I think that lots of advice given is bad - it isn’t practical, it isn’t insightful, and it is often something that is amazingly obvious to the person who is receiving it. Take, for example, these two pieces of advice from Sam Altman (someone who I admire and think can be very insightful - his
Many folks who give out what is labeled "bad advice" are unaware that their success is mostly due to randomness. Thats why that sort of advice lacks true insight - the giver did not become successful because of something they did, they just happened to be standing in the right spot when the bullets went flying. Its also why the advice seems super obvious - because the giver cant tell that their success is just a random selection and does not come directly from their own choices. There advice amounts to, "just stand there and hope you dont get taken out, thats what I did!"
More subtly, some of the advice doesn't fall into the "random" category, so much as "necessary, but not sufficient." It probably is true, that if successful person didn't do X, then they wouldn't have achieved success. But also probably true that many people do X and aren't successful. Hard work is the probably the classic example. But that doesn't mean you don't need to care about something like hard work, it just isn't sufficient. There is other secret sauce - some random - but also some other factor that just are difficult to articulate from the inside position of a participant.
If this was true, the world's successful people would be a random sampling of the population. But they're not. Clearly certain groups are more represented than others on the Forbes Rich List, for example.
Reminds me of this paragraph from Scott Alexander's "You don't hate polyamory, you hate people who write books."
> Advice is disproportionately written by defective people. Healthy people perform naturally and effortlessly. You walk so gracefully that a million man-hours into bipedal robots fail to match your skill. But if some stroke patient or precocious one-year-old asked your secret, you would just say “I put one foot in front of the other.”
If you want good advice about how to walk, ask someone with cerebral palsy. They experience walking as a constant battle to overcome their natural constitution, and so accumulate tips and tricks throughout their lives. Or ask a physical therapist who works with these people and studies them. Just don’t ask someone you see walking especially briskly down the street.
Relationships work the same way. Go to an elderly couple who have been happily married for fifty years, and they’ll give you vapid old-person advice like “Treat every day as a gift from God.” But go to someone who’s struggled with every one of their last thirty-seven relationships, and they’ll be full of suggestions! They’ll tell you all sorts of fascinating things about boundaries and gaslighting and the four-hundred-and-ninety-four principles of nonviolent communication.
A major, and hopefully relevant, gripe I have a long had about public advice is that it is too rarely conditional. I do think there are situations where “work harder“ it’s good advice, we could debate whether it is 1%/10%/80%, but the more productive framing would be “what kind of person would benefit from this, now?”
Maybe one heuristic is “If you have had an early success, but recently a failure, and this goal is not dominated by noise, you should *try* working harder for *one week* and evaluate the outcome”. As an example I finally got a personal trainer after believing my energy level was insensitive to exercise - turns out I just wasn’t doing enough of it.
Being human, I had a strong prior that exercise should be helpful. The cause and effect relationship is not nearly as noisy as entrepreneurship, and the reason I was not trying harder was pretty clear: it’s not fun and it is too easy to turn off a YouTube video.
But this is already way too complex to get across in a verbal public interview. Anything more complex than a single if/then statement would also get pretty dry for a written article. I don’t really have a solution to the problem that the algorithmic complexity requires to give good advice might simply overflow the capacity of most channels through which we receive it
Broadly agree with all of this, but I would say (and I'm not even sure we're disagreeing on this point) that receiving advice from a PT about the amount of time someone should exercise and the intensity with which they should exercise seems categorically different from generic advice to 'work harder'. The reason I think 'work harder' is generally bad advice is because it is obvious that putting in more effort will generally lead to better results in the pursuit of some goal - but the amount of hours and the intensity you should put into exercising isn't obvious, and similarly any advice about reaching a specific goal that is 'there is some threshold here that you need to reach before you start making serious progress that you are unaware of' seems very different to 'you have to work harder'. What you refer to as a 'productive framing' of the advice 'work harder' seems to me to just lead to different (and better) advice altogether!
The most important bit of context is almost certainly the receivers prior beliefs (especially the ones that are not consciously acknowledging!)
“Be hard to compete with” would have actually been a revelation for me when graduating college. Competing in undifferentiated groups is the water you are swimming in throughout all of schooling, and then you are released with the vague promise that you are smart, and smart people are successful. But if someone had really pushed me (preferably live, so I could not make excuses - much like exercising to YouTube) to consider my unique value proposition as an employee, I think it would have opened my eyes.
