5 ms·

My favorite Knuth quote: "Email is a wonderful thing for people whose role in life is to be on top of things. But not for me; my role is to be on the bottom of

My favorite Knuth quote: "Email is a wonderful thing for people whose role in life is to be on top of things. But not for me; my role is to be on the bottom of things. What I do takes long hours of studying and uninterruptible concentration. I try to learn certain areas of computer science exhaustively; then I try to digest that knowledge into a form that is accessible to people who don't have time for such study." source: https://www-cs-faculty.stanford.edu/~knuth/email.html https://www-cs-faculty.stanford.edu/~knuth/email.html

samstave 8y ago
Thats an amazing quote. Reminds me of a "hermit-like wizard in his tower in solitary uninterrupted study of their craft."
- marmot777 8y ago
  Or the scholar/monk archetype.
- schwurb 8y ago
  I once used two weeks of vacation to learn category theory - that was a really fun and rich time. It really relaxed me, but it also brought me more knowledge than one year of shallowly engaging with the topic.
  - james_s_tayler 8y ago
    I've heard many times on HN people bring up category theory and how beneficial it has been to learn it, so I've recently ordered a textbook on it. Can I ask how has it benefited your thinking? I'm curious to know not only how it benefits an understanding of software development, programming languages and functional programming, but also has it helped you understand the world at large in new and novel ways?
    - whatshisface 8y ago
      If category theory is your first exposure to proofs and logic it can benefit you a lot, although number theory might be a more fun and more concrete way to get the same result. If you know a lot about more than one field of math, as a great big generalization over everything category theory can also help. I should warn you that there is a social bias towards saying that you found a lot of insight in studying something, and a bias against saying that studying it was a waste of time, irrespective of whether or not the thing is actually helpful. Aristotle could stare at a rock and end up with some insight, so anyone who stares at a rock and admits it hasn't helped them has to publicly admit they are not as smart as Aristotle. They say this is how geology got started. ;)
      - leereeves 8y ago
        Aristotle might have been able to stare at a rock and say something clever, but most of his ideas about science were wrong, starting from the very definition of science, which he believed should be based on unprovable first principles (like mathematical axioms). He rejected atomic theory, believed the Earth was the center of the universe and the four fundamental elements were air, fire, water, and earth, and thought heavy objects fell faster and moving objects would tend to come to a stop if nothing interacted with them. These were observations that Aristotle and his audience accepted as "obvious". It took more careful and patient observers to discern the less obvious truth.
        ci5er 8y ago
        It's an iterative process. Anthro-centricism is powerful in terms of what it makes "obvious". The universe spins about the earth. That atoms are made up of objects. That we are the pinnacle of local evolution. That reality is really real.
    - rfergie 8y ago
      Not qualified to speak on this myself, but I asked a similar question to a friend and he said something like: Part of good software development is knowing where and when to create abstractions. Category theory has a collection of interesting and very general abstractions - if you can spot these patterns in your software development then they will be good candidates for abstraction/generalisation
  - leethargo 8y ago
    Now imagine having a year of this kind of intensive study of category theory (or similar), that is, graduate-level classes in math.
    - schwurb 8y ago
      I envy those. I pity not having done my undergrad in maths! I have to say tho that we the examples learned in math department category theory are often not that important or interesting to programmers, and the same is true for the abstractions being studied.
      - leethargo 8y ago
        I did actually do graduate studies in math. Early on I went to Paris for 2 semesters where I tried to follow a course on category theory, in preparation to algebraic geometry. Even though I did lots of abstract algebra before that, the intensity was too much and I could not keep pace. Interestingly, the Chinese students in that class did fairly well, even though they could not follow the actual lectures (given in French) and were studying exclusively from the lecture notes (given in English).
JimboOmega 8y ago
I feel completely the opposite. Not that email or any of the million other apps don't produce constant interruptions - but in terms of what I want my role to be and what I find pleasure in. A world of high stimulation and constant social interaction? A million things shouting at me and I have to sort through them and grab the most important one and hop from thing to thing fast? That's what I like. Focusing on one topic for hours, days, even weeks to really understanding every facet of it? I think it'd kill me. I know people theoretically need these big blocks of uninterrupted time to get work done, but I'd much rather have a bunch of meetings where we can figure out the path forward. I have mad respect for him... but I never could have been an academic. I like to take in all those digested kernels of knowledge, I like to digest as many of them as I can; I know I lose the details, but integrating them all into something bigger has always been my dream; it's just how I live my life.
- nambit 8y ago
  It's only after you've developed a deep understanding that any efforts at integration would be successful. If you just like orchestrating, that's fine but I've met too many managers/product managers who think up lofty ideas on how to integrate various pieces without a deeper understanding of the underlying technology.
  - disishhsha 8y ago
    I think trusting and open relationships between the big picture people and the small detail people can go a long way, if you don’t have a unicorn who can do both.
  - sharadov 8y ago
    Yup, been around a lot of them, they are the ones who love to sling words around and are a prime definition of tech bros always looking to debate not on the basis of firm knowledge but as a cover-up for their inadequacies.
    - JimboOmega 8y ago
      One of the signs that you're actually good at operating at this level is you spend more time listening than talking. You need people who actually know more than you - if you are always arguing and trying to prove you actually do know best, you're definitely not looking for that.
freewilly1040 8y ago
I like how it's a nice articulation of a preference that doesn't denigrate people who have the opposite preference.
- nolok 8y ago
  There are actually two of those in that sentence. The one about emails, and the one about doing the long understanding study.
hodgesrm 8y ago
Stanford CS has at least a couple of people who seem to be really decent humans as well as great computer scientists. Besides Don Knuth, John Ousterhout is another person who comes to mind. Both of them have a humility combined with humor that I find quite admirable.
- black-tea 8y ago
  Really? So everyone else in CS isn't even a "decent human being" by default for you?
  - jnicholasp 8y ago
    They were simply commenting that there were a couple people they especially admired, who they think are really decent (i.e., better than average at being decent). It is neither a necessary nor a natural reading to think that means they believe everyone else is below average.

