6 ms·

People love self-study materials, but do you really have the time to absorb it? Yeah you could do this stuff every day after work, but then you may not go to th

by mjfl 4y ago

People love self-study materials, but do you really have the time to absorb it? Yeah you could do this stuff every day after work, but then you may not go to the gym, go on that date, etc... Not enough time in the world to go through even a single learning resource people promote.

susam 4y ago
In 2021, a bunch of us, software engineers, did a small book club to learn analytic number theory from a book written by Apostol. We met for 40 minutes a day and read every line of the book and every page of the book in great detail. We did it for the joy of learning. It took us 79 hours spread across 120 days to complete reading the book and gain a good understanding of analytic number theory along with the analytic proof of the prime number theorem. Meetup log: https://susam.net/maze/meet/iant/log.html https://susam.net/maze/meet/iant/log.html A few blog posts about it: https://susam.net/maze/tag/iant.html https://susam.net/maze/tag/iant.html
- wheelinsupial 4y ago
  I skimmed your blog posts and logs, so apologies if I missed it. How many problems or exercises did you work through in the book? Did people put in any additional time individually between sessions?
  - susam 4y ago
    I can't speak for all participants of these meetings and I haven't kept an exact count of the problems I solved but I believe I must have solved somewhere between 160 to 200 problems from the exercises in the book. See also https://news.ycombinator.com/item?id=34908191 https://news.ycombinator.com/item?id=34908191 for more details on this topic.
- FiberBundle 4y ago
  Sorry, but you don't properly learn any math topic in 80 hours. I've spent more than 10 hours on some problem sets in difficult courses. Solving hard problems takes a lot of time and is absolutely essential to learn math. When you read a book or a script you feel as if you have grasped the content, but most of the time you really haven't. It's only when you attack problems from different angles that you really build intuition for the concepts taught in that book.
  - susam 4y ago
    The time recorded in the meeting log includes only the time spent together in reading the book and working through the proofs. There are some more details about our reading style here: https://susam.net/maze/journey-to-prime-number-theorem.html https://susam.net/maze/journey-to-prime-number-theorem.html I have kept an archive of all our meeting notes here: https://offbeat.cc/iant/boards/ https://offbeat.cc/iant/boards/ I very much agree that attacking problems from different angles is very important in building intuition for the concepts taught in the book. However, that's something that we did not do within the 40 minute meetings. It wouldn't make sense too because I believe solving problems is very much a personal journey where different people need different amount of time to solve problems. I believe solving problems is best done on our own time. Sometimes though we did discuss the solutions of some problems in the meetings just to take a break from the theorem-and-proof style of meetings. I solved most of the problems in the book in my own time. I know another participant who did too. Of course, that took a lot more than 80 hours. I must have spent an additional 2 to 3 hours everyday for solving the problems.
nohaydeprobleme 4y ago
It's not so hard to find time. Many people with full-time jobs practice martial arts an hour a day for three to five times a week, while others practice a musical instrument for half an hour to an hour a day. Other people also complete part-time Master's degrees in highly technical subjects while working full-time (many programs like these are designed for full-time workers). I've personally enjoyed language learning as a hobby, and found time in my commutes to listen to audio programs (I enjoy this more than anything else I could do on a commute in a crowded train, so it's no loss of time). For more casual gym-goers like myself, oftentimes running on a treadmill doesn't require full attention, so it's possible to watch video lectures or listen to more audio programs. Neither of these are as ideal as concentrated study in a quiet room with a desk, but they're relatively lower-effort ways to get useful practice with a specific subject (so it's easier to practice consistently). I then find more time on the weekends for more dedicated study. If you're dating someone who also shares your interests in studying, they can also spend time studying alongside you (alongside more fun and relaxed dates). To quantify this, that is about 13 hours a week of non-concentrated study (assuming a 1-hour commute each way every weekday for 10 hours, plus another 3 hours assuming studies during a 1-hour workout three times a week, at the lower end of how often you can consistently exercise). Add in another 1 hour of concentrated study a day (or if you'd like, 3 hours of extended study sessions each weekend day, plus additional scattered study during the week), and you hit 20 hours a week. You may not be as fast as the writer of the submitted article with this, but you can at least get pretty far from consistent practice over long periods of time.
- foo42 4y ago
  it sounds like you have a good attitude for making smart choices about how you use your discretionary time. I feel I should just point out however that generalisations about how much discretionary time people have don't always apply as widely as you may believe. There's probably a certain bias to the "average" person you meet at, for example, marital arts clubs, as they're the people with that time. In other circumstances or phases of life people count their discretionary time in minutes per day, and that is probably right at the end of the day. Not criticising, just hoping to widen your perspective a touch.
paulpauper 4y ago
If you're smart enough you can retain and master material very fast. People like Ed Witten and Terrance Tao probably mastered year's worth of material in months . IQ does not scale linearly, meaning going from 130 to 166 is not a 20% gains more more like multiple orders of magnitudes. It's hard to appreciate or comprehend what truly smart people are capable of doing. It's like magic, to put it bluntly.
- mjfl 4y ago
  Can you tell a field that either Ed Witten or Terrance Tao knows a significant amount about outside their field of specialization? And don't say math for Ed Witten because his field is mathematical physics.
  - paulpauper 4y ago
    Ed Witten mastered bridge on a cruse ship , winning a tournament despite knowing nothing about it https://www.baltimoresun.com/news/bs-xpm-1995-02-12-1995043206-story.html https://www.baltimoresun.com/news/bs-xpm-1995-02-12-19950432... But that is not the point. My point is super-smart people can pick up and master concepts very fast.
- whatshisface 4y ago
  >going from 130 to 166 is not a 20% gains more more like multiple orders of magnitudes. I have always wondered what the truth is about this but I have never seen any real answers. On the "linear" side you have median incomes by IQ percentile. On the "not even quantitatively comparable" side you have the fact that your score on an IQ test will not increase that much if you are allowed much more than the standard time to finish it, indicating that there is a wall (defined by your ability) that you can hit. On the "sublinear" side, there's the fact that a lot of moderately smart people have made major scientific discoveries, which you wouldn't think could ever happen when there are 10,000 people with IQ > 160 in the US today - easily more than enough to take all of the places in the history books written about this decade. I do not think there is any clear-cut answer.
  - 4y ago
antegamisou 4y ago
Yeah and my observation is that the self-study proponents very often heavily overrate their own intelligence/learning ability.
wayeq 4y ago
not to mention some of us have tiny little cloned versions of ourself that are doing their best to perish for 20 hours a day

