If you've used a command for the past 10[0] years, and it's always worked, and you go to use it again, and you're using it in a similar way to how you always have, do you read the man page?
No, of course you don't. You expect things to behave the way they always have.
Then when something goes wrong you don't immediately suspect the command you've successfully been using for the past 10 years and read the man page for all the commands you've been using for the past 10[1] years.
"If you've used a command for the past 10[0] years, and it's always worked, and you go to use it again, and you're using it in a similar way to how you always have, do you read the man page?"
I'm trying to find where the cut-off point might be.
Do you read the man page every time you use "ls"?
Do you read the man page every time you invoke the compiler for your favourite language?
If you use someone else's machine, do you read the man page for every command you type? The distro could be different, so you can't be sure it will be the same.
I suspect the answer here in each case is "no", so I'm curious as to where you draw the line. If you used a command yesterday, and the day before and the day before that, do you read the man page today?
Again, I suspect the answer is "no", but again, I'm curious as to where that line really is.
No, of course you don't. You expect things to behave the way they always have.
Then when something goes wrong you don't immediately suspect the command you've successfully been using for the past 10 years and read the man page for all the commands you've been using for the past 10[1] years.
No, you suspect the new code.
[0] For some value of 10[1].
[1] In my case significantly longer than 10.