samstave 8y ago

Thats an amazing quote. Reminds me of a "hermit-like wizard in his tower in solitary uninterrupted study of their craft."

marmot777 8y ago
Or the scholar/monk archetype.
schwurb 8y ago
I once used two weeks of vacation to learn category theory - that was a really fun and rich time. It really relaxed me, but it also brought me more knowledge than one year of shallowly engaging with the topic.
- james_s_tayler 8y ago
  I've heard many times on HN people bring up category theory and how beneficial it has been to learn it, so I've recently ordered a textbook on it. Can I ask how has it benefited your thinking? I'm curious to know not only how it benefits an understanding of software development, programming languages and functional programming, but also has it helped you understand the world at large in new and novel ways?
  - whatshisface 8y ago
    If category theory is your first exposure to proofs and logic it can benefit you a lot, although number theory might be a more fun and more concrete way to get the same result. If you know a lot about more than one field of math, as a great big generalization over everything category theory can also help. I should warn you that there is a social bias towards saying that you found a lot of insight in studying something, and a bias against saying that studying it was a waste of time, irrespective of whether or not the thing is actually helpful. Aristotle could stare at a rock and end up with some insight, so anyone who stares at a rock and admits it hasn't helped them has to publicly admit they are not as smart as Aristotle. They say this is how geology got started. ;)
    - leereeves 8y ago
      Aristotle might have been able to stare at a rock and say something clever, but most of his ideas about science were wrong, starting from the very definition of science, which he believed should be based on unprovable first principles (like mathematical axioms). He rejected atomic theory, believed the Earth was the center of the universe and the four fundamental elements were air, fire, water, and earth, and thought heavy objects fell faster and moving objects would tend to come to a stop if nothing interacted with them. These were observations that Aristotle and his audience accepted as "obvious". It took more careful and patient observers to discern the less obvious truth.
      - ci5er 8y ago
        It's an iterative process. Anthro-centricism is powerful in terms of what it makes "obvious". The universe spins about the earth. That atoms are made up of objects. That we are the pinnacle of local evolution. That reality is really real.
  - rfergie 8y ago
    Not qualified to speak on this myself, but I asked a similar question to a friend and he said something like: Part of good software development is knowing where and when to create abstractions. Category theory has a collection of interesting and very general abstractions - if you can spot these patterns in your software development then they will be good candidates for abstraction/generalisation
- leethargo 8y ago
  Now imagine having a year of this kind of intensive study of category theory (or similar), that is, graduate-level classes in math.
  - schwurb 8y ago
    I envy those. I pity not having done my undergrad in maths! I have to say tho that we the examples learned in math department category theory are often not that important or interesting to programmers, and the same is true for the abstractions being studied.
    - leethargo 8y ago
      I did actually do graduate studies in math. Early on I went to Paris for 2 semesters where I tried to follow a course on category theory, in preparation to algebraic geometry. Even though I did lots of abstract algebra before that, the intensity was too much and I could not keep pace. Interestingly, the Chinese students in that class did fairly well, even though they could not follow the actual lectures (given in French) and were studying exclusively from the lecture notes (given in English).

JimboOmega 8y ago

I feel completely the opposite. Not that email or any of the million other apps don't produce constant interruptions - but in terms of what I want my role to be and what I find pleasure in. A world of high stimulation and constant social interaction? A million things shouting at me and I have to sort through them and grab the most important one and hop from thing to thing fast? That's what I like. Focusing on one topic for hours, days, even weeks to really understanding every facet of it? I think it'd kill me. I know people theoretically need these big blocks of uninterrupted time to get work done, but I'd much rather have a bunch of meetings where we can figure out the path forward. I have mad respect for him... but I never could have been an academic. I like to take in all those digested kernels of knowledge, I like to digest as many of them as I can; I know I lose the details, but integrating them all into something bigger has always been my dream; it's just how I live my life.

nambit 8y ago
It's only after you've developed a deep understanding that any efforts at integration would be successful. If you just like orchestrating, that's fine but I've met too many managers/product managers who think up lofty ideas on how to integrate various pieces without a deeper understanding of the underlying technology.
- disishhsha 8y ago
  I think trusting and open relationships between the big picture people and the small detail people can go a long way, if you don’t have a unicorn who can do both.
- sharadov 8y ago
  Yup, been around a lot of them, they are the ones who love to sling words around and are a prime definition of tech bros always looking to debate not on the basis of firm knowledge but as a cover-up for their inadequacies.
  - JimboOmega 8y ago
    One of the signs that you're actually good at operating at this level is you spend more time listening than talking. You need people who actually know more than you - if you are always arguing and trying to prove you actually do know best, you're definitely not looking for that.

freewilly1040 8y ago

I like how it's a nice articulation of a preference that doesn't denigrate people who have the opposite preference.

nolok 8y ago
There are actually two of those in that sentence. The one about emails, and the one about doing the long understanding study.

hodgesrm 8y ago

Stanford CS has at least a couple of people who seem to be really decent humans as well as great computer scientists. Besides Don Knuth, John Ousterhout is another person who comes to mind. Both of them have a humility combined with humor that I find quite admirable.

black-tea 8y ago
Really? So everyone else in CS isn't even a "decent human being" by default for you?
- jnicholasp 8y ago
  They were simply commenting that there were a couple people they especially admired, who they think are really decent (i.e., better than average at being decent). It is neither a necessary nor a natural reading to think that means they believe everyone else is below average.