4 ms·

>I don’t expect engineers to be better than me. I respect your opinion but I find this strategy very strange. (Or maybe I don't exactly understand what you're

by jasode 9y ago

>I don’t expect engineers to be better than me. I respect your opinion but I find this strategy very strange. (Or maybe I don't exactly understand what you're saying.) In my view, it would be a dream scenario if the next 10 programmers I hire were all superior to me. If I was the worst programmer on the team, that would be an ideal outcome. Yes, I've been programming for 30+ years but I'm also self-aware of my limitations. This gives me the ability to recognize better programmers and learn from them. If I'm not hiring interns, I do expect (or at least hope) that they are better than me. Am I competent enough to write a "do/while" loop and knock out FizzBuzz in 5 minutes?!? Sure, but that doesn't mean there aren't better programmers than me. I would consider it a great business skill if I'm able to consistently attract superior programmers -- either to work with me or for me. Larry Page attracted programmers better than him (e.g. Jeff Dean). Yes, Larry earned a masters degree in computer science but his java programming (the rudimentary 1996 crawler) wasn't considered very good. His employees rewrote the whole thing in C++ and made it scale for billions of pages. Bill Gates hired superior programmers. Mark Zuckerberg hired superior programmers. Etc. etc.

reinhardt 9y ago
You actually agree with each other. You're saying "it would be a dream scenario if the next 10 programmers I hire were all superior to me"; it's an ideal outcome, even if unlikely. The parent's "I don’t expect engineers to be better than me" merely points out the unlikeliness, you point out the desirability.
- ljw1001 9y ago
  When I was at one company (later acquired by Google), one of the developers created a model to show just how unlikely it was that every new hire would continue to raise the average level of competence for the team as a whole (as our hiring traditions espoused). After years of hiring very good programmers, it just wasn't going to be possible to hire many more who could pass that test.
  - andai 9y ago
    Could you elaborate on the hiring process? Was it calibrated to hire people more competent than X% of the current engineers?
- jasode 9y ago
  >You actually agree with each other. It doesn't seem like it. The next 3 sentences from the thegayngler: > I have but one qualification. Can they do the job? Are they strong enough that I can guide them into the position I need them at if it is required. In my mind, programmers superior to me have abilities that would be beyond my knowledge to "guide" them. They can come up with programming solutions I could never think of. It's very realistic that "the qualification to do the job" essentially means they are a better programmer than me.
  - kbenson 9y ago
    > In my mind, programmers superior to me have abilities that would be beyond my knowledge to "guide" them. Presumably, in your mind, "programming" skill is along one spectrum and one focus, as well? I think that's part of what's being talked about here. There are many, many aspects to software engineering/programming (depending on whether you want to classify them differently), and many roads that could have been taken to get to the point someone is at. I see no reason to think that someone that is better or worse than me in one aspect will also be so in every one of the other aspects of the job.
    - jasode 9y ago
      >There are many, many aspects to software engineering/programming [...] I see no reason to think that someone that is better or worse than me in one aspect will also be so in every one of the other aspects of the job. Let me try to restate that in more general terms and you can tell me if I interpreted it correctly: Since programming skill is multi-dimensional / multi-faceted, and it is unlikely for a programmer to be superior in _all_ dimensions compared to another, it is therefore not possible to conclude if one programmer is superior to another. Is that a fair rewording? My response is... sure, programming skill can be looked at as multi-dimensional. But so can other real-world comparisons. We prioritize the dimensions that are more important and collapse all aspects into a composite