[r6rs-discuss] [ANN] scheme-reports.org

[Pavel Dudrenov]

On Mon, Aug 24, 2009 at 1:21 PM, Thomas Lord<lord at emf.net> wrote:
> On Mon, 2009-08-24 at 12:56 -0700, Pavel Dudrenov wrote:
>
>> I'm all for having standard module system in small scheme. I'm also
>> all for keeping that module system with small and simple public
>> interface.
>
> A SRFI compatible with small scheme can provide that.
>
> Back in the day, when it was uphill to school in both
> directions, people would sometimes write "modules"
> like:
>
> (define export1 '())
> (define export2 '())
> ...
> (letrec* ( ...module definition ...)
>  (set! export1 ...)
>  (set! export2 ...)
>  ...)
>
> If you syntactically abstract that a bit
> you get a perfectly nice standard for a module
> system for small scheme.
True, but I think that it would be more beneficial for implementations
to provide a module system. But not required like you said. One of the
main benefits, apart from leaving implementations to represent modules
as they feel free internally, is that people would have a standard
minimalistic module system already provided, so, hopefully, no one
will have the urge to write their own module system as to better split
code for projects they are working on.
>
>
> (John C.: This half-answers (but enough?) your
> request to flesh out my claim that a module system
> can be just syntax.)
>
> -t
>
>
>