5 ms·

It was a pleasure to read, but of course it should be taken with a grain of salt and not applied to every field :-) Imagine being in a software team and disreg

It was a pleasure to read, but of course it should be taken with a grain of salt and not applied to every field :-) Imagine being in a software team and disregarding what others are doing? Screw the rest of the team, I'll work on my own better solution and ignore the parts that are already working well! You might end up with a good solution to the problem, but you'll also manage to piss off a lot of people ^^ A more valuable 'quote' from Feynman would be to "Test everything yourself". Don't just take for granted the results of others but verify it yourself. That (in the field of software) can be quite valuable :-) Still, loved reading that blogpost!

telesilla 9y ago
More to the point, consider being the person who decides what that team builds. If you are constantly looking around you, worried about the competition and not trusting your own instinct, you may never build that singular wonderful thing that helps define your career (and gives you energy to pursue difficult things).
sdrothrock 9y ago
https://en.wikipedia.org/wiki/Trust,_but_verify https://en.wikipedia.org/wiki/Trust,_but_verify
srean 9y ago
> Imagine being in a software team and disregarding what others are doing? Screw the rest of the team, I'll work on my own better solution and ignore the parts that are already working well! If you are competent, have a realistic estimate of your competence, this might not be a bad idea if what you need is a purely technical solution. In many big organizations (lets leave them nameless) most of your time might be going in navigating the politics, guess-estimating different people's internal incentives and how to steer with or around them, not enough time might remain for thinking about the actual technical problem. In these organizations its necessary to take care of these things, because even if you figured out a great solution that ticked all the technical boxes, including technical compatibility with others, all that will come to nought if it is not in alignment with the social and political aspects.
- mannykannot 9y ago
  Unfortunately, for every one who is that person in that situation, there are several who merely think they are, at least in my experience (the most common form seems to be the covert rewriter whose work is more 'pure', 'object-oriented | functional', 'general', 'flexible', 'performant', 'reusable', or my favorite, 'simple' (for some definition of 'simple') than will ever be needed, at the cost of considerable additional (and undocumented) complexity.)
  - srean 9y ago
    Not only do I agree with you, I think it might actually be worse. Seen enough primadonnas who think they can walk on water and have.
- dotancohen 9y ago
  > If you are competent, have a realistic estimate of your competence, this might not be a bad idea if what you need is a purely technical solution. This. Just _yesterday_ some xls-to-json library that has been used in the past was choking on an innocent-looking file. After fiddling with the thing for over an hour, including using LibreOffice to change the file format, I still could not get the library to read the file. In five minutes I had a Python script which read the CSV version of the file and converted it to the necessary format. I chanced writing the script because that I knew that I could. In other situations, gambling company time on "I _might_ be able to help" usually leads to wasting company time.
nabla9 9y ago
>Imagine being in a software team and disregarding what others are doing? Screw the rest of the team Exactly that has been the goal of large scale programming. To find a way to structure programs so that you will be able to disregard others. More accurately: to minimize communication needed to build software.
jamesrcole 9y ago
> A more valuable 'quote' from Feynman would be to "Test everything yourself". Don't just take for granted the results of others but verify it yourself. IMO it's more about "don't follow the path others take, where it's easy to make the same assumptions they did. work through things from first principles yourself".
- noir_lord 9y ago
  The first principle thing is interesting because I've seen Elon Musk hammer that point in many of his talks. SpaceX was started because he sat on a plane and worked out the cost to orbit is limited only by fuel costs if you can make things reusable. It's something I've considered frequently since. Though I'm nowhere near as smart as Feynman or Musk.
  - jamesrcole 9y ago
    I wonder, how many people could actually work through things from first principles, if they were taught how? I don't know, but I suspect a lot more than currently can/do. The education system doesn't currently make any attempt to teach these kinds of skills (AFAIK). I also wonder how much working through stuff like that might actually contribute to making people smart. It would be a valuable thing to find out.
    - noir_lord 9y ago
      When I was in high school I had a teacher who taught us to consider lateral thinking when approaching a problem, it was valuable and something that's come in useful ever since. You are right, schools don't encourage it enough imo.
erasmuse 9y ago
>Imagine being in a software team and disregarding what others are doing? I think the application of 'disregard others' to this case would be to ignore what other software teams working on the same problem are doing.
indigochill 9y ago
>Imagine being in a software team and disregarding what others are doing? I think there's an extent to which this is actually healthy. If we're to build modular code, we shouldn't concern ourselves with the internal workings of modules built by others. We do need to agree on how the modules will interact, but if we need to concern ourselves with how their parts work then we're insufficiently modular. This maps to the social interpretation of this advice as well. You need a healthy interpersonal interface to accept useful input while ignoring useless input.

