don't see how that's not strictly better than not taking it.

Unlike OP, I think a world where people can only solve their problems by becoming addicted to complex and expensive drugs is a bad one.

The hell is a "complex" drug? Does it have a hard to pronounce name? Does it have a large molecular weight? Does it act on more than one signaling pathway?

Ozempic isn't particularly expensive. Most middle class people in the West can afford it, if it's not covered by insurance. There are legal or grey market sources that are significantly cheaper. And as more alternatives arise, including generics, it'll only get cheaper.

Gym memberships cost money too.

And Ozempic isn't "addictive". Do people not know what that word means?

Is insulin addictive to a diabetic, because they'd fall sick or even die if they stopped?

Weening yourself off of the drug should be the ultimate goal, otherwise you're just embracing a different kind of slavery.

I would like to see you wean yourself off oxygen and water. Perhaps, to be less challenging or immediately lethal, clothing or shelter. Otherwise what are you but a slave to biological necessity?

This is all such immensely confused thinking that I don't know how such beliefs can even arised. At the very least, it is factually incorrect.

I'll allow it. It doesn't matter if it's rare, if it happens at all.

Does that make your guts come out? I would hope not! Without looking it up, I believe it's a rather superficial, if widespread, infection.

(You're Japanese, so I apologize for raising the specter of radiation poisoning in your presence.)

We have very few guns. That makes mugging much harder, you can run away from a knife most of the time. Mugging is very rare in these parts too, never happened to anyone I know.

In the US, being poor in the first place is a far stronger sign of serious dysfunction and inadequacy than it would be in India, where the majority of people are poor. So you have many more industrious, hard working poor people around and fewer degenerates, in relative if not absolute terms.

Of course, I think the Texan approach to such things is laudable. I'd be happy living there, even if California still has my heart.

I don't have a sufficiently strong intuition about what the typical base-rate is in the West to have a very helpful answer.

If I had to guess, I'd say it's nowhere near as bad as say, a bad neighborhood in SF, where leaving something expensive in the car is incredibly foolish. Copper stealing isn't as bad as in a ghetto in a big city, maybe.

I've never had to worry about my car being broke into. Never happened to any friends or family. I've never had the power go out because someone stole the cables. I probably wouldn't leave a large amount of gold bullion visibly sitting in the backseat, but I'm sure you wouldn't either.

If the chatbot doesn't quite have your 30 years of memories, but can make an impression that would fool anyone else, what's the difference?

It makes a difference to me. I'm the customer here! If I'm physically around to evaluate the claim, then presumably there's some kind of non-destructive mind upload going on. I wouldn't consent to a destructive one unless I had no choice, or if I was sufficiently convinced by evidence that it highly accurately captures almost all behavior and internal state.

If I died without hope of recovery, then I have no control over what others get up to. If they want to run a fine-tune of GPT o5 that mimics me via text, in an unending simulation of The Motte, and names that thing self_made_human, what can I do about it? Even I think that's a strict improvement over being dead and entirely forgotten.

It's just that I feel like your arguments prove too much, as the expression goes. If there can be such a thing as "not enough data", then indeed how can you place a cutoff point? There's never all data. You of today don't have all the data on the you of yesterday.

As far as I am aware, there is no principled and rigorous way to answer that question, at present.* It rarely comes up in normal life, because humans can't trivially clone themselves with their memories.

Some people think they live on through their children. Some think it's the books they write, or the good they do? That's good for them, or at least good enough. Highly inadequate for me.

And the prompting question was why I seek immortality, proper immortality and not word-play. That's my answer.

*I have strong intuitions on the matter, and seek to see if science and engineering can make them rigorous. I strongly believe that there's a meaningful and objective way to compare similarity between minds, in the manner you can generate embeddings on text.

If I go to bed right now, and wake up again, the new self_made_human and I will be virtually indistinguishable, even to ourselves. So we have no qualms about calling ourselves the same person. If you want proof, ask me this question again tomorrow, I guess.

