If I had a nickel for every bad boy programmer whose code I've had to clean up...
I find the people who are the worst about playing fast and loose with pointers are contractors. New grads out of school might get careless, but if you catch them and reminds them they usually straighten up. Contractors don't have to maintain the code they write; they just want to get paid as quickly as possible so they can move on to the next job.
My quibble with this comic is that good memory management practices are so fundamental that it shouldn't even need to be in the coding standard. That and with modern C++ a failed call to new should throw so there's no need to explicitly check.
Those guys aren't the ones that get me. What gets me are the people who think they're doing it right, but they're using pointers when they have absolutely no need. They have no clue what a const & is.
The ironic part is that this is so right, that even I see it as a slice of life. And I'm a hardware engineer who never touches code at work. Not just a hardware engineer, an analog/RF hardware engineer - you can't get more removed from SW than I am.
I am not telling you how many of these cartoons I have looked at this evening. But I went back to the first one and started forward. Every once in a while I have to stop and mail one out to someone.
If I had a nickel for every bad boy programmer whose code I've had to clean up...
ReplyDeleteI find the people who are the worst about playing fast and loose with pointers are contractors. New grads out of school might get careless, but if you catch them and reminds them they usually straighten up. Contractors don't have to maintain the code they write; they just want to get paid as quickly as possible so they can move on to the next job.
+1 on Dave, and when you DO catch them, they want to be paid for fixing their own screwups... sigh
ReplyDeleteMy quibble with this comic is that good memory management practices are so fundamental that it shouldn't even need to be in the coding standard. That and with modern C++ a failed call to new should throw so there's no need to explicitly check.
ReplyDeleteThose guys aren't the ones that get me. What gets me are the people who think they're doing it right, but they're using pointers when they have absolutely no need. They have no clue what a const & is.
The ironic part is that this is so right, that even I see it as a slice of life. And I'm a hardware engineer who never touches code at work. Not just a hardware engineer, an analog/RF hardware engineer - you can't get more removed from SW than I am.
ReplyDeleteThat's how bad it is.
I am not telling you how many of these cartoons I have looked at this evening. But I went back to the first one and started forward. Every once in a while I have to stop and mail one out to someone.
ReplyDeleteThanks for a fun link.