telesilla 9y ago

More to the point, consider being the person who decides what that team builds. If you are constantly looking around you, worried about the competition and not trusting your own instinct, you may never build that singular wonderful thing that helps define your career (and gives you energy to pursue difficult things).

sdrothrock 9y ago

https://en.wikipedia.org/wiki/Trust,_but_verify https://en.wikipedia.org/wiki/Trust,_but_verify

srean 9y ago

> Imagine being in a software team and disregarding what others are doing? Screw the rest of the team, I'll work on my own better solution and ignore the parts that are already working well! If you are competent, have a realistic estimate of your competence, this might not be a bad idea if what you need is a purely technical solution. In many big organizations (lets leave them nameless) most of your time might be going in navigating the politics, guess-estimating different people's internal incentives and how to steer with or around them, not enough time might remain for thinking about the actual technical problem. In these organizations its necessary to take care of these things, because even if you figured out a great solution that ticked all the technical boxes, including technical compatibility with others, all that will come to nought if it is not in alignment with the social and political aspects.

mannykannot 9y ago
Unfortunately, for every one who is that person in that situation, there are several who merely think they are, at least in my experience (the most common form seems to be the covert rewriter whose work is more 'pure', 'object-oriented | functional', 'general', 'flexible', 'performant', 'reusable', or my favorite, 'simple' (for some definition of 'simple') than will ever be needed, at the cost of considerable additional (and undocumented) complexity.)
- srean 9y ago
  Not only do I agree with you, I think it might actually be worse. Seen enough primadonnas who think they can walk on water and have.
dotancohen 9y ago
> If you are competent, have a realistic estimate of your competence, this might not be a bad idea if what you need is a purely technical solution. This. Just _yesterday_ some xls-to-json library that has been used in the past was choking on an innocent-looking file. After fiddling with the thing for over an hour, including using LibreOffice to change the file format, I still could not get the library to read the file. In five minutes I had a Python script which read the CSV version of the file and converted it to the necessary format. I chanced writing the script because that I knew that I could. In other situations, gambling company time on "I _might_ be able to help" usually leads to wasting company time.

nabla9 9y ago

>Imagine being in a software team and disregarding what others are doing? Screw the rest of the team Exactly that has been the goal of large scale programming. To find a way to structure programs so that you will be able to disregard others. More accurately: to minimize communication needed to build software.

jamesrcole 9y ago

> A more valuable 'quote' from Feynman would be to "Test everything yourself". Don't just take for granted the results of others but verify it yourself. IMO it's more about "don't follow the path others take, where it's easy to make the same assumptions they did. work through things from first principles yourself".

noir_lord 9y ago
The first principle thing is interesting because I've seen Elon Musk hammer that point in many of his talks. SpaceX was started because he sat on a plane and worked out the cost to orbit is limited only by fuel costs if you can make things reusable. It's something I've considered frequently since. Though I'm nowhere near as smart as Feynman or Musk.
- jamesrcole 9y ago
  I wonder, how many people could actually work through things from first principles, if they were taught how? I don't know, but I suspect a lot more than currently can/do. The education system doesn't currently make any attempt to teach these kinds of skills (AFAIK). I also wonder how much working through stuff like that might actually contribute to making people smart. It would be a valuable thing to find out.
  - noir_lord 9y ago
    When I was in high school I had a teacher who taught us to consider lateral thinking when approaching a problem, it was valuable and something that's come in useful ever since. You are right, schools don't encourage it enough imo.

erasmuse 9y ago

>Imagine being in a software team and disregarding what others are doing? I think the application of 'disregard others' to this case would be to ignore what other software teams working on the same problem are doing.

indigochill 9y ago

>Imagine being in a software team and disregarding what others are doing? I think there's an extent to which this is actually healthy. If we're to build modular code, we shouldn't concern ourselves with the internal workings of modules built by others. We do need to agree on how the modules will interact, but if we need to concern ourselves with how their parts work then we're insufficiently modular. This maps to the social interpretation of this advice as well. You need a healthy interpersonal interface to accept useful input while ignoring useless input.