by mjfl 4y ago

susam 4y ago

In 2021, a bunch of us, software engineers, did a small book club to learn analytic number theory from a book written by Apostol. We met for 40 minutes a day and read every line of the book and every page of the book in great detail. We did it for the joy of learning. It took us 79 hours spread across 120 days to complete reading the book and gain a good understanding of analytic number theory along with the analytic proof of the prime number theorem. Meetup log: https://susam.net/maze/meet/iant/log.html https://susam.net/maze/meet/iant/log.html A few blog posts about it: https://susam.net/maze/tag/iant.html https://susam.net/maze/tag/iant.html

wheelinsupial 4y ago
I skimmed your blog posts and logs, so apologies if I missed it. How many problems or exercises did you work through in the book? Did people put in any additional time individually between sessions?
- susam 4y ago
  I can't speak for all participants of these meetings and I haven't kept an exact count of the problems I solved but I believe I must have solved somewhere between 160 to 200 problems from the exercises in the book. See also https://news.ycombinator.com/item?id=34908191 https://news.ycombinator.com/item?id=34908191 for more details on this topic.
FiberBundle 4y ago
Sorry, but you don't properly learn any math topic in 80 hours. I've spent more than 10 hours on some problem sets in difficult courses. Solving hard problems takes a lot of time and is absolutely essential to learn math. When you read a book or a script you feel as if you have grasped the content, but most of the time you really haven't. It's only when you attack problems from different angles that you really build intuition for the concepts taught in that book.
- susam 4y ago
  The time recorded in the meeting log includes only the time spent together in reading the book and working through the proofs. There are some more details about our reading style here: https://susam.net/maze/journey-to-prime-number-theorem.html https://susam.net/maze/journey-to-prime-number-theorem.html I have kept an archive of all our meeting notes here: https://offbeat.cc/iant/boards/ https://offbeat.cc/iant/boards/ I very much agree that attacking problems from different angles is very important in building intuition for the concepts taught in the book. However, that's something that we did not do within the 40 minute meetings. It wouldn't make sense too because I believe solving problems is very much a personal journey where different people need different amount of time to solve problems. I believe solving problems is best done on our own time. Sometimes though we did discuss the solutions of some problems in the meetings just to take a break from the theorem-and-proof style of meetings. I solved most of the problems in the book in my own time. I know another participant who did too. Of course, that took a lot more than 80 hours. I must have spent an additional 2 to 3 hours everyday for solving the problems.

