auganov

Mainstream psychology is basically a more popular version of scientology with "therapy" instead of "auditing". Can you guess what my politics are more likely to be based off that sentiment? If so, then that tells you all you need to know.

The insidious thing is "you aren't prompting it correctly" is kind of a truism. For every possible output there almost certainly is a prompt that produces it (at worst you can just tell it exactly what to output verbatim). The true believers can already do all their programming via ChatGPT regardless of whether or not there are real productivity gains. Not that different from all the other tools people claim turn them into 10x programmers while others remain unconvinced. So as long people enjoy the format, it's here to stay.

These kinds of arguments come up a lot when it comes to homeschooling. But would anyone follow the same logic when picking a school? There are certainly nice schools that don't have too many of these problems. Are these bad choices then? Should one find the most ghetto public school? If your country has only nice schools, how about sending your kids to a rougher country?

I'll have to agree with this comment by "Jason":

   Victor Mair quoting a colleague said:

   "In 1998 when the reformist president, Mohamad Khatami, came to power he tried to eliminate this slogan from the political scene and he suggested to replace any 'marg bar' slogan with 'zendeh baad my opponent'."

   Why would he try to do this if "marg bar X " neutrally means "down with X", not "death to X". Similarly:

   "Alireza is right to try to downplay the slogan — for two reasons: 1) This is an early revolutionary slogan that is quite exaggerated. 2) Not many Iranians actually subscribe to this view…The official government uses this slogan, but not many actually take it seriously."

   Again, why is your colleague /so concerned to downplay the significance of this slogan/ as "exaggerated" (whatever that means — do they mean "hyperbolic") and "not taken seriously" if it merely means "down with America?"

   Frankly this piece reads like pure political drivel of the type that's increasingly infecting Language Log of late, when it seems perfectly obvious that:

   1) Marg bar Amrika literally means "Death to America"
   2) Native Persian speakers are perfectly aware of this fact, or else why try to euphenize/euthemize the expression or downplay how "seriously" Iranians take it?

   Reza Mirsajadi's effort in sophistry reminds me of David Irving's attempts to prove that "ausrotten" doesn't mean "exterminate". And indeed, given what happened to the Shah, the fact that the phrase originated in the Iranian revolution is the opposite of comforting.

   Violent metaphors have a long provenance in all langauges, but in the age of terrorism and the twitter soundbite every English speaker has had to learn to avoid them /precisely because/ you can never know if such threatening language is sincere or not. I fail to see why Iranians should get a free pass and a translational obfuscation courtesy of Mirsajadi on such language. Can anyone trust Mirsajadi to translate honestly after this particular effort?

You may as well claim "go kill yourself" is an idiom that means "I don't like you". An idiom is when it's nearly impossible to infer the actual meaning based off the literal meaning of the words. A hyperbole or a metaphor does not make an idiom. Death is invoked here exactly because of what it implies. "Death to X country" is not any less metaphorical in English. A country obviously cannot die.

These two parameters you're mentioning were chosen to optimize for startup performance. By definition, they have to be preselected. I may be wrong, but it sounds to me like you're under the false impression that this quoted stuff is somehow subject to jitting. You could be more sophisticated about picking these parameters for sure, but that would be orthogonal to jitting.

If you're content with where you are then that's fine. I won't claim it's going to be easy to change things. But I do think there are few things you could change that will nearly guarantee success:

- There's no good or bad food. The number is the only thing that matters. Don't feel bad about eating chips and candy everyday. That's pretty much what I do. All you need is to hit the number.

- Always keep track of how much you've eaten. A big binge is not an excuse to stop counting. After a year or so, counting will become second nature and can be done purely with mental math. But initially, you want to write all the numbers down. Date | calories in | morning weight.

- Don't tie the counting efforts to any other effort. You mention walking 5 hours a day. That's some other unrelated thing. Don't ever mentally link this to your weight loss effort. If you want to do it, do it, but it should have nothing to do with losing weight. Counting and keeping records is your primary job.

