[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [r6rs-discuss] [Scheme-reports] [scheme-reports] Scheme pattern matching & R*RS

Andre van Tonder scripsit:

> I don't like overriding.  

Then you would prefer + for integer addition and +. for float addition,
as in Bliss?

> The meaning of any code snippet becomes highly contextual and 
> nonlocal.  

Any language that allows local syntax extension can have no complaints
to make about contextual meaning.  I personally would never write a
let-syntax, but I don't urge its removal from Scheme.

> the failure of object-oriented languages to deliver on their initial
> promises in the industry.

$DEITY couldn't have designed a language, or anything else, that justified
that level of hype.

> There is a long, and in my experience excellent, tradition in Scheme
> of not using names in libraries that clash with core names.

That is not a tradition, but rather a necessity, at least for portable
code, since standard Scheme had only one namespace until R6RS.

In addition, it sort of undercuts the position of Schemers against
Common Lisp that CL doesn't allow redefinition of core names at all
(which is one of the features that mitigates the failure of CL to provide
hygienic macros).

> If a user disagrees with this and wants to override in his own code,
> that is fine.  Library renaming provides a mechanism for it.  But I
> do not think this failed philosophy  should be forced onto us by
> a standardization.

It's not forced, just the default.  If you want, you can do simple
prefix renaming and get back match-lambda etc.  If you refer to the idea
of making lambda pattern-matching by default in "R8RS", I don't
advocate that.

                Si hoc legere scis, nimium eruditionis habes.

r6rs-discuss mailing list