José,

> To be honest, I'm also thinking about withdrawing because I'm
> not sure to have much time in the next months for it and it may
> wait R7RS-large finalization and more srfi-213 identifier
> properties) implementations.

I personally think that this SRFI should be finalized (after a round of
review and edits, of course—the others raised valuable concerns!) The
hygiene and identifiers problem, I think, should be left to
implementors.

This is, at least, the approach I took with SRFI 253 and 273—the sample
implementation is brittle in that it compares identifiers literally and
translates them to implementation-specific types on e.g. Chicken and
Kawa. So something like `integer?' will always resolve to `integer' on
Kawa, regardless of the current binding of `integer?' in a given library
/ module.

I realize that providing such a sample implementation is bad, but
finalizing an SRFI is more valuable than having a good SRFI withdrawn
purely out of implementation concerns. Finalized SRFIs get implemented
and improved. Withdrawn—never even implemented.

So I advise you to

• Do a round of edits for issues raised by others

• And amend the SRFI document (and Implementation section in particular)
  with a note on this issue

That should be enough for finalization, I believe.

Best of luck,
--
Artyom Bologov
https://aartaka.me