I think this is a clear-cut case where keywords would be better than inferring which argument is which by its type (and then falling back on order after all to discriminate the two strings).  That is not to say that keywords are the *best* solution.

On Sat, Mar 7, 2020 at 10:58 AM Arthur A. Gleckler <xxxxxx@speechcode.com> wrote:
On Sat, Mar 7, 2020 at 6:54 AM John Cowan <xxxxxx@ccil.org> wrote:
The publication process for this SRFI and the adjacent ones was very long, so they reflect ideas from much earlier in SRFI history.  Arthur will have the exact dates, of course. 

Joo considered publishing a version of this SRFI in late 2017, but ultimately decided against it then.  The first draft of SRFI 159: Combinator Formatting was published just a few months before that.

My impression is that the ideas are entirely independent.