From ... From: Erik Naggum Subject: Re: Lisp is alive, was "Re: Common LISP: The Next Generation" Date: 1996/09/18 Message-ID: <3052030527425472@naggum.no>#1/1 X-Deja-AN: 181339229 sender: erik@arcana.naggum.no references: <842908513snz@wildcard.demon.co.uk> <323f7b48.3843721@news.primenet.com> organization: Naggum Software; +47 2295 0313; http://www.naggum.no newsgroups: comp.lang.lisp [Glenn Ehrlich] | Isn't the semantics more important the syntax? generally speaking, if this were true, we would not see the syntax-fixation that the whole industry is built on. syntax _distinguishes_ programming languages from each other; the semantics is only a question of how much of the syntax you need to implement it. e.g., if CFRONT can produce C code from C++, so can a human. syntax can help automate the expression of the semantics in very important ways, or serve to frustrate the same process. however, certain operations are possible (that is, sufficiently easy) on certain syntaxes, which leads me to the more specific interpretation. one part of the reason why I like Lisp is that I and the compiler can use the same functions to read and write the source code, not only in macros, but in codewalkers and automated editing tasks. `read'ing Java or any of the C-family languages is dreadfully painful. writing out what you have read in is somewhat easier, but not much. | I think one of the really valuable lessons that Dylan has shown is that | you can have a successful infix syntax *and* have all of the Lisp | goodies. _all_ of the Lisp goodies? except macros, except `read' on source code, except `pprint', except ... #\Erik -- those who do not know Lisp are doomed to reimplement it