It also makes sense to have a condition type &errno, which just
packages the errno symbol.

Am Mo., 17. Aug. 2020 um 16:48 Uhr schrieb Lassi Kortela <xxxxxx@lassi.io>:

> We could have &foreign (a subtype of &condition) and &foreign-error (a
> subtype of &error) separately, and `foreign-status?` would test for
> either of those. It's a bit iffy, but could be worse?

We also have the I/O errors in section 8.1 of the R6RS Standard
Library report. I guess it makes sense to do some merging here as
well.