“Pattern matching”, “next token prediction”, “tensor math” and “gradient descent” or the understanding and application of these by specialists, are not useful models of what LLMs do, any more than “have sex, feed and talk to the resulting artifact for 18 years” is a useful model of human physiology or psychology.

My understanding, and I'm not a specialist, is there are huge and consequential utility gaps in our models of LLMs. So much so, it is reasonable to say we don't yet understand how they work.

There's no doubt that inherent physical attractiveness, extroversion, confidence and charisma, winning the genetic lottery, creates huge advantages. I've seen it in action, seen it open doors that will never open for an asymmetric, scrawny, socially erratic, introvert nerd with a high-pitched voice like myself. Seen how it affords choices and outcomes I will never have, no matter what I do.

Like so much of life, it is patently unjust and painful.

But it turns out there are enough other doors to open and ways of opening them, that me and all my similarly hopeless friends, got laid, got long-term partners and went on to experience the same relationship joys and failure modes as the beautiful people.

I wish I could warn my young self I'd be the comic relief in the mainstream mating game, but that it's OK, that I should try it, study it, realize and grieve my inadequacy, just get over it, get comfortable with who I am and what's achievable for me, and then go and play a slower, deeper game in a smaller, more congenial league.

What works? Sure, do the self-improvement stuff. Get fit, get a purpose. Find and live by your values. Practice fearlessness and not giving a crap what others think and leverage that into charisma (which is really hard but doable and I'm still working on it).

Find your tribe or social ecosystem and learn to love it (I had contempt for nerds that I had to get past. It was internalized shame at being mainstream-inadequate and it held me back for decades.) Explore adjacent social ecosystems. Get out there. Engage. Do not care about winning any specific outcome.

Have faith that it gets better, almost certainly, in most cases.

Most important, and in accordance with your values and self-respect, work hard to make ongoing net-positive social contributions (all kinds will work, interpersonal, social, material, intellectual, ethical, time, effort) with no strings attached. Use your strengths and interests.

I know this works. I am a weird intense socially-incompetent pedantic misanthrope, but I make the effort to spare my friends the downsides and I deliberately contribute positives. For example, I lead with my values in word and deed. They aren't a matter of consensus and I am contrary to my friends often enough. My super power is that I am entirely comfortable around opposing values. I will not conform to other people's and I don't care, on a personal level, if they adopt mine. Oddly, that combination results in a lot of approval and status. (I didn't plan this, it's a byproduct of my misanthropy, I figured it out after the fact, and realized I was on to a good thing.)

Sometimes at gatherings I can tell that I've just gone off the range, and I can see the look of friendly acceptance. “He's a freak, but he's our freak.” That group judgement of me reflects well and raises my social status considerably. I know my wife is proud of the social respect I've earned, even if the means are a mystery to her. It means nothing at all to some of the ladies, but it does to her.

This is just one example of how a skinny inconsequential nerd can project dominance without cash outlays, big muscles or a strong jawline. And there are so many other ways to be a social contributor or leader. Which equals reproductive fitness!

You, and every other “ugly person” out there has strengths. Find them. Develop them. Use them. Get over the fact that you can't succeed in mainstream competition. Just let it go. Look around, explore, and play in arenas you can win.

Remember (paraphrased) “behind every high-status female is a high-status male who is sick of her shit.” There's no particular advantage in winning one, and there are likely serious downsides. (And before y'all get mad at me the reverse is obviously true too. Married with kids, you know. And I can't speak to the generalization across gender permutations.)

I'm aware that online dating is a problem, and maybe there's a big problem in unrealistic expectations. Going to absolutely support the advice that you get your head out of that world. Play in the real world like it's 1970, 1980, 1990, whatever works for you. I have direct knowledge there are women doing the same. Society will figure it out in the long term and you will be ahead of the curve.

Does the educational system make age-inappropriate almost inhumane demands on children? Yes. Do many kids have attentional and behavioural issues that are not ADHD? Yes. That can be improved without medication? Yes. Should these be distinguished from true AHDH and treated differently? Yes. Should kids with ADHD try non-medical interventions first? Yes. Should kids who have clear diagnostic markers for ADHD and who respond well to stimulants be prescribed stimulants? Hell yes. Should psychiatrists try to weed out non-ADHD people looking for legal sources of speed? Hell yes. Are the fields of psychopathology and psychotherapy problematic? Somewhat. Are they "right"? Not exactly, no. Are they "wrong"? Not exactly, no. Do they relieve suffering? Sometimes, Do they create unintended consequences? Yes, more often than we'd like. Is the whole field conducted in bad faith? No! Is it a work in progress? Yes. Are there bad faith practitioners? Yes. Are there incompetent practitioners? Yes. Are there competent practitioners acting in good faith? Absolutely yes. Are there reliable signals to tell these apart? No, unfortunately, no. Is there an epidemic of performance-enhancing prescription drug abuse in the circles in which you run? Apparently so, and I've heard this elsewhere. Is there performance-enhancing prescription drug abuse in the circles in which I run? Not to my knowledge.

Life is complicated. You do no good by mashing all this stuff together and disposing of it with simplistic conclusions.

You have no clue because you have no pertinent experience.

You are talking about greedy irresponsible doctors and bad faith "patients".

These are not the same thing.

The truth is that lots of people manage their stimulants just fine. Across everyone I know on stimulants, 8 friends and family, none are as you describe. These are people I physically see on a regular basis. No one is addicted, high, unable to function or experiencing material personality changes. Every single one of them is taking low to moderate doses, as needed, and says their stimulants changed their life for the better, allowing them to function "normally" at work or school.

