[r6rs-discuss] Temerity

[David Rush]

2009/8/31 Joe Marshall <jmarshall at alum.mit.edu>:
> On Sat, Aug 29, 2009 at 1:10 PM, Aubrey Jaffer<agj at alum.mit.edu> wrote:
>> Your classification adds no objective information to the table,
>
> Admittedly it doesn't add much, but there is no point in pretending that
> all implementations are equal.  PLT Scheme is clearly a more important
> implementation than VSCM.

By what measure? There is some sense in which I am the custodian of
the VSCM source code (having been blessed by Matthias Blume to be
allowed post it on SourceForge many moons ago). It has lovely
internals, and IIRC it was also a pretty complete R4RS (which was and
still is rarer than one might think) - but yes it is also moribund.

It surely looks like you're metric is simply: "Joe Marshall's
favorites". And it also looks like an opening shot in the pissing
match over the features of LargeScheme. Is there anything actually
*measured* by this "metric"?

For what it's worth, My "Metric A" would have Gambit, Larceny, Stalin,
and maybe Chicken in the top tier. All of those Implementations have
made *significant* technical contributions to understanding the
essence of the language and its implementation. If you wish to include
moribund Implementations then you;d have to include Rabbit and T. This
is opposed to PLT, which, IMO has made very large strides in
demonstrating the *application* of Scheme's principles to practical
programming (an exercise any competent engineer should be able to
perform).

david rush
-- 
GPG Public key at http://cyber-rush.org/drr/gpg-public-key.txt