polygamous_bat

> In Europe, this is often handled by government purchasing discretion (not price caps as many simplistically believe). If X treatment is too expensive, it simply isnt covered. If generics medication get 99% of the job done, that is what you get.

The amount of times I’ve proposed this only to be responded with “government death panels” is astounding.

> If a CT cost is 10x at one hospital, and 1x at the other, the patient needs an incentive to seek out the 1X cost.

Healthcare is weird because the demand is inelastic and you aren’t always in a place to price shop. However, I agree with your broader point.

> The Sacklers aren’t the problem here, corruption is.

Why not both?

Bitcoin is Schrödinger’s asset, in the sense that it is whatever it needs to be to counter the latest criticisms from non-believers.

For example, if you say that bitcoin has 7 tps and even with lightning will take decades to onboard everyone, it will become a store of value that you are supposed to buy and hold for long term. But if you point out how bitcoin has crazy volatility, it may become a currency that you don’t hold, just use to transact.

How many people fall into the intersection of “wants to do matrix multiplication really fast, and willing to pay for it”, “only knows pure python and can’t touch numpy”? Is that segment of the market worth investing $100M into?

You could, however, easily block the game and render a message directing users to your actual site if you detect they are using the site in an iframe. What legitimate reason can there be for that?

I am curious, why is this an illusion? I have experienced the same (or at least similar) feeling of getting better after getting my hands on a search engine like google. I never stopped using a search engine since then to check if I am suffering from this “illusion” you’re talking about. Should we?

> Could GPT-4V be used for robotic applications?

I have a quote I like in this context, “Could you learn gymnastics just from watching videos?” However intelligent an internet trained model may be, I strongly believe you have to have some interaction with the real world to learn more complicated actions. So far, Pick and place, is all that’s been shown with the bigger models, so my hypothesis seems to be holding for now.

Yes, and as an added bonus, following the same reasoning, the restaurant can put everything on the menu as $0! Isn’t that a brave new world.

It always surprises me how some people are shocked a plot of P(student learned the class material well) is starkly different from P(student learned the class material well | student was hard-working enough to secure admission through one of the most selective* processes).

* Minus the legacies and the atheletic admissions I guess.

Any computational method that relies on a function that converts m bits (an image) to n bits (a fingerprint) where m > n will always be vulnerable to such an attack. The smaller n is compared to m, the easier it is to counterfeit something with that signature. It is not new knowledge. The only way to be certain, unfortunately, is for a human to look at it.

When the allegation is as serious as CSAM, I would rather be very very certain before accusing someone, rather than taking a shotgun approach and just randomly searching phones. But what do I know.

> The goal of such a platform should not be to make money, but to make the market more efficient.

This. Proponents of fully free markets often conveniently ignore that it’s almost always cheaper to sell products of unethical means, absent outside intervention. It’s always cheaper for nestle to sell slavery-produced chocolate than not, and nothing in free market capitalism can effectively address questions of ethics.

One, insulting and name calling doesn’t further the discussion.

Two, is there any concrete evidence that may change your mind from the current position of “businesses can’t have done anything bad so if anything has gone bad it’s the government’s fault”? If yes, and if the standards for proof are reasonable, others may try to find such evidence or fail to, thus furthering the conversation. If not, consider how dogmatic your position is, and how you can open up your mind for fruitful conversations.

Yes of course, it lets you wash your hand off of all responsibility.

Reminder to anyone else who is not so trigger happy to dox people: Stochastic Terrorism exists [0] and so does the tale of the “Meddlesome priest”[1]. It’s a trick as old as time, you’re not doing anything novel or clever, and your tactics are obvious to everyone.

[0] https://en.m.wikipedia.org/wiki/Stochastic_terrorism

[1] https://en.m.wikipedia.org/wiki/Will_no_one_rid_me_of_this_t...

> The people doing the amplification here aren’t forcing anyone to do anything, which is a critical difference.

Cool, maybe we can have a robot walk around behind them holding a sign? Where do you draw the line?

> It’s also an anonymous source to a student newspaper. I wouldn’t exactly hold them to the highest editorial standard.

It’s the Harvard Crimson, not Boondook High Weekly. I would be incredibly surprised if their editors are not graduating and entering the countries top newspapers every year.

You don’t think if Elon Musk tweets tomorrow “@polygamous_bat doesn’t like X, or dogecoins! His phone number is 0123 and address is XYZ”, that’s harassment? None of what he would say is false, but by amplifying it to his audience of trolls he just increases the chance of something “accidentally” happening to me.

Same here. Whether you say something or not, putting your face on a billboard is guaranteed to attract crazies. If your opponent is spending a million dollar to give that treatment to literal college kids, it doesn’t take a genius to see who the evil is.

Great! If you read the article instead of trying to incite a flame war you would know that they have already doxxed people falsely, and so your statement doesn’t apply at all.

Doxxing is always weird because there is no real violence until there is – it’s all implied violence – so online trolls can hide behind the facade of “I didn’t touch you!” The people signing the doc are already public, a truck roaming around with their face on it is just organized harassment at that point.

Didn’t think I would see a pro-doxxing content here, but what do I know.

That’s the problem when you get too big, isn’t it? Musk now has $149 million extra that he can spend on fighting legal battles with the SEC and still come out at least a million ahead. Does the SEC have $149m in spare resources to fight this fight?

Unfortunately this is the math I believe the enforcement agencies have to do.

Unfortunately, in my experience, to people who are concerned about “immigrant people with kids but not-interested in having a full Western-life experience (which necessitates having a car)”, this is what counts as a logical complaint.

> only way to really deal with those types is to always escalate and get more eyes on the situation.

Want to echo this, and say that it is in the psychopath’s best interest to make you feel like you are overreacting. Bringing more eyes on the situation is not overreacting, and if someone who has power over you is trying to convince you of that, don’t trust them.

Craziest thing is if you tried to institute the rules of health insurance on credit card fraud (where banks have full liability) heads would roll. However no one bats an eye when it’s for something as crucial as people living or dying.

What scares some people is not that rationing care may already exist but that there may not be any fast-track line you can buy your way into.

Now normalize those numbers by cost of living.

How can we expect people with nuance to win elections if the population has been trained to expect solutions that fit in 280 characters and/or 15 seconds of video?

I've never used WeChat, but are signal/telegram/Whatsapp considered "social media" now? I thought they belonged to a separate category of "messaging apps", which consist of people messaging others that they already possibly know from a different channel.

In my experience, the answer comes down to "does your code use classes liberally?"

If no, you're just passing things between functions, then go ahead with Jax! But converting larger codebases with classes is just significantly better with PyTorch even if they use different method names etc.

Is there a reason you are limited to JAX/tensorflow and can't use pytorch?

For a lot of people I know whose main job was not to write code, switching from tensorflow to pytorch was something that saved them ten to hundreds of hours in the long run, even accounting for the initial learning time.

Those vaguely phallic shaped rockets flying billionaires to space won't pay for themselves, you know!

But if you roll over and let them own 0your entire life, I guess you can have a pizza party from time to time.

"Unquantized" is the key word here: with quantization you can get a 4-8x improvement without much performance degradation.