On 16 Aug 2010, at 17:14, Truls Becken wrote: > On 2010-08-16, at 13:59, Jan-Paul Bultmann wrote: > >> Nu seems pretty awesome, and this >> (http://programming.nu/files/DeliciousFunWithNu.mov) seems just perfekt for >> etoile. > > Nu looks cool, but it is an interpreted language, and code written in Nu is > therefore likely to be much slower than code written in Pragmatic Smalltalk.
Nu is very similar to the ObjLisp concept that someone (alex?) implemented a few years ago on top of the old Étoilé runtime. Unfortunately, it was the only language ever implemented on top of that runtime, so it didn't get to benefit from any interoperability and was never maintained. Implementing something like Nu would be possible with LanguageKit though. > I actually did think about implementing a Nu parser for Language Kit, but > concluded that it is not possible. The reason is that with an expression such > as "(document library)" there is no way to know at compile time whether > "document" is a function with one argument or an object receiving the message > "library". Nu does this at runtime, by asking document to > evalWithArguments:context: every time that piece of the program is executed. > Objects and Blocks then implement the method differently. That's not necessarily a problem. For example, we could always implement evalWithArguments:context: in categories on NSBlock, BlockClosure, and NSObject to get the same effect, although it would still come with a performance penalty. We could maybe do some static analysis or use type feedback to try to optimise it, or maybe just insert a compile-time test on each Nu expression to see whether the receiver is function-object. The nicest thing about Lisp-like languages is that the S-expression representation makes source-to-source transforms easy. If I (or someone else?) ever get around to finishing the OMeta implementation in svn, we get something similar because we can implement AST transforms in terms of pattern matching, which is even more flexible (and, potentially, language-agnostic). > Maybe somebody else has something more favourable to say? Perhaps with actual > experience using Nu + GNUstep to back it up? I never actually used Nu - it's one of those things that appeals to me a lot more in theory than in practice. David -- Send from my Jacquard Loom _______________________________________________ Etoile-discuss mailing list [email protected] https://mail.gna.org/listinfo/etoile-discuss
