5 ms·

I never took his rule too literally. I always thought of it as an explanation for how people become the best at something: practice. I still believe it is true.

by Oculus 12y ago

I never took his rule too literally. I always thought of it as an explanation for how people become the best at something: practice. I still believe it is true. Though you can't study entrepreneurship, practicing (i.e. gaining experience) makes you better at running companies. The same can be applied to other activities which don't involve rigid structure which can be learned.

a_bonobo 12y ago
Your definition of 'practicing' is different from the publication's - the publication specifically looks only at deliberate practice: >This view holds that expert performance largely reflects accumulated amount of deliberate practice, which Ericsson et al. defined as engagement in structured activities created specifically to improve performance in a domain. As an entrepreneur or programmer, I think that it's very hard to do deliberate practice - there are few structured activities you can do to improve your skills, as both fields, even in their more basic tasks, incorporate various skills and approaches. For example, you wouldn't say 'today I practice for-loops' and then do nothing but write for-loops all day. You can say 'today I write a program that uses while loops instead of for-loops', but I think that wouldn't count as deliberate practice anymore - the structure of the task changes with the focus and task of your program (you'd also have to do more like de-bugging, tests etc.) As a tennis player, however, you can say 'today I practice serves' and then do nothing but serve balls all day, that's a structured task that doesn't change. Apparently it just doesn't help that much.
- sousousou 12y ago
  I'm coming from music as a profession to programming as a profession. In music you can craft repetitive exercises to improve small aspects of your skillset, but in computer science you can't do that exactly (or sanely). However, I do find myself directing my study in a similar way. I have various high-level goals and I try to find some way to get at them on a daily basis. I use concrete projects like reading a book, doing exercises, or taking on a hobby project. It feels very similar to practicing music, but I wouldn't call it practicing. It's my practice!
  - nickff 12y ago
    Thanks for your insight, I really appreciate hearing comments from people who weren't born coding, because these often contain very interesting insights! I would suggest that revisiting code which you have previously written, and trying to re-factor it, or at least considering how this could be done (and its impacts), is a very useful tool to help you 'practice' coding. I know that this is imperfect, but I have personally learned a lot from applying new lessons (from books, experience, or code reviews) to my own prior work.
    - Dewie 12y ago
      > Thanks for your insight, I really appreciate hearing comments from people who weren't born coding, because these often contain very interesting insights! How nice that we outsiders can be of some use.
  - jmalicki 12y ago
    Actually, you can. Sites like topcoder and Project Euler are actually very good at building skills... they're much like playing arpeggios or similar exercises... they're small, may take 30 minutes to two hours to complete, and the more you do the more used to solving problems you get, and you also find out what you need to improve upon.
x1798DE 12y ago
The article is about the degree to which deliberate practice helps you in various fields, it's not actually about the specific 10,000 hour rule. The thesis of the article is that, essentially, deliberate practice does not have a large effect on your effectiveness in fields which do not have unchanging, rigid structures (like games whose rules don't change). As a rule, I reserve judgement about any "new study", but also as a rule I think that anything Malcolm Gladwell says is bullshit, so I'm torn on this one.
- pjscott 12y ago
  I had the same torn reaction, but upon reading the full text (thanks, Gwern!) I'm no longer conflicted: it's not really a new study, so much as it is a systematic review of roughly a decade of relevant literature. Malcolm Gladwell seems to have based his claims mainly on a widely-cited Hot New Study from 1993, which has since turned out to have been overblown: later studies on the importance of deliberate practice have almost all come up with lower values for the correlation between deliberate practice hours and performance. So, it looks like ignoring Malcolm Gladwell was the right answer here. It usually is.
  - gwern 12y ago
    You're welcome. > later studies on the importance of deliberate practice have almost all come up with lower values for the correlation between deliberate practice hours and performance. What's also really interesting is the differences in the kind of data used to measure deliberate practice: daily logs, which you'd think be the most accurate, showed the least correlation with performance, while retrospective interviews showed the most. Biases in recollection or researcher allegiance?
- Hasknewbie 12y ago
  I believe he goes by the name Explaino The Clown these days.
- PM_Tech 12y ago
  I am wary of anyone who makes blanket statements like "anything Malcolm Gladwell says is bullshit." You simulataneously discount his entire canon of work and also every single underlying scientific paper he has quoted, cited or referenced. That is a lot of work to label bullshit.

