6 ms·

It's only true for integration, and not even all of it. Differentiation can be done mechanically, and killer app of calculus (finding maximum) only requires dif

It's only true for integration, and not even all of it. Differentiation can be done mechanically, and killer app of calculus (finding maximum) only requires differentiation.

iraqmtpizza 3y ago
I dropped out before vector calculus and differential equations but i'm confident they have literally nothing to do with making money in software development. in fact I would wager money that the more math you take during your computer science education the less money you make. say goodbye to startups. say hello to academia and top ramen
- sobriquet9 3y ago
  If you are talking about CRUD, sure. But there is software in self-driving cars, 3D printers, drones, etc. that uses vector calculus and differential equations heavily. Same with applied statistics. For example, if your software has anything to do with quantitative models used in trading, it would be pretty hard to work on it without calculus. Even reading a paper on effectiveness of COVID-19 vaccines requires one to know what p value is and isn't, which in turn requires PDF and CDF (calculus again).
  - iraqmtpizza 3y ago
    >self-driving cars, 3D printers, drones The chance of working on one of these projects is small. The chance of working on the advanced algorithms themselves is even smaller. The chance of getting paid well to do so is 50/50 at best. On average it's a bad bet to go math heavy in computer science. But if it's what you love then go for it. >quantitative models used in trading Probably better off with a degree other than computer science here. Calculus doesn't teach about p-hacking so it's probably not as useful as you think.
    - sobriquet9 3y ago
      Calculus is not heavy math. It describes everyday physics - coordinate, velocity, acceleration, curvature. Stuff you can touch. And it describes unrelated abstract concepts like PDF and CDF, in a way you can intuitively understand by rearranging stacks of Lego bricks. I agree that a degree in a quantitative field other than computer science is better. A CS degree is like saying "I can drive a car and use Microsoft Excel". You need to either be a really good driver, or be ready to work harder than many others to get ahead.
      - iraqmtpizza 3y ago
        I don't care what it describes. Six advanced math courses is math heavy for computer science insofar as private sector employment is concerned. This is a descriptive fact about the world. Go nuts with it. We're talking statistics, differential calc, integral calc, vector calc, differential equations, discrete math. Especially when only one of the six (discrete math) is useful for making money, which is the yardstick I selected. Curvature has literally nothing to do with making money, on average. Being able to describe physical reality at a fundamental level with specificity is not adaptive on the level of the individual. If you want to donate your time to humanity and eschew money, go ahead. But don't be sour grapes about your educational choices.

It's only true for integration, and not even all of it. Differentiation can be done mechanically, and killer app of calculus (finding maximum) only requires differentiation.

iraqmtpizza 3y ago

I dropped out before vector calculus and differential equations but i'm confident they have literally nothing to do with making money in software development. in fact I would wager money that the more math you take during your computer science education the less money you make. say goodbye to startups. say hello to academia and top ramen

sobriquet9 3y ago
If you are talking about CRUD, sure. But there is software in self-driving cars, 3D printers, drones, etc. that uses vector calculus and differential equations heavily. Same with applied statistics. For example, if your software has anything to do with quantitative models used in trading, it would be pretty hard to work on it without calculus. Even reading a paper on effectiveness of COVID-19 vaccines requires one to know what p value is and isn't, which in turn requires PDF and CDF (calculus again).
- iraqmtpizza 3y ago
  >self-driving cars, 3D printers, drones The chance of working on one of these projects is small. The chance of working on the advanced algorithms themselves is even smaller. The chance of getting paid well to do so is 50/50 at best. On average it's a bad bet to go math heavy in computer science. But if it's what you love then go for it. >quantitative models used in trading Probably better off with a degree other than computer science here. Calculus doesn't teach about p-hacking so it's probably not as useful as you think.
  - sobriquet9 3y ago
    Calculus is not heavy math. It describes everyday physics - coordinate, velocity, acceleration, curvature. Stuff you can touch. And it describes unrelated abstract concepts like PDF and CDF, in a way you can intuitively understand by rearranging stacks of Lego bricks. I agree that a degree in a quantitative field other than computer science is better. A CS degree is like saying "I can drive a car and use Microsoft Excel". You need to either be a really good driver, or be ready to work harder than many others to get ahead.
    - iraqmtpizza 3y ago
      I don't care what it describes. Six advanced math courses is math heavy for computer science insofar as private sector employment is concerned. This is a descriptive fact about the world. Go nuts with it. We're talking statistics, differential calc, integral calc, vector calc, differential equations, discrete math. Especially when only one of the six (discrete math) is useful for making money, which is the yardstick I selected. Curvature has literally nothing to do with making money, on average. Being able to describe physical reality at a fundamental level with specificity is not adaptive on the level of the individual. If you want to donate your time to humanity and eschew money, go ahead. But don't be sour grapes about your educational choices.