A combination of factors made me decide to stop taking Vyvanse: a probably unrelated health issue, 40 mg of dextroamphetamine per day gave me unpleasant physical side effects, and my realization that 5 years of medication had profoundly hanged my habits and self-concept. I decided to see if I could go off meds and sustain the benefits. So I tapered off (no dependence!) and for 6 years I relied on habits, strategies, exercise, nutrition, healthy sleep and meditation.

But I am now taking Vyvanse again and there is no doubt in my mind now that no amount healthy living can do what Vyvanse does.

My current dosing: on a day to day basis, I take 0, 15, 30 or 45 mg of Vyvanse depending on what I need to get done, my energy level, and my habituation. I drink coffee with it because I find that caffeine helps me keep the Vyvanse dose lower. I organize my life to maximize the value of my medicated time. I experience a clear step function in executive functioning between "slightly too little" and "enough" Vyvanse and I can feel the gears shift in my head, takes about 5 seconds, as the drug concentration in my body crosses the threshold. I move the dose up and down in 15 mg steps and when reducing my dose I spend 2-3 days at each level to avoid crashing. I take Vyvanse holidays two to six weeks long, two or three times a year to reset my habituation and fix any sleep deficit. After a holiday I get the full therapeutic effect from 15 mg. My goal is to generate enough high-quality executive function hours to fulfil my responsibilities, but keep my dose as low as possible. It seems to be working so far.

I have ADHD, my kids have ADHD, and most of my friends have ADHD. Was on Vyvanse for 5 years, then off for 6, and I just recently started taking it again.

During my first 5 years on Vyvance, I had my first experience of what was truly possible for me and it changed my life.

Then, triggered by unrelated health issues and unhappiness with side effects, but mainly because I wanted to run the experiment of seeing how far I could get without meds, I stopped for 6 years. I focussed on lifestyle, structure, health, exercise, nutrition, meditation, And it was pretty good, much better than before my diagnosis.

Then as my life moved forward it presented me with increasing responsibility and challenge, and this year I decided to go back on meds. Of course it was immediately obvious that meds improve my ability to focus and my general self-regulation to a level not possible through healthy lifestyle and earnest intentions. But… now I am getting a multiplier effect of an ADHD-optimized lifestyle working together with my medication, and its amazing. I'm taking a lower dose than before, but sustaining a better level of productivity and mental health.

I have seen in myself and others where meds on their own have failed because their benefits were squandered by a careless lifestyle. Now I get to see how this plays out with my kids, And guess what? Meds are essential, but they aren't enough. The hard part is changing their attitudes, habits and lifestyle.

Individuals within a group want to understand the terms under which a group operates, and in particular, how to extract their share of value from the group and how to maintain or enhance their extractable share of value.

The more the rules of a group are perceived to be fair, reciprocal and consistently applied, the more members can trust the group and each other. This relieves the participants of significant cognitive and emotional burden, allows longer-term collective action, and reduces transaction costs dramatically.

But I can intellectually understand the value of opportunistic defection strategies that "cheat" the group, extracting both an unfair share of value (as well as some of the "embodied trust value" resulting in an incremental loss of trust across the group).

I have a personal commitment to honest behavior and I actively avoid low-trust friends, groups and choices. Possibly because I find low-trust situations too stressful and too much work, as well as morally horrifying (whatever that means!) I'd like to believe that I'm fundamentally a "good person" but I acknowledge that it might just be that I'm unwilling to leave the local maximum high-trust situation I've self-selected for throughout my life, or I'm afraid of the risks of defector strategies (or I'm just cognitively and emotionally unsuited to them).

Here's where it gets complicated, because organizations change, trust levels change, the signifiers of trustability change, organizations lie, sometimes organizations are specifically operated to trap and exploit high-honor people, and people have very different ideas about what constitutes "fairness" or "exploitation". I find it hard to criticize someone who chooses to use low-trust tactics against a low-trust or deceptive group.

It seems, looking around different cultures and organizations in the world that there is a huge variation in the principles under which groups function. Apparently "low-trust" is a viable option, although I'm supportive of the idea that high-trust brings a competitive advantage both for groups and individuals, and is worth building and sustaining.

I'm horrified that we seem to have reached some kind of tipping point in the west where a critical mass of elites (who already extract enormous value!) have decided they can extract even more value through high-order defection than by building modern and durable foundations of trust.

It covers a lot that other commenters have mentioned.

So with respect to the properties you've listed, it is rare for me to find someone where there's a mutual recognition of "sameness".

I'm not in anyone's tribe and so I had to learn to not exhibit tribal signifiers at work, which I think gets me half a point?

  1. Be good at your job (aka write good code).
  2. Be easy to work with.
  3. Be a member of the prevailing tribe.
  4. Be charismatic.

Only one is directly functional, three can be trained to some degree, and one can be faked adequately if absolutely necessary, but it can be mentally unhealthy.

If you don't understand why you aren't getting ahead, apply this model and train a weak category. I came up with this 30 years ago when a young man with limited skills and experience asked me “how do I succeed?” and I was looking for universal factors.

It also shows how to have a great career while being bad at the job. Oh, and it doesn't cover sociopathic strategies. :-)

It should be obvious that your advice is not in any way useful or actionable. And rituals are a very well accepted strategy for dealing with situations which demand consistent and good attention, where human cognitive variability causes problems.

A couple of famous examples--checklists used by aircraft pilots, or the pointing rituals used on Japanese railways.