What I can say for sure is that my career took off about a year later when I saw an HBR cover article “Data Science: Sexiest job of the 21st century”. I’d somehow developed enough sense at that point to realize that (at the time) nobody else had more data science credentials than I did (none), so if I took a run at building a portfolio I would be in a good supply/demand position. This few minutes of insight has defined the last 10 years of my career - I imagine a good coach could’ve gotten me there at least a few months earlier
An example I often think about is whether to “take more risks“. If I had to take a generalized stance it would be “pro“ - there is even a cool study to support this! https://www.nber.org/papers/w22487
But I, and I suspect everyone, has an acquaintance to whom you would strongly suggest taking fewer risks.
Moreover, if it could be conditional on the category of opportunity/risk I could give even better advice. I’d advise more friends to switch jobs than start companies, for example. That’s a function of both the person and opportunity.
If I could see their financial situation I’d be even more confident, etc. But again that would make for a bewildering TED talk, so it never happens.
Actually I just made a connection: the article “the unreasonable effectiveness of data“ argues that ML’s success in NLP comes less from finding elegant principles and more from being able to see every little corner case of language use. But each edge case is more parameters, and there is simply no elegant compression - which will never fit in a witty interview answer. I hate it when it venture capitalists say “pattern matching“, but maybe the truly best they can do is have 10,000 highly specific rules of thumb for success
I view a lot of the advice as more bland, generic and not useful rather than actively bad (which seems more like a terminology difference than a substantive disagreement). People are often incentivized to write it because they can get attention and rewards for giving advice, because they're famous and popular enough. Or because e.g. giving advice for other CEOs may convince investors that they know how to be a CEO. So I think the main reason they write bland and generic advice is because they don't have any special insight to share, but they want to write advice anyway. Similarly, a lot of mediocre books exist because people have reasons that they want to be an author other than having something novel and important to say.
When people give advice publicly, it also suffers from social desirability bias. People say stuff that sounds appealing. I think this is particularly the case with dating advice. People are often unwilling to speak candidly especially if people are hostile.
"It’s also very unclear that it’s even possible to take this advice - conscientiousness is almost certainly partially genetic, so I’m not sure you can just jump up to the 99th percentile in hard-work just by willing it" Why not? If you truly have free will, you can change your behavior willingly. Similarly, despite your genes, you can be in an environment so outside the norm that the environmental effect dominates. Heritability is the proportion of variation attributable to genetic differences. All people are how they are because of genes and environment. You cannot have one without the other. Imagine a rectangle. No width means no rectangle. No length means no rectangle. Heritability is SOMEWHAT like comparing all rectangles. If most rectangles are wide but short in length, then you could have a low area if you are short in width and short in length. However, if you believe you can expand length willingly, you should be able to expand length so far that you achieve 99% percentile in area.
I think 'we truly have free will' is already an assumption I'm not comfortable making, but even assuming that it is true, I don't think it necessitates that a person can change their work ethic just through will-power. I see work ethic as more comparable to a talent - if someone has an extremely poor work ethic it is extremely unlikely that they will ever jump to the 99th percentile in work ethic even if they really want to be conscientious, just like someone who is an awful musician is very unlikely to ever become an incredibly good musician just through willpower.
First, I'm pretty sure there is a limited set of "stuff" that is valuable to success, so it's basically impossible to come up with truly novel general advice. That's why I'm not so critical of whether advice is insightful or not: It might not seem so for you, but it could be for me because I either hadn't heard of it before or (far more likely) because I didn't think to apply it to this situation. Take your example of one-on-ones being useful. To me that's not insightful, it's just a different flavor of Rubber Duck Debugging or grandma's constant reminder that sometimes the best way to solve a problem is to talk through it.
I look at general advice as throwing stuff at a wall and seeing if anything sticks: Something will work for me and something else will work for you.
But general advice is only the tip of the iceberg, most advice has to be specific to the person and the problem to be useful. And you'll never get that from a book or Youtube video or blog post. I think this is closely related to your concern that advice needs to be actionable. Specific advice is still going to be related to one of the set of useful "things," but it's going to give you an idea of how to apply that "thing" to your specific circumstances.
Really liked this, one thing I would add (that may be implicit) in giving advice: Make sure it's specific to a situation/person, and not broad (e.g. "How to be successful").
So for the grad school example, this would be how to help someone who is feeling unproductive in grad school (and may not be helpful for anyone who is feeling unproductive);
For Anki, it is probably more effective when tailored specifically in how to effective remember knowledge (and not for example just "being more productive").
(The irony that I'm reading your blog instead of focusing on my grad school research is killing me.)
Many folks who give out what is labeled "bad advice" are unaware that their success is mostly due to randomness. Thats why that sort of advice lacks true insight - the giver did not become successful because of something they did, they just happened to be standing in the right spot when the bullets went flying. Its also why the advice seems super obvious - because the giver cant tell that their success is just a random selection and does not come directly from their own choices. There advice amounts to, "just stand there and hope you dont get taken out, thats what I did!"