The same holds for SMH from last week, less so from last year, even less so than 20 years ago. It will also continue to become less true with time. But I consider such divergence both unavoidable at present, and also entirely acceptable. I want to be able to grow and improve, ideally in a self-directed fashion.

There are operations I could undergo that wouldn't preserve identity. Say developing Alzheimer's, or having a lobe of my brain removed.

Even a clone of me with no shared memories would be a very similar person. We'd get along well, I'd treat him like family. I'd probably give him money if he asked. But he wouldn't meet my threshold for being the same person.

Strongly agreed. Blindsight is in my top 3 list, and Lovecraft? The man was afraid of miscegenation, his own shadow, and presumably, General Relativity, given how much he hates non-euclidean geometry.

I found the general concept of unknowable, eldritch entities interesting, but his execution lackluster. The Laundry Files does it way better, if we're sticking to the Mythos.

but it never comes up again and has no impact on anything.

Isn't the whole point of the novel that God is a virus, infecting our simulation?

I'd say that's pretty plot relevant.

In fact, from a realism perspective, it is entirely believable that we might discover clear evidence that the universe is a sim, while simultaneously not being able to do anything about it. I assume the people with the capability to simulate an entire universe would have better sandboxing and intrusion hardening than AWS.

There's a reason why the blurb/introduction has the following quote:

Whenever I feel my will to read becoming too strong, I read Watts

(Great book. Up there as an all-time fave.)

So why are Indian tourists, from a country where pickpockets are routinely beaten, not beating these pickpockets on the metro, where they are surrounded by other Indians, who (presumably) routinely beat pickpockets? Why is this dog not barking?

I never claimed that. I presume that they do, in fact, raise hell should they catch the culprit in the act. Unfortunately for tourists, they tend to be found in crowded places, some of which might be called tourist traps, where it's harder to notice, or figure out who the culprit was, while the latter has an easier time vanishing into the crowd.

To be clear, I'm not talking about online Indians, I'm talking about actual Indians I've met IRL, with who I've talked about life in India

Did you specifically ask them about the topic? It rarely comes up unprompted.

When Ashwin was arrested by the police in New Friends Colony, the seeds of doubt were sown in his mind.

This doesn't say anything about whether or not it was a victim who caught him, whether he was roughed up during the process, or after by the police. I've never claimed that standard means of law enforcement don't exist in India.

You claim it is, but again, I have met many Indians and I have never heard any of them claim that thieves are guaranteed an ass-whooping on the spot.

That is not a claim I've made. I've said such an outcome was "guaranteed", merely that the risk is high enough to be a real deterrent for would-be criminals.

We don't have that many Indians on The Motte, so I don't see how it makes a difference. In the wider internet, I think I'm quite unusual in being open-minded about the benefits of such extrajudicial punishment, compared to the kind of Indians you would pay attention to online.

It's not a video, it's an article. The guy mostly picks pockets of people on public transit, which tourists are usually wealthy enough to avoid and has a critical mass of witnesses and people who would be available for administering an impromptu beating

India is a poor country! Western tourists are probably not taking public transport except for the sake of it. That is not nearly as true for Indian tourists, in India.

Further, there's obvious selection-bias at play: Vice didn't choose to interview an ex pickpocket, did they? If someone is still up to their shenanigans, then they're either skilled or lucky. An ideal pickpocket doesn't want to be caught, if they did, they'd be a bandit.

I can't even imagine what it means to be "too dumb to care" about being beaten to a pulp. It is possible that he's lying, but why would he lie about this? Why would he continue to be a pickpocket if he's getting beaten every day by mobs of Indians? More importantly, how could he continue to do anything but lie in a hospital?

Criminals aren't particularly known for their keen ability to do fine risk-benefit calculations. If they did, they'd probably be more likely to look for a job. That doesn't mean that they're immune to incentives.

