[r6rs-discuss] What's up with the library names?
On May 24, 2007, at 12:22 PM, Ludovic Court?s wrote:
> Hi,
>
> "R. Kent Dybvig" <dyb at cs.indiana.edu> writes:
>
>> I believe that the current proposal actually gives the r7rs
>> editors more
>> rather than less flexibility.
>
> Perhaps (although I'm not quite convinced).
>
> At any rate, it does seem harmful for a program to not specify which
> version of the report it is targeting since it may eventually break in
> unexpectable ways when the next report comes out.
Right. And since it's far more convenient to write (rnrs)
than it is to write (rnrs (6)), people are going to do that
by default (I would). Next thing that happens is that r7rs
comes out and any library you download would have an import
of (rnrs) which says nothing about what they target. So,
you start the trial and error process (which is what r6rs
libraries were supposed to handle automatically). If, on
the other hand, a library imports (r6rs) and your system
complains about "(r6rs) library not found", then you would
at least get a clue about where to look. And if you do
change the import from (r6rs) to (r7rs), you would at least
do that with caution instead of having your import bomb by
being automatically upgraded.
In short, I view (r6rs) as the good, (rnrs) as the bad, and
(rnrs (6)) as the ugly.
Aziz,,,
Received on Thu May 24 2007 - 12:50:18 UTC
This archive was generated by hypermail 2.3.0
: Wed Oct 23 2024 - 09:15:01 UTC