score. Multidimensions such as choosing a new job: Company X is a 5-minute commute with 4 weeks vacation but Company Y has the newest technology and has lucrative stock options. Yes, it already goes without saying that there's more than 1 aspect of "desirable working conditions" at each company and no single company can claim a superior score on _all_ aspects. We can still weigh all aspects and eventually pick one (and only one) which is superior. If I hired a "Jeff Dean" type of programmer, it would be an upgrade to my team and my company. Could I search for an "aspect" that I'm better at than him? Well, looking over his expertise[1], it looks like he doesn't have much frontend GUI programming experience. So yes, I guess I could claim superior skill on writing event handlers in C# language for buttons. That's not important though for my assessment. My point is I don't need for programmers to be superior to me in every aspect for me to determine that they are superior overall. (Same as determining which company is "superior" to work for.) If I was founding a startup and I was the "ceiling" of programming skill instead of the minimum "baseline", my company would fail. I think I'm pretty decent at programming but I have no Dunning-Kruger delusions that I can "train" anybody that can pass FizzBuzz into a "Jeff Dean". I know of no 6-week bootcamp or even a 6-month corporate training program that can claim that feat. What I can do is recognize the programming skills in others that are (overall) superior to mine and hopefully convince the candidates to come work with me. [1] https://scholar.google.com/citations?user=NMS69lQAAAAJ&hl=en https://scholar.google.com/citations?user=NMS69lQAAAAJ&hl=en
danielvinson 9y ago
I believe he is referring to the case of the interviewer being a peer rather than a manager. In my experience the manager cares about fit and attracting you to the team, whereas the peers are all looking for reasons NOT to hire you.
ljw1001 9y ago
I recall one of my friends, when I was at Google, sending me a screenshot of code written by Jeff Dean with an obvious bug. Jeff is, of course, extraordinary, but I wonder how many companies might have passed on him if he submitted that code in an interview setting.
- sp527 9y ago
  This anecdote made my day. Thank you for helping assuage my Imposter Syndrome.
- animal531 9y ago
  Absolutely. There exists no single human being that does not make any mistakes. The trick is to find the people that make as few as possible, know how to track them down, and that can figure out how to hot-fix them as needed.
thegayngler 9y ago
Of course you want the best you can get. I’m arguing 10x is constantly influx for any number of reasons. So it’s unlikely you’ll get one. If you have a process for making engineers into the best then you don’t need to look for the unlikely 10x because you can make them into the 10x high performers you need. Ive seen companies literally turn good engineers into underperforming engineers...stifling them with process, tech debt, politics, etc... Then they get angry some people can perform and others can’t under the circumstances and conditions the management created. I’ve watched perfectly good engineers get fired or forced out for not being miracle workers only for people to later realize there was nothing wrong with the engineer in the first place. You don’t know who is going to be the best until you work with them. I would advise people to pick people they will like even under the worst of conditions and train them to the way you want them to be.
- freedomben 9y ago
  I have observed exactly the same thing. It's truly mindblowing. I also find it remarkable how quick people are to blame the engineer, rather than considering the whole picture (process, tech debt, politics, etc). It's a complicated equation with a large number of variables.
hw89y24y89 9y ago
Java in 1996 was version 1! Doesn't necessarily say much about Larry'd code-foo :) Jokes aside; I agree with the advice of hiring people smarter than you in the domains where you won't be paying attention should the company actually take off. If you're looking to be a founder/CEO level type, doesn't seem to matter what your skill in Java and OOP are, yes? No need to be a water walker. That's the "genius" of management; they sell an idea and get someone else to do the work.

by jasode 9y ago

reinhardt 9y ago

You actually agree with each other. You're saying "it would be a dream scenario if the next 10 programmers I hire were all superior to me"; it's an ideal outcome, even if unlikely. The parent's "I don’t expect engineers to be better than me" merely points out the unlikeliness, you point out the desirability.

ljw1001 9y ago
When I was at one company (later acquired by Google), one of the developers created a model to show just how unlikely it was that every new hire would continue to raise the average level of competence for the team as a whole (as our hiring traditions espoused). After years of hiring very good programmers, it just wasn't going to be possible to hire many more who could pass that test.
- andai 9y ago
  Could you elaborate on the hiring process? Was it calibrated to hire people more competent than X% of the current engineers?
jasode 9y ago
>You actually agree with each other. It doesn't seem like it. The next 3 sentences from the thegayngler: > I have but one qualification. Can they do the job? Are they strong enough that I can guide them into the position I need them at if it is required. In my mind, programmers superior to me have abilities that would be beyond my knowledge to "guide" them. They can come up with programming solutions I could never think of. It's very realistic that "the qualification to do the job" essentially means they are a better programmer than me.
- kbenson 9y ago
  > In my mind, programmers superior to me have abilities that would be beyond my knowledge to "guide" them. Presumably, in your mind, "programming" skill is along one spectrum and one focus, as well? I think that's part of what's being talked about here. There are many, many aspects to software engineering/programming (depending on whether you want to classify them differently), and many roads that could have been taken to get to the point someone is at. I see no reason to think that someone that is better or worse than me in one aspect will also be so in every one of the other aspects of the job.
  - jasode 9y ago
    >There are many, many aspects to software engineering/programming [...] I see no reason to think that someone that is better or worse than me in one aspect will also be so in every one of the other aspects of the job. Let me try to restate that in more general terms and you can tell me if I interpreted it correctly: Since programming skill is multi-dimensional / multi-faceted, and it is unlikely for a programmer to be superior in _all_ dimensions compared to another, it is therefore not possible to conclude if one programmer is superior to another. Is that a fair rewording? My response is... sure, programming skill can be looked at as multi-dimensional. But so can other real-world comparisons. We prioritize the dimensions that are more important and collapse all aspects into a composite score. Multidimensions such as choosing a new job: Company X is a 5-minute commute with 4 weeks vacation but Company Y has the newest technology and has lucrative stock options. Yes, it already goes without saying that there's more than 1 aspect of "desirable working conditions" at each company and no single company can claim a superior score on _all_ aspects. We can still weigh all aspects and eventually pick one (and only one) which is superior. If I hired a "Jeff Dean" type of programmer, it would be an upgrade to my team and my company. Could I search for an "aspect" that I'm better at than him? Well, looking over his expertise[1], it looks like he doesn't have much frontend GUI programming experience. So yes, I guess I could claim superior skill on writing event handlers in C# language for buttons. That's not important though for my assessment. My point is I don't need for programmers to be superior to me in every aspect for me to determine that they are superior overall. (Same as determining which company is "superior" to work for.) If I was founding a startup and I was the "ceiling" of programming skill instead of the minimum "baseline", my company would fail. I think I'm pretty decent at programming but I have no Dunning-Kruger delusions that I can "train" anybody that can pass FizzBuzz into a "Jeff Dean". I know of no 6-week bootcamp or even a 6-month corporate training program that can claim that feat. What I can do is recognize the programming skills in others that are (overall) superior to mine and hopefully convince the candidates to come work with me. [1] https://scholar.google.com/citations?user=NMS69lQAAAAJ&hl=en https://scholar.google.com/citations?user=NMS69lQAAAAJ&hl=en

danielvinson 9y ago

I believe he is referring to the case of the interviewer being a peer rather than a manager. In my experience the manager cares about fit and attracting you to the team, whereas the peers are all looking for reasons NOT to hire you.

ljw1001 9y ago

I recall one of my friends, when I was at Google, sending me a screenshot of code written by Jeff Dean with an obvious bug. Jeff is, of course, extraordinary, but I wonder how many companies might have passed on him if he submitted that code in an interview setting.

sp527 9y ago
This anecdote made my day. Thank you for helping assuage my Imposter Syndrome.
animal531 9y ago
Absolutely. There exists no single human being that does not make any mistakes. The trick is to find the people that make as few as possible, know how to track them down, and that can figure out how to hot-fix them as needed.

thegayngler 9y ago

Of course you want the best you can get. I’m arguing 10x is constantly influx for any number of reasons. So it’s unlikely you’ll get one. If you have a process for making engineers into the best then you don’t need to look for the unlikely 10x because you can make them into the 10x high performers you need. Ive seen companies literally turn good engineers into underperforming engineers...stifling them with process, tech debt, politics, etc... Then they get angry some people can perform and others can’t under the circumstances and conditions the management created. I’ve watched perfectly good engineers get fired or forced out for not being miracle workers only for people to later realize there was nothing wrong with the engineer in the first place. You don’t know who is going to be the best until you work with them. I would advise people to pick people they will like even under the worst of conditions and train them to the way you want them to be.

freedomben 9y ago
I have observed exactly the same thing. It's truly mindblowing. I also find it remarkable how quick people are to blame the engineer, rather than considering the whole picture (process, tech debt, politics, etc). It's a complicated equation with a large number of variables.

hw89y24y89 9y ago

Java in 1996 was version 1! Doesn't necessarily say much about Larry'd code-foo :) Jokes aside; I agree with the advice of hiring people smarter than you in the domains where you won't be paying attention should the company actually take off. If you're looking to be a founder/CEO level type, doesn't seem to matter what your skill in Java and OOP are, yes? No need to be a water walker. That's the "genius" of management; they sell an idea and get someone else to do the work.