As I've made clear, I never claimed that pickpocketing is guaranteed to lead to a beating. If literally every pickpocket was caught right away and beaten, then I think the number of pickpockets out there would be, if not literally zero, within spitting distance. Mostly because of kleptomaniacs I suppose.

I know, which is why I've resisted the blandishments of doctors trying to sell this to me. I know I'd be one of the patients who didn't stick to the stringent lifestyle changes you have to make along with the surgery, and I'd be one of the ones who over-eat to the extent of bursting the sleeve.

https://pubmed.ncbi.nlm.nih.gov/18586571/

Those high in dietary adherence had lost 4.5% more weight at postoperative week 92 than those low in dietary adherence.

In other words, it doesn't really matter if you're a good boy/girl and listen to your doctors after you've had most of your stomach removed. Of course, bariatric surgery isn't a truly permanent solution, weight tends to come back after several years, but it was a good option before Ozempic made it somewhat obsolete.

That telling people to make lifestyle changes is highly ineffective? That implies nothing about whether or not the changes themselves don't work. I consider it clear enough, in the context of the comment. In fact, the very next sentence is:

Telling people to use their will to get over depression or diabetes doesn't, and the same is true for obesity.

The only reason we even bother trying is that telling people to do things is rather cheap and low-effort. In rare cases, they might even listen. It also makes us feel good, and ticks boxes.

I do not necessarily claim that fat-shaming doesn't work. I just think that it's cruel, and the efficacy is far less than ideal.

There are societies, like China or Korea, where the social opprobrium for gaining weight is so strong that most people will do just about anything to avoid it. It likely also drives people into depression, eating disorders, while also sucking.

Locking people in a fat camp and beating them with sticks for eating will, I strongly believe, reduce the obesity rate. I don't endorse such tactics, even if they work.

(You don't see no fatties coming out of Auschwitz, do you?)

Yes, I'm on Ozempic now. I haven't lost weight, but it'd doing good for my blood sugar. Weirdly, I'm eating both less and more, since I don't eat as much at one sitting, but now I'm constantly eating small snacks and meals. No idea what the hell is going on there

I'm not an endocrinologist, but it might be worth asking them to increase the dose. More Ozempic is, very roughly speaking, more weight lost. There are also alternative drugs that work even better, but they're harder to get.

Jokes aside, given that you were, presumably, at some point a transhumanist, I would expect that you have better arguments against it than the average person. I would be happy to discuss them, if you cared to.

It seems trivial to me that a society that can actually achieve that goal, is on the cusp of being able to simulate every mind that ever lived, and any arbitrary number of minds that didn’t. So if you think it’s inevitable that simulation will happen, what’s your concern about dying? The pain will suck, I’m sure and that’s fair, I’d enjoy a golden god body too, but it seems highly likely given your priors that you’ll just go to sleep and then wake up a simulation at some unspecified point in the future, with no sensation of loss.

A post-Singularity civilization with Dyson spheres and Matrioshka brains has a lot of energy and computational power, but it is not infinite. The sheer number of computations needed to accurately simulate a single human brain is a subject of considerable debate. In their landmark roadmap on Whole Brain Emulation, Anders Sandberg and Nick Bostrom surveyed estimates that ranged from 10^18 to 10^21 floating point operations per second (FLOP/s). Let's be charitable and take a figure on the lower end, say 10^18 FLOP/s, just for a real-time simulation.

