3 ms·

I am surprised by how much time CS undergrads are required to waste on irrelevant subjects such as Physics. Here's an example CS curriculum at University of Ca

I am surprised by how much time CS undergrads are required to waste on irrelevant subjects such as Physics. Here's an example CS curriculum at University of California at Los Angeles: https://www.seasoasa.ucla.edu/curric-19-20/80-comscicur19.html https://www.seasoasa.ucla.edu/curric-19-20/80-comscicur19.ht... English Composition 3 Differential and Integral calculus Integration and Infinite Series Physics 1A — Mechanics Calculus of Several Variables Oscillations, Waves, Electric and Magnetic Fields Calculus of Several Variables UCLA Samueli Ethics Course Linear Algebra and Applications Introduction to Discrete Structures Electrodynamics, Optics, and Special Relativity Differential Equations Probability Elective That's a whole bunch of useless courses. These courses are intended to make sure you're well-rounded in your education. But spending more than 50% of the time making sure you're well-rounded in Physics is useless for a software engineer. 99.9% of software engineers don't use physics in their career (except maybe some game or graphics programming). 99% of software engineers don't need Calculus (except maybe for innovating some ML core algorithms). A better way to be a well-rounded software engineer is to make sure graduates know more than just algorithms and a couple of programming languages. They should know Networking, Databases, Operating Systems, Machine Learning, Big Data, Distributed Systems, etc. Because of the time spent on general education (that will be forgotten as soon as they graduate), CS graduates are missing out on valuable topics such as those I listed.

908B64B197 6y ago
You are looking for a bootcamp, not a Software/Computer Engineering/Science degree. English Composition 3 - Core skill to effectively communicate in a professional setting. Differential and Integral calculus - Required for AI/Complex algorithms. Integration and Infinite Series - Series are key to many algorithms and results. UCLA Samueli Ethics Course - Fashionable. Linear Algebra and Applications - Core of ML/AI/Graphics/High performance computing Introduction to Discrete Structures - Core and fundamental for algorithms and data structures. Differential Equations - Required for any serious ML/AI Probability Elective - Pretty much required to do or read anything AI related. They should take what you listed as well. Should be possible in 4 years.
- souprock 6y ago
  I don't see it as "looking for a bootcamp", and that sure sounds classist in a bad way. Bootcamps are highly non-standard and usually weak, with very little time spent on the fundamentals of computing. (building an OS or compiler, theory of computation, etc.) Wanting to avoid college nonsense isn't the same as wanting a bootcamp. It is legitimate to want a 4-year degree that is 100% meat, no fluff. The desired degree would be like a BS plus MS, minus the AA that is built into every BS. There are an awful lot of people wanting that. Schools don't want to offer it because the instructors would be costly, because demand for BS/MS combined degrees would go down, because administrators personally love non-technical subjects, and so on. Standardization and accreditation are additional barriers to giving students what students want, though conveniently both are under the influence of the schools.
  - 908B64B197 6y ago
    A lot of the "fluf" is really the foundation on which a lot of courses is built. Linear Algebra and Probabilities for instance are pretty much required for any level of AI. Compiler relies heavily on Discrete Math. Maybe the physics is a little bit useless. But it helps by providing a practical application for the math.
    - flowerlad 6y ago
      > A lot of the "fluf" is really the foundation on which a lot of courses is built. What is the point of a massive foundation if the superstructure you're building on top of it is only barely livable? Maybe we need better balance.

908B64B197 6y ago

You are looking for a bootcamp, not a Software/Computer Engineering/Science degree. English Composition 3 - Core skill to effectively communicate in a professional setting. Differential and Integral calculus - Required for AI/Complex algorithms. Integration and Infinite Series - Series are key to many algorithms and results. UCLA Samueli Ethics Course - Fashionable. Linear Algebra and Applications - Core of ML/AI/Graphics/High performance computing Introduction to Discrete Structures - Core and fundamental for algorithms and data structures. Differential Equations - Required for any serious ML/AI Probability Elective - Pretty much required to do or read anything AI related. They should take what you listed as well. Should be possible in 4 years.

souprock 6y ago
I don't see it as "looking for a bootcamp", and that sure sounds classist in a bad way. Bootcamps are highly non-standard and usually weak, with very little time spent on the fundamentals of computing. (building an OS or compiler, theory of computation, etc.) Wanting to avoid college nonsense isn't the same as wanting a bootcamp. It is legitimate to want a 4-year degree that is 100% meat, no fluff. The desired degree would be like a BS plus MS, minus the AA that is built into every BS. There are an awful lot of people wanting that. Schools don't want to offer it because the instructors would be costly, because demand for BS/MS combined degrees would go down, because administrators personally love non-technical subjects, and so on. Standardization and accreditation are additional barriers to giving students what students want, though conveniently both are under the influence of the schools.
- 908B64B197 6y ago
  A lot of the "fluf" is really the foundation on which a lot of courses is built. Linear Algebra and Probabilities for instance are pretty much required for any level of AI. Compiler relies heavily on Discrete Math. Maybe the physics is a little bit useless. But it helps by providing a practical application for the math.
  - flowerlad 6y ago
    > A lot of the "fluf" is really the foundation on which a lot of courses is built. What is the point of a massive foundation if the superstructure you're building on top of it is only barely livable? Maybe we need better balance.