as a matter of curiosity, i know some projects (of...
# general
a
as a matter of curiosity, i know some projects (often gnu projects) tend to keep per-file copyright notices up to date for the current year. i actually appreciate our policy of keeping the copyright date at the year the file was introduced as a low-key form of historical record that persists across renames (which is also useful for copyright purposes). i have also seen (and made the mistake myself) of copying over the heading from another python file when making a new file and accidentally using a year before the current one (case in point -- every one of my PRs this year so far). would it be useful to add a pre-commit check that extends the current file header check to ensure any new files have the correct year?
h
I’ve forgotten to update year I think 50% of the time I’ve added files. All of those times Stu caught it for me 😂 I think this would be nice to have, although not a big priority
a
i think something that can get lost in discussion of priorities is the benefit/effort ratio -- but i'm speaking less about open source projects when i say that. it is true that github highlights added files enough that this isn't a necessity. i just worry that the inconsistency with git history is confusing, but not more than anyone else
👌 1
a
IIRC technically the year is meant to be the first year that novel work was done in the file, and updating the year is actually bad
h
Yes, don’t update years in existing files.
Copyright notices on source files are ~90% BS, since you have copyright whether or not you have such a notice. But if we’re going to have them, then the logic of noting the year is that it establishes the earliest date for which this file represents protected work. So updating it seems just wrong.
a
oh absolutely
hence just making sure that new files have the correct year instead of just checking that it matches YYYY
i’m 100% agreed on not updating them
i will find an example of a project that does this if i have time to look for it but i didn’t bring it up at all because i thought we should do it, and should have made that clear