- Your family should know what you're up to. They're there to help you achieve your goals. If you've already eaten, they shouldn't let you have a family meal with them. If they just let you get away with this, let them know you don't appreciate it. When you mess up, they should care and think it's a bad thing.

- Get this idea of a "cycle" out of your head. There is no cycle, every day is a new day. Your behavior only looks like some cycle caused by external forces post facto. Everybody trying to form a habit has a similar experience. There is nothing weight loss specific going on here. When you break a good streak everything can go out of the window. This is why it's important to keep the required actions as simple as possible. The more you couple different efforts the harder it's going to be. When you mess up you have to take the loss and move on. A good streak helps but don't dwell on breaking it.

- You have to have a sense of urgency and importance. Those around you should too. "I guess that means I'll get some obesity disease at some point" - just think how ridiculous that sounds. You know you can potentially prevent a serious disease (most people never have that opportunity) but you're just gonna sit there and let some cycle run its course. I mean, come on! How can you wake up everyday and not want to stop it today? What about your children? Perhaps they will inherit your eating habits (stats do show high heritablity), perhaps it would be nice to show them how to keep it under control? You owe it to them. Why doesn't your spouse think that's a huge deal?

It's one of the few things in life where you are literally guaranteed to get a good result if you do what you're supposed to do. There's a lot of advice on how to do this or that, get rich or whatever. But you can never prove these things are guaranteed to work so I can understand if people have a hard time motivating themselves. With weight loss you have a 100% guarantee of success. Personally, I don't see it as some random little problem. Pretty much any effort in life looks like this. You want an outcome, you think of the necessary actions and then the entire ballgame is about actually implementing these actions.

> The thing is, I'm never full. I can eat until I physically can't eat anymore (not something I do regularly, of course), and as soon as my stomach has emptied a bit, I feel fairly hungry again. "Eat until you feel full" is literally a human experience I had never really had.

I feel exactly the way you do, and yes, I was fat, lost it in one go and never gained it back. I try to keep my BMI around 18, never went over 20 since losing weight. It's been around 15 years now. I still have the exact same unlimited appetite I have always had. If you put food I like in front of me, I'm going to eat it.

I don't do anything special. I eat the same foods I always have, just less of them. Got really good at counting and never get more food than I need. The trick is to not put yourself in a situation where you have to stop eating in the first place. It's hard to just stop eating something you want to eat. Not that hard to stop yourself from going to the store again or ordering more (especially when other people are around).

I feel like most people never get serious enough about counting, try to eat stuff they don't enjoy, waste effort on exercise or let other people interfere.

Check it for rust. If you've got structural rust, you can't even assume the original safety characteristics anymore.

That's fine we can call it discipline (which some in the weight loss debate believe is impossible to cultivate, not myself). The point is, if there is indeed a conceptually simple way to increase intelligence, there's no inherent contradiction in most being unable to do it and statistical evidence looking the way it does.

The most trivial brain training exercises routinely show major increases in IQ scores. Those who favor a static hypothesis (and they seem to be the majority) like to simply assume these are fake gains. But without ground truth, there's no way to know. You'd also be surprised how few studies there are that have people directly practice IQ test taking.

The static hypothesis seems to rely on common sense rather than hard evidence. It's rare for somebody who seems dull to become a genius all of a sudden. And when it does happen it's easy to assume they seemed dull rather than actually were.

There is no disconnect between intelligence being simple yet hard to improve. Losing weight is very simple. But it's hard. Obesity is only slightly less heritable than IQ (both highly heritable), yet we know for a fact it's 100% controllable by the individual (unless one is strongly against the notion of free will).

Either the language model would need to know what it's doing or the host program would have to know what the AI is doing. Both seem out of reach. The latter seems more doable since you could hack something up for simple scenarios, but you'd effectively have to match the capabilities of the neural network in a classical way to handle every case (which would render using a neural net moot).

I don't think you'll find much love between the two nowadays. Modern science likes to make allusions to being the "truth" or at least based on some logical rigor. Back in the day you had someone like Godel be buddies with all the greats of science despite his critical takes. I can't imagine this happening today.