Now, let's consider the scale. The Population Reference Bureau estimates that about 117 billion modern humans have ever been born. Of those, about 109 billion have died. If we wanted to bring them all back for, say, a simulated century, the total compute required would be 109 billion people * 100 years * 3.154e+7 seconds/year * 10^18 FLOP/s. This works out to something on the order of 10^41 total floating point operations. That's a big number, but within budget for a Kardashev-II. But this calculation completely ignores the monumental task of getting the data in the first place. This is, as I see it, the fatal flaw:

For the vast majority of those 109 billion deceased humans, the information that constituted "them" is gone. Utterly. This is what cryonicists call "[information-theoretic https://en.wikipedia.org/wiki/Information-theoretic_death)". It is the point at which the physical structures in the brain that encode memory and personality are so thoroughly disrupted that no technology, even highly-speculative, could in principle recover them. Think cremation, advanced decomposition, or having your constituent atoms scattered across the globe by the nitrogen cycle.

Let give an example:

Unga and Bunga were, someone claims with authority, Siberian cavemen from 27,000 years ago. They were unfortunately, not lucky enough to be frozen in permafrost. They have decayed, and all we're left with is maybe a fragmentary sample of DNA or the odd bone. We know next to nothing about their lives, their language, or their culture.

Even with an unlimited budget, we could not bring them back. I don't think even the most powerful ASI around could.

The challenge is not just that ancient DNA is typically fragmented and damaged, but that DNA is merely the blueprint for the hardware. The "software," the connectome, the specific synaptic weights, the epigenetic modifications, the entire lifetime of learned experiences that made Unga Unga and not just some generic instance of his genome, has been overwritten by entropy. The information is lost, and the second law of thermodynamics is a harsh mistress. Reconstructing them from scratch would be, if not a physical impossibility, close enough for government work.

Even for a modern person with a vast digital footprint, the problem remains. You could fine-tune an LLM on my blog, my comments, and every email I've ever sent. You could supplement it with video, audio, and detailed recollections from my family. The result would be a very convincing "self_made_human" chatbot, but it would not be me. It would be a sophisticated imitation, a high-fidelity echo that lacks my internal continuity of consciousness.

I demand far higher fidelity. I want to test such an entity in both a black box, and ideally, look at the simulated neurons and the information they contain.

The same problem applies to cloning. You could take my DNA, create a clone, and raise him in an environment designed to replicate my own. But how would you transfer my almost 30 years of memories, my specific neural pathways? He would be, at best, my identical twin brother, separated by a generation, not a continuation of my existence. He and I would agree that he is not me. Not in the same manner as the me of yesterday or from next year. There just isn't enough data.

(If he doesn't, then something definitely broke along the way)

This brings us to cryonics. It's not a guarantee, not by a long shot. We do not know for certain what level of structural and chemical detail must be preserved to retain the information essential for personal identity. But it is, at present, the only method that even attempts to bridge the gap between clinical death and information-theoretic death by preserving the brain's physical structure.

As others have argued, the decision to sign up can be framed as a question of expected value. If you assign even a tiny, non-zero probability to the success of future revival technology, the potential payoff is a lifespan of indeterminate, potentially astronomical, length. It's a bet against the finality of decay, a wager that our descendants will be much, much smarter than we are. I find that a slim hope is infinitely better than no hope at all, but I would still very much prefer to not die in the first place.

What do you think? Do you want to spin this conversation off somewhere else?

Well, that's what I think haha.

I now mildly regret that whopper of a post being so far down thread. I usually prefer to write out in public instead of DMs (I'm vain enough to want the public to praise me for the work I do). You're welcome to reply here, or post a top-level comment elsewhere, or whatever you please. Even DMs, though I would be saddened by going through all this effort behind closed doors.

I'm not being circumspect. I am merely elaborating on a point I've made, which was interpreted in a sense I didn't intend (I'm not accusing you of an intentional, bad faith claim, misinterpretations happen). I don't see any additional hedging, or caveats at play.

I would like to think I'm usually quite clear in what I mean via what I say, not that this is any guarantee of people interpreting it exactly as intended. Even legal documents and contracts, intended to maximize clarity and leave no room for error, often end up in the courts. They also make for riveting reading.

The kind of adoption process that involves traveling the world to find the perfect orphan is straight-up child buying, and has some moral similarities to eugenic embryo modification.

