CWD and other unresolved issues Lassi Kortela 07 Feb 2020 15:34 UTC

I could also take another look at the draft.

We discussed adding 170 to Gambit with Marc, and he again made the case
that requiring the Scheme CWD to be equal to the OS CWD is not practical
for multi-threaded programs.

Would it make sense to add language to the SRFI saying that
implementations may use pathnames relative to a custom CWD if they want to?