On 2020-07-16 20:14 -0700, Arthur A. Gleckler wrote:
> > Sounds good.  If Arthur will accept it, I'll send a PR removing them.
>
> Since both authors were surprised by the behavior under these
> circumstances, it definitely counts as an error, so I have to say yes to an
> erratum.

Thanks.  Pull request opened.

Essentially, until commit 851ae058, the working specification for
maybe-let* and either-let* was little more than "the Maybe/Either
equivalent of SRFI 2 and-let*", so the implementation of these forms
followed SRFI 2 in all its oddities.  John then added cleaner
semantics for these forms, but I unfortunately assumed that nothing
significant had changed.  Hopefully everything is satisfactory now.

--
Wolfgang Corcoran-Mathe  <xxxxxx@sigwinch.xyz>

"In the military more is not better." --_Sun Tzu_