On Fri, 04 Sep 2009 13:14:35 -0400, John Cowan <[email protected]> wrote:
> Ben Goetter scripsit:
>> John Cowan wrote:
>> >
>> >R4 is very nearly dead; the implementations that only
>> >support R4RS are pretty much unmaintained.
>> Stalin is unrepentently R4RS, and it is still maintained. Last release
>> 2 Oct 2006.
>
> Three years is a long time in Internet time. And Stalin is manifestly
> a dead end: it does what it does, but it's a nightmare to maintain for
> anyone but the author, especially given that the internal documentation
> consists of "LaHaShem HaAretz U'Mloah" at the beginning and "Tam
> V'Nishlam
> Shevah L'El Borei Olam" at the end of each source file. :-)
>
> Still, someone could provide Stalin with a preprocessor that did
> syntax-rules and added support for (probably) everything in R5RS except
> dynamic-wind. (Multiple values would have to cheat, as is the case in
> chibi-scheme, where values returns a list of values with a unique object
> in the car, and call-with-values merely checks for this.)
An attitude like this is very detrimental to commercial Schemes which need
to emphasize backwards compatibility. There is no reason to break
compatibility with older systems in Scheme, since not that much has really
changed. We've added things, sure, but it should still be possible to
express that a program is R4RS instead of R5RS, IMO. Using the right set
of modular libraries, I bet we could also allow differing semantics to
some extent for backwards compatibility sake, though I don't see any
reason to do this since nearly everything is R5RS now, and there are
precious few programs left in the Scheme world relying on, say, nil == #f.
Aaron W. Hsu
--
Of all tyrannies, a tyranny sincerely exercised for the good of its
victims may be the most oppressive. -- C. S. Lewis
_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
