I'm implementing this SRFI and have noticed an ambiguity in the
specification.

string-unfold-right specifies that the results of the mapper procedure
are accumulated from right to left in the result.  When the mapper
returns a character, that's not a problem, and the specification is
clear.

However, when the mapper returns a string, it's not specified how the
characters of the string are accumulated.  One might imagine the
characters being prepended to the result one at a time in reverse order,
as if it was returned by multiple calls to the mapper in the string
order, or the string might be simply prepended to the output being
built.

I see that Kawa implements the latter.

I don't care what the actual semantics are, as long as they are clearly
specified.

I suppose it's worth spelling out for string-unfold too, though in that
case it seems like the obvious solution is to just append the string,
since the two orders are congruent.

Another concern would be whether someone expects that string-unfold and
string-unfold-right have some kind of deep symmetry, in which the
resulting strings are the reverse of one another, even when some mapper
values are strings.  In that case it might be more elegant to prepend
the string's characters in reverse order rather forward order.  I'm not
making that argument, though I can understand the appeal.  I just want a
clear specification.