by Oculus 12y ago

a_bonobo 12y ago

Your definition of 'practicing' is different from the publication's - the publication specifically looks only at deliberate practice: >This view holds that expert performance largely reflects accumulated amount of deliberate practice, which Ericsson et al. defined as engagement in structured activities created specifically to improve performance in a domain. As an entrepreneur or programmer, I think that it's very hard to do deliberate practice - there are few structured activities you can do to improve your skills, as both fields, even in their more basic tasks, incorporate various skills and approaches. For example, you wouldn't say 'today I practice for-loops' and then do nothing but write for-loops all day. You can say 'today I write a program that uses while loops instead of for-loops', but I think that wouldn't count as deliberate practice anymore - the structure of the task changes with the focus and task of your program (you'd also have to do more like de-bugging, tests etc.) As a tennis player, however, you can say 'today I practice serves' and then do nothing but serve balls all day, that's a structured task that doesn't change. Apparently it just doesn't help that much.

sousousou 12y ago
I'm coming from music as a profession to programming as a profession. In music you can craft repetitive exercises to improve small aspects of your skillset, but in computer science you can't do that exactly (or sanely). However, I do find myself directing my study in a similar way. I have various high-level goals and I try to find some way to get at them on a daily basis. I use concrete projects like reading a book, doing exercises, or taking on a hobby project. It feels very similar to practicing music, but I wouldn't call it practicing. It's my practice!
- nickff 12y ago
  Thanks for your insight, I really appreciate hearing comments from people who weren't born coding, because these often contain very interesting insights! I would suggest that revisiting code which you have previously written, and trying to re-factor it, or at least considering how this could be done (and its impacts), is a very useful tool to help you 'practice' coding. I know that this is imperfect, but I have personally learned a lot from applying new lessons (from books, experience, or code reviews) to my own prior work.
  - Dewie 12y ago
    > Thanks for your insight, I really appreciate hearing comments from people who weren't born coding, because these often contain very interesting insights! How nice that we outsiders can be of some use.
- jmalicki 12y ago
  Actually, you can. Sites like topcoder and Project Euler are actually very good at building skills... they're much like playing arpeggios or similar exercises... they're small, may take 30 minutes to two hours to complete, and the more you do the more used to solving problems you get, and you also find out what you need to improve upon.

x1798DE 12y ago

The article is about the degree to which deliberate practice helps you in various fields, it's not actually about the specific 10,000 hour rule. The thesis of the article is that, essentially, deliberate practice does not have a large effect on your effectiveness in fields which do not have unchanging, rigid structures (like games whose rules don't change). As a rule, I reserve judgement about any "new study", but also as a rule I think that anything Malcolm Gladwell says is bullshit, so I'm torn on this one.

pjscott 12y ago
I had the same torn reaction, but upon reading the full text (thanks, Gwern!) I'm no longer conflicted: it's not really a new study, so much as it is a systematic review of roughly a decade of relevant literature. Malcolm Gladwell seems to have based his claims mainly on a widely-cited Hot New Study from 1993, which has since turned out to have been overblown: later studies on the importance of deliberate practice have almost all come up with lower values for the correlation between deliberate practice hours and performance. So, it looks like ignoring Malcolm Gladwell was the right answer here. It usually is.
- gwern 12y ago
  You're welcome. > later studies on the importance of deliberate practice have almost all come up with lower values for the correlation between deliberate practice hours and performance. What's also really interesting is the differences in the kind of data used to measure deliberate practice: daily logs, which you'd think be the most accurate, showed the least correlation with performance, while retrospective interviews showed the most. Biases in recollection or researcher allegiance?
Hasknewbie 12y ago
I believe he goes by the name Explaino The Clown these days.
PM_Tech 12y ago
I am wary of anyone who makes blanket statements like "anything Malcolm Gladwell says is bullshit." You simulataneously discount his entire canon of work and also every single underlying scientific paper he has quoted, cited or referenced. That is a lot of work to label bullshit.