Hello,

I’ve just published the third version of my review of named and optional parameters. <https://dpk.gitlab.io/presrfis/arguments.html>

The summary of the changes and updates is:

  - Mention the use of keyword arguments in foreign-function interfaces to languages where keywords are needed.
  - Propose a low-level but unoptimized `lambda-keyword*` form for use in cases, such as foreign-function interfaces, where the performance-necessitated restrictions on `lambda-keyword` require another approach.
  - Discuss the implications of adopting compiler macros and/or identifier syntax for optimization of keyword argument binding.
  - Discuss the approach suggested by Common Scheme and Aaron Hsu.
  - Clarify and extend the definition of ‘as soon as possible’.
  - Explicitly clarify that the RFC 2119 meanings of key words are used.

And you can see a full diff here:
<https://gitlab.com/dpk/presrfis/-/compare/keywords-draft-2...keywords-draft-3#807b342342fc391537165de0134a947b5faaaa99>
(The relevant file is srfi-89/REVIEW.md, itself a legacy filename from the initial stages of this project. You can ignore the other two files — unfortunately GitLab doesn’t seem to let you compare only one file between two Git revisions.)

Further comments are welcome. In the absence of any further feedback, I expect this will be the last revision until the R7RS ballot on syntax-case, identifier-syntax, etc. gets voted on.

Many thanks,

Daphne