Re: Changing the name John David Stone 20 Feb 2002 16:18 UTC

        Michael Sperber writes:

 > So why can't we fix that by a perhaps improved explanatory paragraph
 > in the SRFI?  I have a hard time seeing this specific "incident" as
 > enough reason to warrant the change.

        It's not this specific incident that bothers me.  I'm fretting
about the dozens, hundreds, or thousands of future incidents of the same
general sort -- not just questions in comp.lang.scheme, but questions in
classrooms.  The fact that we're starting to get them only two weeks into
the proposal process for SRFI-26 is significant because it indicates that
such questions are going to be annoyingly frequent in the future.

        Imagine the apologetic little postlude that teachers like me are
going to be adding to our lectures next year when we explain currying to
Scheme-speaking students: ``So that's how currying works.  And, oh yes,
some of you may also be acquainted with CURRY-expressions in Scheme.  Those
implement something different, the operation that we've been calling
`sectioning'.  A procedure that really curries binary procedures looks like
this in Scheme:

           (define curry
             (lambda (procedure)
               (lambda (first-argument)
                 (lambda (second-argument)
                   (procedure first-argument second-argument)))))

CURRY-expressions in Scheme use the keyword CURRY because the people who
proposed and adopted the construction couldn't agree on a better name for
it.  Yes, they all know that CURRY-expressions don't do currying -- if you
read the original SRFI on the subject, you'll see that the author says so
explicitly.  Yes, they went ahead and used the keyword CURRY anyway.  Yes,
I know it's confusing, but it's just one of those things.''

 > <joke>You can't seriously be advocating a change with the intention of
 > making less people see the SRFI, right?</joke>

        I want fewer of the people _who are trying to figure out how
currying works_ to see this SRFI.  Using the name SECTION might conceivably
attract more people who are trying to figure out how sectioning works; that
would be another advantage.

--
   John David Stone - Lecturer in Computer Science and Philosophy
           Manager of the Mathematics Local-Area Network
           Grinnell College - Grinnell, Iowa 50112 - USA
     xxxxxx@cs.grinnell.edu - http://www.cs.grinnell.edu/~stone/