The `sox.1' man page introduces the term `sample size', but then goes on to use `bit-depth' in a few places. This commit attempts to apply the term `sample size' uniformly across the project, including in the code of `src/stats.c'.
I'm not sure that is the right way to go. Even though sample size and bit depth are related concepts, they are not always interchangeable; I think sample size refers more to a "container" that has a maximum capacity, but doesn't need to be completely filled. Bit depth can mean both the container and the fill level.
In any case, even if "sample size" was the better term in some places, it is not advisable to change the output of "stats" that is often parsed by scripts, nor is it appropriate to reword past changelog entries. (And apart from that, I'm quite convinced it should be bit depth in stats.)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The terms are largely synonymous, except when they are not. I do not believe the current wording is confusing. Some of the changes in the patch are, however, definitely incorrect.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Apply with `git am'
I'm not sure that is the right way to go. Even though sample size and bit depth are related concepts, they are not always interchangeable; I think sample size refers more to a "container" that has a maximum capacity, but doesn't need to be completely filled. Bit depth can mean both the container and the fill level.
In any case, even if "sample size" was the better term in some places, it is not advisable to change the output of "stats" that is often parsed by scripts, nor is it appropriate to reword past changelog entries. (And apart from that, I'm quite convinced it should be bit depth in stats.)
The terms are largely synonymous, except when they are not. I do not believe the current wording is confusing. Some of the changes in the patch are, however, definitely incorrect.