3 ms·

And this applies especially to code.

by vrode 16y ago

lmkg 16y ago
At my college, the "gatekeeper" CS course was notorious, even among non-CS majors. Your code could compile, pass all the unit tests, have optimal algorithmic efficiency and highly optimized runtime... and you could still get a C- because you used inconsistent capitalization for your variable names, or had misspellings in your comments[1]. May the ghost of Dijkstra have mercy on you if you did anything "clever" at the high level. The goal, of course, is not that later on you will spend half your coding time thinking about variable capitalization. The goal is that when you're writing your code, you pick good variable names automatically because you've built up the reflex, by having the idea pounded into your head with the wrath of an angry god. Anecdotally, the class was largely successful in its goal of setting up good habits before the bad ones settled in. After taking that class, people wrote code that was significantly more readable and maintainable than they had written before, and the habits tended to stick. I suspect that at least part of the problems with effective communications is that we develop our early communications largely ad-hoc, and tend to build up a lot of bad habits in the process. Without concentrated effort to correct this habits, they remain and reinforce themselves through repetition. And of course, concentrated effort doesn't scale well. [1] Granted, part of this was because the class was graded on a negative curve. If everyone got 145/150, then 145 is the average, which means a C.
- count 16y ago
  In high school, my senior AP Eng. Comp class was similar. We had to write 2-3 page paper, in class, every day. The papers were all literary analysis (ugh!), but it definitely sharpened the 'writing muscle'. I can write much faster, with more clarity, than most people I know today - all because of that one year of English.

3 ms·

And this applies especially to code.

by vrode 16y ago

And this applies especially to code.

lmkg 16y ago
At my college, the "gatekeeper" CS course was notorious, even among non-CS majors. Your code could compile, pass all the unit tests, have optimal algorithmic efficiency and highly optimized runtime... and you could still get a C- because you used inconsistent capitalization for your variable names, or had misspellings in your comments[1]. May the ghost of Dijkstra have mercy on you if you did anything "clever" at the high level. The goal, of course, is not that later on you will spend half your coding time thinking about variable capitalization. The goal is that when you're writing your code, you pick good variable names automatically because you've built up the reflex, by having the idea pounded into your head with the wrath of an angry god. Anecdotally, the class was largely successful in its goal of setting up good habits before the bad ones settled in. After taking that class, people wrote code that was significantly more readable and maintainable than they had written before, and the habits tended to stick. I suspect that at least part of the problems with effective communications is that we develop our early communications largely ad-hoc, and tend to build up a lot of bad habits in the process. Without concentrated effort to correct this habits, they remain and reinforce themselves through repetition. And of course, concentrated effort doesn't scale well. [1] Granted, part of this was because the class was graded on a negative curve. If everyone got 145/150, then 145 is the average, which means a C.
- count 16y ago
  In high school, my senior AP Eng. Comp class was similar. We had to write 2-3 page paper, in class, every day. The papers were all literary analysis (ugh!), but it definitely sharpened the 'writing muscle'. I can write much faster, with more clarity, than most people I know today - all because of that one year of English.