3 ms·

Tooling is not computer science.

by Veuxdo 5y ago

rsj_hn 5y ago
But most people study computer science in order to work as software engineer, not computer scientists. Particularly at the undergrad level. This massive hidden subsidy to computer science also results in graduates who don't have a proper background in software engineering and at some point both government officials and the public at large is going to figure out that these are different skillsets with different curricula, and that we need to hire professors who can teach good software engineering principles and practices in our universities rather than just focusing on the computer science aspect and forcing the graduates to figure the engineering out for themselves on the job.
- Veuxdo 5y ago
  Vim and git aren't Software Engineering either. They're just tools that some people (really) like. They aren't truly industry standards, especially Vim. Universities shouldn't be producing graduates who are dependent on specific tools.
  - rsj_hn 5y ago
    Learning how to use professional tools is just scratching the surface of what should be taught. How to design a program, how to document a program, software design patterns and methodologies, how to refactor, secure coding patterns including at a minimum authentication, authorization, and audit systems, how to work with bug tracking systems and repositories, IDEs, unix and shell scripts, how to debug code, etc. Then they should display proficiency with assembly, c or c++, at least one scripting language and at least one higher level OO language and one functional language. Then they should also work on large software projects for practice. Then for electives they can focus on different areas of emphasis such as real time systems, AI, systems programming, database design, networking, web applications, etc. Imagine you study civil engineering but all your classes are taught by physicists and mathematicians, and you are studying physics. Yes, this is all important foundational knowledge for engineering, and you can't be a good engineer if you don't know any physics. But engineering is not physics and to graduate civil engineers without actually teaching them engineering seems like a massive fail. Adding a "missing semester" in which they learn how to use some engineering software is just scratching the surface of what an civil engineering degree should require. That we have gotten away with this in CS-land is just because of the immaturity of the field, not because this is a good way of producing the software engineers that society needs.
    - toomanybeersies 5y ago
      > Then they should display proficiency with assembly, c or c++, at least one scripting language... So graduates are expected to show proficiency in a minimum of 5 different programming languages? That sounds like an enormous waste of time.
      - lostcolony 5y ago
        5 different programming paradigms. A single language can have multiple paradigms. Do you think there are tradeoffs worth considering between language paradigms? Is there a reason why you might want to write something close to the metal, vs something in a scripting language? Is there a reason you might want FP over OO for certain problems? I mean...my curriculum didn't explicitly call out languages anywhere, but along the way I picked up Python (intro class), Java (most classes), C and an x86 assembly (embedded class), C++ (graphics class)...and within two years of graduating I was wanting to learn a functional programming language, and changed jobs just to do it. Had I gone AI track I would have picked up a Lisp. It's not really that big an ask.
        toomanybeersies 5y ago
        When you put it that way, I sort of agree with you. I don't think whether a language is close to the metal or high level is an important distinction to be made in the context of learning computer science, but I do think that learning about different programming paradigms and concepts in the abstract is important. Learning x86 assembly for the sake of it isn't useful or educational. However, building an x86 compiler (or in my case JVM compiler) as part of a languages and compilers course is something worth doing.
        lostcolony 5y ago
        Well, I will say that the class that had us learn assembler started with (software) FPGAs, then assembler, then C. The idea being that it demystifies a lot of what is going on underneath your code. Does it actually help you write better code in most instances? Probably not. It's helpful if you're writing embedded things (and that was the whole gist of the class), but for many devs it's not that useful. But that knowledge is still comforting; I've met devs who didn't have it, and they felt a little uncomfortable that they didn't understand how the code they wrote actually affected the underlying machine. shrug Certainly, "our CS degree is good for anything that isn't too low level" isn't the vibe most universities are going for.
  - lostcolony 5y ago
    Every developer should learn an installed-by-default POSIX standard text editor, when you need to make remote changes to a box (yes, cattle not pets, but the build server is probably a pet. Your personal workstation is probably a pet. Etc). Every developer should learn to work with an SCM tool. Yes, Vim and git are specific choices, but they are reasonable, extremely popular defaults. The first will work basically everywhere, the latter at least gets you the concepts necessary to quickly learn and use something else if needed (while being the most popular of its kind)
    - Veuxdo 5y ago
      > when you need to make remote changes to a box This isn't something you do in CS ever. If you want to learn a technical skill like system administration or git, great. It isn't CS though.
      - rsj_hn 5y ago
        Right, and that's my point. Almost no one gets a CS degree to be a computer scientist. They get a CS degree to be a software engineer. So dontcha think we might want to teach them what software engineers do, rather than what computer scientists do? We don't require civil engineers to get a degree in physics and then pick up civil engineering on the job site. Why do that to software engineers?
      - lostcolony 5y ago
        Why write compilable code at all then? Teach everything in a hypothetical language like (M)MIX, like Knuth does in the Art of Computer Programming. Why bog students down with engineering 'details', like actually compiling their code and getting it to run, or stepping through a debugger, or any such thing. That's not computer science.
yjftsjthsd-h 5y ago
And measuring tools aren't science, but good luck doing physical sciences without measuring anything.