You say that like it's a bad thing. I consider both of them to be fine, not that I'm looking to adopt, or at the very least "not my business".

But that may simply be a further question of degree. If we really did build a genuine superintelligence, unfettered by "alignment" to some other human's political agenda, would I not be wise to submit myself to it? I feel grateful to doubt that I will ever face such a choice.)

I'm sure that I'm notorious enough that my own subjective likelihood of facing an ASI in my lifespan doesn't need elaboration, and I'll skip over my usual arguments.

I think that, compared to life as it is right now, accepting the rule and oversight of a benevolent superintelligence would be grossly superior, and beats rule by humans in just about every metric (barring your ability to rebel, should you have strong feelings on the matter). They are likely incorruptible, smarter than the average politician, and thus far better placed to consider the likely outcomes of their policies. They might even, at least theoretically, be democratic and defer to the opinions of us retarded humans. I'd hope so, at least, since we're the ones building them to fulfill our whims.

(If they're not benevolent, gg I guess)

That being said, that's not what I consider an ideal world. I'd much rather use the kind of technology available in a post-Singularity world to improve myself and rapidly bootstrap to the level of an ASI so I can exert agency and be treated as a peer. I'd rather not be beholden to anyone, no matter how kind and wise.

This is of course, a rather aspirational goal. About the same as me saying that utopia-with-free-blowjobs beats utopia-without.

You've just made it clear that even if I were to produce any figures, you wouldn't believe them. Which isn't even the wrong approach, since crime stats from the Third World are notoriously unreliable, and this isn't as cut and dry as murder.

That being said, the gold standard for comparative statistics when comparing crime rates between jurisdictions is the murder rate. Because, well, murder is a pretty big deal, hard to hide, and the cops, even if lazy and incompetent, are usually not that awful.

India, according to UNODC figures, had a murder rate of 2.94/100,000 in 2021.

The global average is 5.19 in 2023.

The World Bank claims 11 for "low income countries". 10.9 for all of Africa.

The United States? 5.9

I'm not bold enough to immediately jump to claims that the same ratio holds for other forms of crime, but yes, you are far less likely to be murdered as the average person in India compared to the global or third world average. We even beat the States, which is unusually awful by Western standards.

I have no objections if you wish to consider my claims to be entirely anecdotal. I stand by them regardless.

It's a fool's errand to trade anecdotal evidence. I'm not claiming that kind of crime happens at all, I'm stating that the amount of crime that happens is lower than it would be, because of the fear of extrajudicial punishment.

You would absolutely face more risk if you were a super-crowded environment, or if you were an obvious tourist. That's true just about everywhere where pickpocketing happens at all.

India is a very big country. I don't know anyone in my friends or extended family who was robbed or pickpocketed in maybe the last 3-5 years. That included both urbanites living in desirable neighborhoods, and family who live in villages and small towns. The latter had burglars rob them multiple times, but the last instance I can remember was at least 15 years ago, and it hasn't happened since.

what you mean by that? Out of lets say 25 people you know - how many were pick-pocketed within last year? How many were robbed?

Zero. And zero. As far as I'm aware, no one I know has been pickpocketed or robbed in the last, uh, maybe 3 to 5 years?

and while petty crime is annoying - what about more serious crime? Lets say that woman goes alone during night though city - is it likely that something bad will happen to her?

Depends on when and where. In just about every major city in India, as is the case for most of the world, there are "good" and "bad" neighborhoods. I can think of a dozen places where I'd be unconcerned about being a woman wandering around after dark, and more where I would be.

There's a reason why I was careful to only talk about petty crime, and mostly property crime within that range, because, as I've elaborated in another comment, this doesn't hold nearly as true for sex crimes.

Every Indian woman I've met seems to have a story about getting groped in public and the offender never gets beaten by the upstanding citizens that you claim inhabit the subcontinent.

