3 ms·

I'm sure if I took the time to read them all, they'd all be very different as everyone has their own opinions on "things every programmer should know". IMO wha

by veb 7y ago

I'm sure if I took the time to read them all, they'd all be very different as everyone has their own opinions on "things every programmer should know". IMO what every programmer should know is how to do their job :)

MaxBarraclough 7y ago
> IMO what every programmer should know is how to do their job :) That's either a non-answer, or it's setting the bar awfully low. Storing passwords unhashed may 'get the job done', but every programmer should know better.
- barbs 7y ago
  Maybe it should be changed to "know how to do their job well"
  - test6554 7y ago
    No, that’s a different list. What every decent programmer should know.
    - nkassis 7y ago
      How about the theoretical minimum one should know to be a professional programmer. Basic security principles would be included.
- larnmar 7y ago
  > Storing passwords unhashed may 'get the job done', but every programmer should know better. Not every programmer has to deal with passwords, though, so the generality is it is limited. A programmer who has spent his whole career writing computational fluid dynamics simulations doesn’t need to know anything about storing passwords, but probably would have an equally idiosyncratic idea of what all programmers should know.
- paxys 7y ago
  I would argue that storing passwords insecurely is not "getting the job done". Following minimum industry standards can be considered a requirement of knowing how to do your job.
onion2k 7y ago
I reckon there must be a few universal truths in programming. Things like "Write testable code" and "Document why your code works." spring to mind immediately but I bet there's more.
- jedberg 7y ago
  > Things like "Write testable code" and "Document why your code works." There are many successful companies that didn't do either of those things in their codebase. Whether or not you consider the people who wrote it good programmers is I suppose up for debate, but objectively, they created code that achieved the stated business purpose. My point being, I don't think there are any universal truths.
  - onion2k 7y ago
    There are many successful companies that didn't do either of those things in their codebase. Commercial success is a very poor measure of code quality.
    - jedberg 7y ago
      Like I said, the ability of the programmer is up for debate, but objectively, they created code that solved the problem.
    - MaulingMonkey 7y ago
      Code quality is a poor measure of product success, commercial or otherwise.
    - UncleMeat 7y ago
      Very few people are paid to write quality code. Instead people are paid to produce commercial success.
ergothus 7y ago
> IMO what every programmer should know is how to do their job :) In truth, though, we don't. And not "oh, a few good ones do, but the masses we pity and mock don't". All of us. The only difference is if we acknowledge it. We're trying (and that's why articles like this are so common!), and arguably we as an industry are making progress, but we don't yet have it figured out. https://xkcd.com/2030/ https://xkcd.com/2030/

by veb 7y ago

MaxBarraclough 7y ago

> IMO what every programmer should know is how to do their job :) That's either a non-answer, or it's setting the bar awfully low. Storing passwords unhashed may 'get the job done', but every programmer should know better.

barbs 7y ago
Maybe it should be changed to "know how to do their job well"
- test6554 7y ago
  No, that’s a different list. What every decent programmer should know.
  - nkassis 7y ago
    How about the theoretical minimum one should know to be a professional programmer. Basic security principles would be included.
larnmar 7y ago
> Storing passwords unhashed may 'get the job done', but every programmer should know better. Not every programmer has to deal with passwords, though, so the generality is it is limited. A programmer who has spent his whole career writing computational fluid dynamics simulations doesn’t need to know anything about storing passwords, but probably would have an equally idiosyncratic idea of what all programmers should know.
paxys 7y ago
I would argue that storing passwords insecurely is not "getting the job done". Following minimum industry standards can be considered a requirement of knowing how to do your job.

onion2k 7y ago

I reckon there must be a few universal truths in programming. Things like "Write testable code" and "Document why your code works." spring to mind immediately but I bet there's more.

jedberg 7y ago
> Things like "Write testable code" and "Document why your code works." There are many successful companies that didn't do either of those things in their codebase. Whether or not you consider the people who wrote it good programmers is I suppose up for debate, but objectively, they created code that achieved the stated business purpose. My point being, I don't think there are any universal truths.
- onion2k 7y ago
  There are many successful companies that didn't do either of those things in their codebase. Commercial success is a very poor measure of code quality.
  - jedberg 7y ago
    Like I said, the ability of the programmer is up for debate, but objectively, they created code that solved the problem.
  - MaulingMonkey 7y ago
    Code quality is a poor measure of product success, commercial or otherwise.
  - UncleMeat 7y ago
    Very few people are paid to write quality code. Instead people are paid to produce commercial success.

ergothus 7y ago

> IMO what every programmer should know is how to do their job :) In truth, though, we don't. And not "oh, a few good ones do, but the masses we pity and mock don't". All of us. The only difference is if we acknowledge it. We're trying (and that's why articles like this are so common!), and arguably we as an industry are making progress, but we don't yet have it figured out. https://xkcd.com/2030/ https://xkcd.com/2030/