On Mon, Nov 3, 2014 at 8:03 PM, Matthias Felleisen <matth...@ccs.neu.edu> wrote:
> > What this really shows is how bad the choice of load was. I would bet a > beer or something that you'll find #lang/require and friend will eventually > make your life easier and happier. > > > Ouch. I'm going to need that beer sooner rather than later :p This has actually become one of our biggest pain points. I'm not at all sure that #lang/require can do what we want. Conceptually it seems fairly clear, but pragmatically I haven't been able to find the right approach. Essentially what 'load' did and continues to do is let us close the last 10%-20% functionality gap in our tool in a quick and straightforward manner (the term 'hack' does come to mind); mainly by allowing those arbitrary side effects to be collected as necessary. -- Dan Liebgold [dan.liebg...@gmail.com]
_________________________ Racket Developers list: http://lists.racket-lang.org/dev
