Re: Rationale question Michael Sperber 10 Jun 2005 06:05 UTC

Sebastian Egner <xxxxxx@philips.com> writes:

> Mike Sperber wrote:
>> I'm not sure what VALUES->LIST is supposed to be---given the symmetry
>> between return values and arguments, I'd say it'd have to be
>>
>> (define (values->list . vals)
>>   vals)
>>
>> This procedure is called LIST in R5RS, so LIST-VALUES isn't a name
>> symmetric with anything.  UNLIST seems fine.
>>
>> Or am I missing something?
>
> I think you do. Please refer to the previous incarnation of this
> discussion:
>
>         http://srfi.schemers.org/srfi-71/mail-archive/msg00015.html
>         http://srfi.schemers.org/srfi-71/mail-archive/msg00019.html
>
> Aparently there is some comment missing in the SRFI text.

Oops, sorry, that must have gone right past me.

However, I think VALUES->LIST is a pretty unnatural name for your
construct.  (Or the construct an unnatural fit for the name---take
your pick.)  So I think adding the definition wouldn't really help
clarifying the rationale.

Given that there's now been two of us who are confused by your
argument for the *other* side, my vote is for eliding that paragraph.
(Not that it's especially important, so I'll shut up now on this.)

--
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla