From: Thomas Lord <[email protected]> Subject: Re: [r6rs-discuss] [ANN] scheme-reports.org Date: Mon, 24 Aug 2009 13:39:01 -0700
> On Mon, 2009-08-24 at 13:33 -0700, Pavel Dudrenov wrote: > > 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. > > You can't grab power over future programmers > that way. Don't even bother trying. And you > shouldn't, anyway - it's rude. > > If small scheme has some anemic module system, > don't act too surprised when people start writing > their own replacements anyway. > > Just give people the essence which, in this matter, > is just enough so that module system_s_ can be defined > in SRFIs. > > We're already committed to some form of syntactic > abstraction and lambda: there is nothing more to > add to the *core* vis a vis modules. I'm all for counting on SRFIs for fancy module features, but I'd like to point out that the syntactic abstraction of module systems is not only for users' convenience, but it is important to leave room for the implementations to treat imported identifiers differently than mere symbols. It would leave the door open to "plug-in" hygienic macros; if we bind the simple module system to be realized as letrec*-and-set! approach, I'm afraid that hygienic macro systems on top of it would require preprocessing whole program. --shiro _______________________________________________ r6rs-discuss mailing list [email protected] http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