Notice that I was very careful not to make this claim about sexual harassment.

Because that wouldn't be true. But why the difference?

Because a pinched ass is not a smoking gun! If someone gropes you up, there's usually no evidence a crime was committed at all, unlike most forms of theft, where you can, at least some of the time, show that the culprit is carrying your wallet or phone. The worst they can do is try and drop the evidence, which is usually not the same as destroying the evidence. My wallet lying ten feet away from you is, shall we say, suggestive.

Also, and I hope this is obvious, being groped is far more embarrassing than being robbed. This is just as true in India than is in the West. How many women cat-called by construction workers in NYC go on to file a complaint for sexual harassment?

Another important factor is that, because of the lack of evidence, it often boils down to a he-said-she-said situation, which bystanders are usually quite loathe to become involved in.

You're the only Indian I've met who claims that people would get beaten in the street for various transgressions and I really find it hard to believe.

My man, did you bother to do something as simple as Google the phrase thief beaten up India?

Or did you look up "man beaten for groping woman India"? Because yes, that has happened.

Here is a vicesplainer on the career trajectory of one Delhi pickpocket. He joins a gang that has so much opportunity for larceny that they're pickpocketing around the clock in shifts. He certainly doesn't fear retaliation from honorable bystanders, the only thing he seems to fear is the gang after he tells them he's out.

I don't have the time to watch the video right now, but my expectation is that they're preying on tourists primarily, which would be easily true in Delhi. They could also, more tentatively, simply be lying about the risks of being caught, or too dumb to care.

  1. You will lose muscle along with the weight, if you solely use it for weight loss without dieting. But the degree of muscle loss is about the same as going on a diet, fasting, or, if memory serves, bariatric surgery. If you exercise alongside, you can stave most of this off.

  2. Indeed. You will regain roughly half the weight you lost in a year of use after a year of disuse. But no other negative effects, to the best of my knowledge.

(I take oral semaglutide, so I'm putting money pills where my mouth is.)

Scott, as always, has an excellent write up:

Why does Ozempic cure all disease?

While I agree with you, for the most part:

This is bullshit. Especially as the beatings would likely be administered by the husband with no judicial oversight. I mean, sure, if the husband had beaten his wife for no reason on the general principle that she should live in terror of him, it would have been very likely that she would not have picked up her hobby of sexting convicts. But this is like suggesting that cobalt bombs are a good way to stop wildfires in California: while technically correct, the cure would be worse than the disease.

Despite what Western media reporting might have you believe, the rate of petty crime in India is surprisingly low. People rarely get pick-pocketed or robbed. Do you know why?

Because if caught in the act, the perpetrator would be rather unceremoniously beaten to a pulp, both by whoever caught them, and any civic minded individuals present. You can get a nice crowd going, it's fun for the whole family.

This is of course, strictly speaking, illegal. Yet any police officer, if asked to intervene, would laugh, shake their head and say the criminal deserved it. If the crook had the temerity to file charges, he'd probably be taken out back and given a second helping to change his mind.

As far as I'm concerned, this is strictly superior to prevailing Western attitudes regarding property crimes or theft. A shopkeeper who discovers someone shoplifting has very little legal recourse, the police rarely do any more than file a report and then give up on pursuing the matter. Giving them the de-facto right to take matters into their own hand and recover their property? The shopkeeper wins. Polite society wins, the only loser is the thief, and in this case the process is quite literally the desired punishment.

Before you ask, the number of false positives is negligible. I've never heard of anyone being falsely accused in this manner (at least with accusations of theft), and I've never had to have that particular fear myself.

I am, in general, against husbands beating their wives. Yet, in this specific scenario, I could hardly fault the poor chap should he be forced to resort to such methods to protect his own family. At the very least, I'd vote to acquit. It's a bit moot, because with prevailing Western norms, he likely didn't even consider a haymaker as a solution to his problems. In general, that's a good thing.