[Just responding to bullets where I have particular opinions:] At Mon, 15 Feb 2010 12:13:46 -0500, Neil Van Dyke wrote: > * I'm ambivalent about having a monolithic "racket" or "rico" command. > A small downside is that monolitic does feel more like a closed > platform. A small upside is that it looks (deceptively) friendly in demos.
I think `rico' should have a plug-in interface (registration via "info.rkt"), so that it can better serve as the one-stop executable for all development tasks. > * The suggested automagical games like transparently translating > filename extensions. A variation on this would be to *stop* having > filename extensions in things like require specs. Instead, to turn what > essentially is a file reference into a filename, an ".rkt" would be > appended. As a backward-compatibility measure, ".ss" would be permitted > in the spec, and either (I'm not sure which) would prevent ".rkt" being > added, or would try one extension first and fallback to the other. There are many normalization and transition problems here. Although I don't like automagical conversions, the transition path looks way too painful unless we equate ".ss" and ".rkt" in module paths. For example, a lot of code that refers to collection- and Planet-based libraries using ".ss" forms. _________________________________________________ For list-related administrative tasks: http://list.cs.brown.edu/mailman/listinfo/plt-dev
