4 ms·

I think there's more to it than that. Languages, including programming languages, affect the way you think. When you're proficient in a language your brain adju

by guelo 10y ago

I think there's more to it than that. Languages, including programming languages, affect the way you think. When you're proficient in a language your brain adjusts to the language's styles and limitations so you can express yourself efficiently using those modes of thought. If you are only proficient in one language you can fall into the trap of thinking it is the only way of expressing your thoughts efficiently, and that's where you get the heated flame wars. This is why it is important to become proficient in languages with completely different paradigms to better understand the trade-offs. When I switch between two languages in which I am proficient I can almost feel my brain shifting gears to a different mode of thought.

grey-area 10y ago
Wittgenstein put this really well : The language that I speak defines the limits of my world. I've always thought it was particularly apt for programming languages.
- PeCaN 10y ago
  I find it a bit funny that the Sapir-Whorf hypothesis, though invented for and now largely discredited for natural languages, turns out to strongly apply to programming languages. https://en.wikipedia.org/wiki/Linguistic_relativity https://en.wikipedia.org/wiki/Linguistic_relativity
  - guelo 10y ago
    Wikipedia says "Currently, a balanced view of linguistic relativity is espoused by most linguists", so I wouldn't call it discredited, just more nuanced.
    - jcranmer 10y ago
      People have used the term linguistic relativity to mean anything on a vast range of specific hypotheses, which makes it hard to figure out what exactly people mean when they say it's been proven or disproven. Nevertheless, it's pretty clear to most people that the stronger forms of the hypothesis--the idea that language is essentially a straightjacket on the kinds of thoughts you can express (or linguistic determination in general)--is thoroughly debunked. I personally rather suspect that the weaker forms (that language generally shapes but not fully limits how you view the world) are also equally fallacious, even if language and world-views might be weakly correlative--instead, I see language and world-view as essentially both being strongly derived from that of those who teach you.
      - grey-area 10y ago
        I think this depends on the language. For example Mathematics is generally accepted to be incomplete it what it can express (Gödel), and programming languages, while expressive and in theory turing complete, still limit the thoughts you can easily express, which I suspect will guide users to do things a certain way. Natural language is a bit different because of borrowed words and the flexibility of the humans parsing it.
- theoh 10y ago
  There's also this, from a similar milieu: "good notation has a subtlety and suggestiveness about it which at times makes it seem almost like a live teacher . . . and a perfect notation would be a substitute for thought." Bertrand Russell. ("diagram chasing" in category theory might be an example of thought being almost eliminated by a notation...)
  - stcredzero 10y ago
    a perfect notation would be a substitute for thought." -- Bertrand Russell. Something that would only be said by someone who never thought about the time and space complexity of theorem proving algorithms. (Given his areas of contribution, I can't be sure if he ever did or he didn't.)
    - theoh 10y ago
      He also said "Most people would rather die than think; many do." In other words, not all human cognition is "thought", for Russell. So I don't think he's making any claim that implies total absence of computation with a perfect notation. I'd say he's in the clear on this one.
xapata 10y ago
> Languages, including programming languages, affect the way you think That's a nice-sounding hypothesis, but it's difficult to falsify. Steven Pinker reviewed some of the scientific literature on that idea in "The Stuff of Thought" and it mostly holds up the null hypothesis. I believe that hypothesis is true in the trivial sense -- nearly everything you learn affects the way you think -- but I'm not convinced it's true in the more interesting sense.

by guelo 10y ago

grey-area 10y ago

Wittgenstein put this really well : The language that I speak defines the limits of my world. I've always thought it was particularly apt for programming languages.

PeCaN 10y ago
I find it a bit funny that the Sapir-Whorf hypothesis, though invented for and now largely discredited for natural languages, turns out to strongly apply to programming languages. https://en.wikipedia.org/wiki/Linguistic_relativity https://en.wikipedia.org/wiki/Linguistic_relativity
- guelo 10y ago
  Wikipedia says "Currently, a balanced view of linguistic relativity is espoused by most linguists", so I wouldn't call it discredited, just more nuanced.
  - jcranmer 10y ago
    People have used the term linguistic relativity to mean anything on a vast range of specific hypotheses, which makes it hard to figure out what exactly people mean when they say it's been proven or disproven. Nevertheless, it's pretty clear to most people that the stronger forms of the hypothesis--the idea that language is essentially a straightjacket on the kinds of thoughts you can express (or linguistic determination in general)--is thoroughly debunked. I personally rather suspect that the weaker forms (that language generally shapes but not fully limits how you view the world) are also equally fallacious, even if language and world-views might be weakly correlative--instead, I see language and world-view as essentially both being strongly derived from that of those who teach you.
    - grey-area 10y ago
      I think this depends on the language. For example Mathematics is generally accepted to be incomplete it what it can express (Gödel), and programming languages, while expressive and in theory turing complete, still limit the thoughts you can easily express, which I suspect will guide users to do things a certain way. Natural language is a bit different because of borrowed words and the flexibility of the humans parsing it.
theoh 10y ago
There's also this, from a similar milieu: "good notation has a subtlety and suggestiveness about it which at times makes it seem almost like a live teacher . . . and a perfect notation would be a substitute for thought." Bertrand Russell. ("diagram chasing" in category theory might be an example of thought being almost eliminated by a notation...)
- stcredzero 10y ago
  a perfect notation would be a substitute for thought." -- Bertrand Russell. Something that would only be said by someone who never thought about the time and space complexity of theorem proving algorithms. (Given his areas of contribution, I can't be sure if he ever did or he didn't.)
  - theoh 10y ago
    He also said "Most people would rather die than think; many do." In other words, not all human cognition is "thought", for Russell. So I don't think he's making any claim that implies total absence of computation with a perfect notation. I'd say he's in the clear on this one.

xapata 10y ago

> Languages, including programming languages, affect the way you think That's a nice-sounding hypothesis, but it's difficult to falsify. Steven Pinker reviewed some of the scientific literature on that idea in "The Stuff of Thought" and it mostly holds up the null hypothesis. I believe that hypothesis is true in the trivial sense -- nearly everything you learn affects the way you think -- but I'm not convinced it's true in the more interesting sense.