More subtly, some of the advice doesn't fall into the "random" category, so much as "necessary, but not sufficient." It probably is true, that if successful person didn't do X, then they wouldn't have achieved success. But also probably true that many people do X and aren't successful. Hard work is the probably the classic example. But that doesn't mean you don't need to care about something like hard work, it just isn't sufficient. There is other secret sauce - some random - but also some other factor that just are difficult to articulate from the inside position of a participant.
If this was true, the world's successful people would be a random sampling of the population. But they're not. Clearly certain groups are more represented than others on the Forbes Rich List, for example.
Reminds me of this paragraph from Scott Alexander's "You don't hate polyamory, you hate people who write books."
> Advice is disproportionately written by defective people. Healthy people perform naturally and effortlessly. You walk so gracefully that a million man-hours into bipedal robots fail to match your skill. But if some stroke patient or precocious one-year-old asked your secret, you would just say “I put one foot in front of the other.”
If you want good advice about how to walk, ask someone with cerebral palsy. They experience walking as a constant battle to overcome their natural constitution, and so accumulate tips and tricks throughout their lives. Or ask a physical therapist who works with these people and studies them. Just don’t ask someone you see walking especially briskly down the street.
Relationships work the same way. Go to an elderly couple who have been happily married for fifty years, and they’ll give you vapid old-person advice like “Treat every day as a gift from God.” But go to someone who’s struggled with every one of their last thirty-seven relationships, and they’ll be full of suggestions! They’ll tell you all sorts of fascinating things about boundaries and gaslighting and the four-hundred-and-ninety-four principles of nonviolent communication.
I liked this a lot! Just published something inspired by it here: https://forge.medium.com/theres-no-such-thing-as-good-advice-bd5dbaf9d450
A major, and hopefully relevant, gripe I have a long had about public advice is that it is too rarely conditional. I do think there are situations where “work harder“ it’s good advice, we could debate whether it is 1%/10%/80%, but the more productive framing would be “what kind of person would benefit from this, now?”
Maybe one heuristic is “If you have had an early success, but recently a failure, and this goal is not dominated by noise, you should *try* working harder for *one week* and evaluate the outcome”. As an example I finally got a personal trainer after believing my energy level was insensitive to exercise - turns out I just wasn’t doing enough of it.
Being human, I had a strong prior that exercise should be helpful. The cause and effect relationship is not nearly as noisy as entrepreneurship, and the reason I was not trying harder was pretty clear: it’s not fun and it is too easy to turn off a YouTube video.
But this is already way too complex to get across in a verbal public interview. Anything more complex than a single if/then statement would also get pretty dry for a written article. I don’t really have a solution to the problem that the algorithmic complexity requires to give good advice might simply overflow the capacity of most channels through which we receive it
Broadly agree with all of this, but I would say (and I'm not even sure we're disagreeing on this point) that receiving advice from a PT about the amount of time someone should exercise and the intensity with which they should exercise seems categorically different from generic advice to 'work harder'. The reason I think 'work harder' is generally bad advice is because it is obvious that putting in more effort will generally lead to better results in the pursuit of some goal - but the amount of hours and the intensity you should put into exercising isn't obvious, and similarly any advice about reaching a specific goal that is 'there is some threshold here that you need to reach before you start making serious progress that you are unaware of' seems very different to 'you have to work harder'. What you refer to as a 'productive framing' of the advice 'work harder' seems to me to just lead to different (and better) advice altogether!
The most important bit of context is almost certainly the receivers prior beliefs (especially the ones that are not consciously acknowledging!)
“Be hard to compete with” would have actually been a revelation for me when graduating college. Competing in undifferentiated groups is the water you are swimming in throughout all of schooling, and then you are released with the vague promise that you are smart, and smart people are successful. But if someone had really pushed me (preferably live, so I could not make excuses - much like exercising to YouTube) to consider my unique value proposition as an employee, I think it would have opened my eyes.
What I can say for sure is that my career took off about a year later when I saw an HBR cover article “Data Science: Sexiest job of the 21st century”. I’d somehow developed enough sense at that point to realize that (at the time) nobody else had more data science credentials than I did (none), so if I took a run at building a portfolio I would be in a good supply/demand position. This few minutes of insight has defined the last 10 years of my career - I imagine a good coach could’ve gotten me there at least a few months earlier
An example I often think about is whether to “take more risks“. If I had to take a generalized stance it would be “pro“ - there is even a cool study to support this! https://www.nber.org/papers/w22487
But I, and I suspect everyone, has an acquaintance to whom you would strongly suggest taking fewer risks.
