4 ms·

He had, what, 10 man-years of experience at this sort of thing, at most? His opposition had what sounds like a man-millenium of experience, and he made A$5M of

by lsb 17y ago

He had, what, 10 man-years of experience at this sort of thing, at most? His opposition had what sounds like a man-millenium of experience, and he made A$5M off of them.

jey 17y ago
The "skill is proportional to experience" idea always puzzles me, especially in fields like finance and software where differences in productivity are measured in orders of magnitude. The most effective investors and programmers easily have 100x greater maximum productivity than the median.
- ryanwaggoner 17y ago
  The most effective investors and programmers easily have 100x greater productivity than the median. I have a hard time believing that for programmers. 10x maybe, not 100x. If this were true, they should be making tremendously more than median programmers. When was the last time you heard about a software engineer employee making over $1m, let alone $10m?
  - lamby 17y ago
    > I have a hard time believing that for programmers. 10x maybe, not 100x. Peopleware claims 10x.
  - jey 17y ago
    > software engineer employee making over $1m Employers don't have any incentive to pay an especially effective programmer at whatever his "market rate" actually is. They just need to pay enough to fill their position with someone skilled, and there's plenty of programmers who'll take the near-guarantee of $MaxSalary/yr and live happily. ("After you've got enough to settle down and buy hardware whenever you need it, what else do you need?") If you want to realize your full "fair market value" (which is just a convenient theoretical fiction anyway), you need to take matters into your own hands. I also don't think there's a consistent 100x variation in overall productivity/output, but maybe something more like maximum productivity (and edited the grandparent post). If you have a super-productive employee it's going to be hard to "fully saturate the pipeline" since any one single project isn't going to constantly require that level of work, and you can't just have him hop project to project within the company since there's a learning curve each time. I originally picked up the "programmer productivity varies a lot" meme from one of PG's essays, but have since seen it in industry: http://paulgraham.com/gh.html http://paulgraham.com/gh.html
    - netsp 17y ago
      If the value a select few brought to a company was > $10m, and this was widely known, they would bid up the price. Of course employers don't have an incentive to pay above market rate, that's almost a tautology. That is what market rate means. However, if there was so much value to be had, market rate would rise.
      - jey 17y ago
        You seem to have misunderstood what I was talking about. I didn't contravene you at all.
  - arohner 17y ago
    "When was the last time you heard about a software engineer employee making over $1m, let alone $10m?" Exactly. employees don't make make over $1m, but plenty of founders who sold startups have.
    - pyre 17y ago
      Does that really qualify as 'programmer' productivity though? Being a productive programmer is different from having a good idea and having the business sense to capitalize on it. Sure it may require productive programming to create it, but I would hardly say that 'hyper-productive programming' is at the core of most successful startups.
  - hristov 17y ago
    You are incorrectly assuming that pay is fair. I can easily believe that 100x number. Especially considering how when a mediocre programmer under-performs he is usually augmented by other mediocre programmers which tends reduce the efficiency even more. Anyway, Paul Graham wrote a pretty convincing essay about this.
  - Dilpil 17y ago
    Measuring things by times-faster is not really the right way to do it. Some software engineers are capable of completing important tasks, some are not. They are (error-divide-by-zero) better.
- hegemonicon 17y ago
  Years of experience gives you much better information to work with, but you're still using your same old thought processes to...process it. You can offset a lower intelligence with better information, but only to a point.
RevRal 17y ago
Yes. The words "man-years" is key. This guy had balls.

by lsb 17y ago

He had, what, 10 man-years of experience at this sort of thing, at most? His opposition had what sounds like a man-millenium of experience, and he made A$5M off of them.

jey 17y ago

The "skill is proportional to experience" idea always puzzles me, especially in fields like finance and software where differences in productivity are measured in orders of magnitude. The most effective investors and programmers easily have 100x greater maximum productivity than the median.

ryanwaggoner 17y ago
The most effective investors and programmers easily have 100x greater productivity than the median. I have a hard time believing that for programmers. 10x maybe, not 100x. If this were true, they should be making tremendously more than median programmers. When was the last time you heard about a software engineer employee making over $1m, let alone $10m?
- lamby 17y ago
  > I have a hard time believing that for programmers. 10x maybe, not 100x. Peopleware claims 10x.
- jey 17y ago
  > software engineer employee making over $1m Employers don't have any incentive to pay an especially effective programmer at whatever his "market rate" actually is. They just need to pay enough to fill their position with someone skilled, and there's plenty of programmers who'll take the near-guarantee of $MaxSalary/yr and live happily. ("After you've got enough to settle down and buy hardware whenever you need it, what else do you need?") If you want to realize your full "fair market value" (which is just a convenient theoretical fiction anyway), you need to take matters into your own hands. I also don't think there's a consistent 100x variation in overall productivity/output, but maybe something more like maximum productivity (and edited the grandparent post). If you have a super-productive employee it's going to be hard to "fully saturate the pipeline" since any one single project isn't going to constantly require that level of work, and you can't just have him hop project to project within the company since there's a learning curve each time. I originally picked up the "programmer productivity varies a lot" meme from one of PG's essays, but have since seen it in industry: http://paulgraham.com/gh.html http://paulgraham.com/gh.html
  - netsp 17y ago
    If the value a select few brought to a company was > $10m, and this was widely known, they would bid up the price. Of course employers don't have an incentive to pay above market rate, that's almost a tautology. That is what market rate means. However, if there was so much value to be had, market rate would rise.
    - jey 17y ago
      You seem to have misunderstood what I was talking about. I didn't contravene you at all.
- arohner 17y ago
  "When was the last time you heard about a software engineer employee making over $1m, let alone $10m?" Exactly. employees don't make make over $1m, but plenty of founders who sold startups have.
  - pyre 17y ago
    Does that really qualify as 'programmer' productivity though? Being a productive programmer is different from having a good idea and having the business sense to capitalize on it. Sure it may require productive programming to create it, but I would hardly say that 'hyper-productive programming' is at the core of most successful startups.
- hristov 17y ago
  You are incorrectly assuming that pay is fair. I can easily believe that 100x number. Especially considering how when a mediocre programmer under-performs he is usually augmented by other mediocre programmers which tends reduce the efficiency even more. Anyway, Paul Graham wrote a pretty convincing essay about this.
- Dilpil 17y ago
  Measuring things by times-faster is not really the right way to do it. Some software engineers are capable of completing important tasks, some are not. They are (error-divide-by-zero) better.
hegemonicon 17y ago
Years of experience gives you much better information to work with, but you're still using your same old thought processes to...process it. You can offset a lower intelligence with better information, but only to a point.

RevRal 17y ago

Yes. The words "man-years" is key. This guy had balls.