Tooling is not computer science.

by Veuxdo 5y ago

Tooling is not computer science.

But most people study computer science in order to work as software engineer, not computer scientists. Particularly at the undergrad level. This massive hidden subsidy to computer science also results in graduates who don't have a proper background in software engineering and at some point both government officials and the public at large is going to figure out that these are different skillsets with different curricula, and that we need to hire professors who can teach good software engineering principles and practices in our universities rather than just focusing on the computer science aspect and forcing the graduates to figure the engineering out for themselves on the job.

Veuxdo 5y ago
Vim and git aren't Software Engineering either. They're just tools that some people (really) like. They aren't truly industry standards, especially Vim. Universities shouldn't be producing graduates who are dependent on specific tools.
- rsj_hn 5y ago
  Learning how to use professional tools is just scratching the surface of what should be taught. How to design a program, how to document a program, software design patterns and methodologies, how to refactor, secure coding patterns including at a minimum authentication, authorization, and audit systems, how to work with bug tracking systems and repositories, IDEs, unix and shell scripts, how to debug code, etc. Then they should display proficiency with assembly, c or c++, at least one scripting language and at least one higher level OO language and one functional language. Then they should also work on large software projects for practice. Then for electives they can focus on different areas of emphasis such as real time systems, AI, systems programming, database design, networking, web applications, etc. Imagine you study civil engineering but all your classes are taught by physicists and mathematicians, and you are studying physics. Yes, this is all important foundational knowledge for engineering, and you can't be a good engineer if you don't know any physics. But engineering is not physics and to graduate civil engineers without actually teaching them engineering seems like a massive fail. Adding a "missing semester" in which they learn how to use some engineering software is just scratching the surface of what an civil engineering degree should require. That we have gotten away with this in CS-land is just because of the immaturity of the field, not because this is a good way of producing the software engineers that society needs.
  - toomanybeersies 5y ago
    > Then they should display proficiency with assembly, c or c++, at least one scripting language... So graduates are expected to show proficiency in a minimum of 5 different programming languages? That sounds like an enormous waste of time.
    - lostcolony 5y ago
      5 different programming paradigms. A single language can have multiple paradigms. Do you think there are tradeoffs worth considering between language paradigms? Is there a reason why you might want to write something close to the metal, vs something in a scripting language? Is there a reason you might want FP over OO for certain problems? I mean...my curriculum didn't explicitly call out languages anywhere, but along the way I picked up Python (intro class), Java (most classes), C and an x86 assembly (embedded class), C++ (graphics class)...and within two years of graduating I was wanting to learn a functional programming language, and changed jobs just to do it. Had I gone AI track I would have picked up a Lisp. It's not really that big an ask.
      - toomanybeersies 5y ago
        When you put it that way, I sort of agree with you. I don't think whether a language is close to the metal or high level is an important distinction to be made in the context of learning computer science, but I do think that learning about different programming paradigms and concepts in the abstract is important. Learning x86 assembly for the sake of it isn't useful or educational. However, building an x86 compiler (or in my case JVM compiler) as part of a languages and compilers course is something worth doing.
        lostcolony 5y ago
        Well, I will say that the class that had us learn assembler started with (software) FPGAs, then assembler, then C. The idea being that it demystifies a lot of what is going on underneath your code. Does it actually help you write better code in most instances? Probably not. It's helpful if you're writing embedded things (and that was the whole gist of the class), but for many devs it's not that useful. But that knowledge is still comforting; I've met devs who didn't have it, and they felt a little uncomfortable that they didn't understand how the code they wrote actually affected the underlying machine. shrug Certainly, "our CS degree is good for anything that isn't too low level" isn't the vibe most universities are going for.
- lostcolony 5y ago
  Every developer should learn an installed-by-default POSIX standard text editor, when you need to make remote changes to a box (yes, cattle not pets, but the build server is probably a pet. Your personal workstation is probably a pet. Etc). Every developer should learn to work with an SCM tool. Yes, Vim and git are specific choices, but they are reasonable, extremely popular defaults. The first will work basically everywhere, the latter at least gets you the concepts necessary to quickly learn and use something else if needed (while being the most popular of its kind)
  - Veuxdo 5y ago
    > when you need to make remote changes to a box This isn't something you do in CS ever. If you want to learn a technical skill like system administration or git, great. It isn't CS though.
    - rsj_hn 5y ago
      Right, and that's my point. Almost no one gets a CS degree to be a computer scientist. They get a CS degree to be a software engineer. So dontcha think we might want to teach them what software engineers do, rather than what computer scientists do? We don't require civil engineers to get a degree in physics and then pick up civil engineering on the job site. Why do that to software engineers?
    - lostcolony 5y ago
      Why write compilable code at all then? Teach everything in a hypothetical language like (M)MIX, like Knuth does in the Art of Computer Programming. Why bog students down with engineering 'details', like actually compiling their code and getting it to run, or stepping through a debugger, or any such thing. That's not computer science.

yjftsjthsd-h 5y ago

And measuring tools aren't science, but good luck doing physical sciences without measuring anything.