nohaydeprobleme 4y ago

It's not so hard to find time. Many people with full-time jobs practice martial arts an hour a day for three to five times a week, while others practice a musical instrument for half an hour to an hour a day. Other people also complete part-time Master's degrees in highly technical subjects while working full-time (many programs like these are designed for full-time workers). I've personally enjoyed language learning as a hobby, and found time in my commutes to listen to audio programs (I enjoy this more than anything else I could do on a commute in a crowded train, so it's no loss of time). For more casual gym-goers like myself, oftentimes running on a treadmill doesn't require full attention, so it's possible to watch video lectures or listen to more audio programs. Neither of these are as ideal as concentrated study in a quiet room with a desk, but they're relatively lower-effort ways to get useful practice with a specific subject (so it's easier to practice consistently). I then find more time on the weekends for more dedicated study. If you're dating someone who also shares your interests in studying, they can also spend time studying alongside you (alongside more fun and relaxed dates). To quantify this, that is about 13 hours a week of non-concentrated study (assuming a 1-hour commute each way every weekday for 10 hours, plus another 3 hours assuming studies during a 1-hour workout three times a week, at the lower end of how often you can consistently exercise). Add in another 1 hour of concentrated study a day (or if you'd like, 3 hours of extended study sessions each weekend day, plus additional scattered study during the week), and you hit 20 hours a week. You may not be as fast as the writer of the submitted article with this, but you can at least get pretty far from consistent practice over long periods of time.

foo42 4y ago
it sounds like you have a good attitude for making smart choices about how you use your discretionary time. I feel I should just point out however that generalisations about how much discretionary time people have don't always apply as widely as you may believe. There's probably a certain bias to the "average" person you meet at, for example, marital arts clubs, as they're the people with that time. In other circumstances or phases of life people count their discretionary time in minutes per day, and that is probably right at the end of the day. Not criticising, just hoping to widen your perspective a touch.

paulpauper 4y ago

If you're smart enough you can retain and master material very fast. People like Ed Witten and Terrance Tao probably mastered year's worth of material in months . IQ does not scale linearly, meaning going from 130 to 166 is not a 20% gains more more like multiple orders of magnitudes. It's hard to appreciate or comprehend what truly smart people are capable of doing. It's like magic, to put it bluntly.

mjfl 4y ago
Can you tell a field that either Ed Witten or Terrance Tao knows a significant amount about outside their field of specialization? And don't say math for Ed Witten because his field is mathematical physics.
- paulpauper 4y ago
  Ed Witten mastered bridge on a cruse ship , winning a tournament despite knowing nothing about it https://www.baltimoresun.com/news/bs-xpm-1995-02-12-1995043206-story.html https://www.baltimoresun.com/news/bs-xpm-1995-02-12-19950432... But that is not the point. My point is super-smart people can pick up and master concepts very fast.
whatshisface 4y ago
>going from 130 to 166 is not a 20% gains more more like multiple orders of magnitudes. I have always wondered what the truth is about this but I have never seen any real answers. On the "linear" side you have median incomes by IQ percentile. On the "not even quantitatively comparable" side you have the fact that your score on an IQ test will not increase that much if you are allowed much more than the standard time to finish it, indicating that there is a wall (defined by your ability) that you can hit. On the "sublinear" side, there's the fact that a lot of moderately smart people have made major scientific discoveries, which you wouldn't think could ever happen when there are 10,000 people with IQ > 160 in the US today - easily more than enough to take all of the places in the history books written about this decade. I do not think there is any clear-cut answer.
- 4y ago

antegamisou 4y ago

Yeah and my observation is that the self-study proponents very often heavily overrate their own intelligence/learning ability.

wayeq 4y ago

not to mention some of us have tiny little cloned versions of ourself that are doing their best to perish for 20 hours a day