Moreover, if it could be conditional on the category of opportunity/risk I could give even better advice. I’d advise more friends to switch jobs than start companies, for example. That’s a function of both the person and opportunity.
If I could see their financial situation I’d be even more confident, etc. But again that would make for a bewildering TED talk, so it never happens.
Actually I just made a connection: the article “the unreasonable effectiveness of data“ argues that ML’s success in NLP comes less from finding elegant principles and more from being able to see every little corner case of language use. But each edge case is more parameters, and there is simply no elegant compression - which will never fit in a witty interview answer. I hate it when it venture capitalists say “pattern matching“, but maybe the truly best they can do is have 10,000 highly specific rules of thumb for success
I view a lot of the advice as more bland, generic and not useful rather than actively bad (which seems more like a terminology difference than a substantive disagreement). People are often incentivized to write it because they can get attention and rewards for giving advice, because they're famous and popular enough. Or because e.g. giving advice for other CEOs may convince investors that they know how to be a CEO. So I think the main reason they write bland and generic advice is because they don't have any special insight to share, but they want to write advice anyway. Similarly, a lot of mediocre books exist because people have reasons that they want to be an author other than having something novel and important to say.
People are genuinely wising up that maybe, advice are social strata specific ("Clear Pill" thinking):
- Platitudes don't work for those who are average or below average https://www.youtube.com/watch?v=SiuDfKLPTtU
- Genuine advice are good, but often provided for the wrong reason https://www.youtube.com/watch?v=SiuDfKLPTtU
When people give advice publicly, it also suffers from social desirability bias. People say stuff that sounds appealing. I think this is particularly the case with dating advice. People are often unwilling to speak candidly especially if people are hostile.
"It’s also very unclear that it’s even possible to take this advice - conscientiousness is almost certainly partially genetic, so I’m not sure you can just jump up to the 99th percentile in hard-work just by willing it" Why not? If you truly have free will, you can change your behavior willingly. Similarly, despite your genes, you can be in an environment so outside the norm that the environmental effect dominates. Heritability is the proportion of variation attributable to genetic differences. All people are how they are because of genes and environment. You cannot have one without the other. Imagine a rectangle. No width means no rectangle. No length means no rectangle. Heritability is SOMEWHAT like comparing all rectangles. If most rectangles are wide but short in length, then you could have a low area if you are short in width and short in length. However, if you believe you can expand length willingly, you should be able to expand length so far that you achieve 99% percentile in area.
I think 'we truly have free will' is already an assumption I'm not comfortable making, but even assuming that it is true, I don't think it necessitates that a person can change their work ethic just through will-power. I see work ethic as more comparable to a talent - if someone has an extremely poor work ethic it is extremely unlikely that they will ever jump to the 99th percentile in work ethic even if they really want to be conscientious, just like someone who is an awful musician is very unlikely to ever become an incredibly good musician just through willpower.
Also, what do you think about this article on IQ vs perseverance? https://freddiedeboer.substack.com/p/grit-or-the-moralists-fable-about-education
I suppose if we don't have free will, then the heritability estimates don't matter. We are determined either way! ha
I think there are a few layers to advice:
First, I'm pretty sure there is a limited set of "stuff" that is valuable to success, so it's basically impossible to come up with truly novel general advice. That's why I'm not so critical of whether advice is insightful or not: It might not seem so for you, but it could be for me because I either hadn't heard of it before or (far more likely) because I didn't think to apply it to this situation. Take your example of one-on-ones being useful. To me that's not insightful, it's just a different flavor of Rubber Duck Debugging or grandma's constant reminder that sometimes the best way to solve a problem is to talk through it.
I look at general advice as throwing stuff at a wall and seeing if anything sticks: Something will work for me and something else will work for you.
But general advice is only the tip of the iceberg, most advice has to be specific to the person and the problem to be useful. And you'll never get that from a book or Youtube video or blog post. I think this is closely related to your concern that advice needs to be actionable. Specific advice is still going to be related to one of the set of useful "things," but it's going to give you an idea of how to apply that "thing" to your specific circumstances.
Really liked this, one thing I would add (that may be implicit) in giving advice: Make sure it's specific to a situation/person, and not broad (e.g. "How to be successful").
So for the grad school example, this would be how to help someone who is feeling unproductive in grad school (and may not be helpful for anyone who is feeling unproductive);
For Anki, it is probably more effective when tailored specifically in how to effective remember knowledge (and not for example just "being more productive").
(The irony that I'm reading your blog instead of focusing on my grad school research is killing me.)
Depends on what you define as success, folks