On Thu, Oct 15, 2009 at 07:26:04PM -0400, John Cowan wrote:
> full-unicode
>         All Unicode characters are supported

Since this is about Thing One, probably there ought to be
a "unicode", and maybe an "ascii" too, to distinguish between
a Scheme that has no unicode support at all and one that has
rudimentary unicode support (whatever that means).

> i386, x86_64, ppc, sparc, sparc64, ...
>         CPU-specific flags.
> 
> ilp32, lp64, ilp64, ...
>         C-memory-model-specific flags.

These sound useful, but are they available on, say a system which
runs on top of Java or another VM?  Or would that VM _be_ the "CPU"?

Also, I propose to add big-endian and little-endian to that list.
The precise CPU architecture isn't usually that important, except
when generating native code/assembly code.  I wouldn't want to list
all CPU types exhaustively when I'm only trying to determine the memory
byte order.

> <name>
>         The name of this implementation.
> 
> <name>-<version>
>         The name and version of this implementation.

How useful is this?  I've seen code that checks if a certain minimum
version of an implementation is in use, but the precise version isn't
*that* useful in cond-expand (except maybe in rare cases where only
that version is known to contain a critical bug which needs to be
worked around).

The rest of the list seems quite useful and sensible. Thanks John, for
drafting this proposal!

Cheers,
Peter
-- 
http://sjamaan.ath.cx
--
"The process of preparing programs for a digital computer
 is especially attractive, not only because it can be economically
 and scientifically rewarding, but also because it can be an aesthetic
 experience much like composing poetry or music."
                                                        -- Donald Knuth

Attachment: pgpLDvtHyRTTI.pgp
Description: PGP signature

_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss

Reply via email to