On Jan 23, 2008 9:12 PM, Jonathan S. Shapiro <[EMAIL PROTECTED]> wrote: > On Wed, 2008-01-23 at 21:01 +0100, Eric Rannaud wrote: > > Hi, > > > > On Jan 23, 2008 5:21 PM, Jonathan S. Shapiro <[EMAIL PROTECTED]> wrote: > > > As we contemplate an s-block syntax for BitC, one question that emerges > > > is initialization rules. In particular, BitC really wants to require > > > that pointer slots (a) be initialized, and (b) be non-null. > > > > Out of curiosity, and even tough it was already suggested in BitC's > > specifications, I never quite understood why you would want to move to > > an s-block syntax: is there a particularly compelling reason? > > Yes. Human factors. The desire to deploy this language in the real world > basically requires ditching the current syntax, and our original reason > for introducing it no longer applies.
Well, you know what they say: "languages designed for other people to use have been bad [...] the good languages have been those that were designed for their own creators"... http://www.paulgraham.com/javacover.html > > Isn't loosing the ability to have real macros somewhat on the down side? > > Oh no. Permanently eradicating the remotest possibility that some bright > soul might contrive to introduce a macro system is one of the few > genuinely positive points about shifting to an s-block syntax. In the > presence of a macro expansion system, human code inspection is > hopelessly compromised. Hmm, unless the sometimes real need for macros is resolved with copy-and-paste... But that's your call, obviously. Eric. _______________________________________________ bitc-dev mailing list [email protected] http://www.coyotos.org/mailman/listinfo/bitc-dev
