If you have "troubleshooters" like Bill putting garbage into your source repository, then it's YOUR fault.
Do you have any standards, procedures, or conventions? Do you QA the source code, or is User Acceptance Testing enough?
If anyone working for me tried to squeeze hard coded error codes by me, I'd keep sending them back until either: a. It was done properly according to standards, b. They quit because they couldn't be "creative", or c. I'd fire them.
Sorry for being a little testy today. I just replaced a 92 line internal bubble sort with 3 lines of code. Of course, I didn't realize it was a bubble sort until I resolved 23 variables named "a", "aa", "aaa", "b", "bb", etc.
You see, I'm the poor schmuck that has to clean up your "best people's" mess.
I think it's important not to overgeneralize this argument though.
The example works because it's very specific to a particular way of programming. Extend that insulated idea to everyone, and you won't have much of a company.
I think this was actually a part of someone's management thinking for a while. I think I remember Nicholas Taleb mentioning something about this in Fooled by Randomness (not the programming part, just about management thinking that advocated firing your best people and promoting the lazy ones)
If you have "troubleshooters" like Bill putting garbage into your source repository, then it's YOUR fault.
Do you have any standards, procedures, or conventions? Do you QA the source code, or is User Acceptance Testing enough?
If anyone working for me tried to squeeze hard coded error codes by me, I'd keep sending them back until either: a. It was done properly according to standards, b. They quit because they couldn't be "creative", or c. I'd fire them.
Sorry for being a little testy today. I just replaced a 92 line internal bubble sort with 3 lines of code. Of course, I didn't realize it was a bubble sort until I resolved 23 variables named "a", "aa", "aaa", "b", "bb", etc.
You see, I'm the poor schmuck that has to clean up your "best people's" mess.