Path: utzoo!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!cs.utexas.edu!rutgers!aramis.rutgers.edu!paul.rutgers.edu!gaynor From: gaynor@paul.rutgers.edu (Silver) Newsgroups: comp.lang.misc Subject: Re: The Universal Language Message-ID: Date: 8 Aug 90 08:23:44 GMT References: <24013@megaron.cs.arizona.edu> Organization: Rutgers Univ., New Brunswick, N.J. Lines: 29 Me writes: > Any `universal language' must be very small but very versatile. The > components of the language itself should be objects in the language and easy > to modify. gudeman@cs.arizona.edu (David Gudeman) responds: > This is an approach that I favor, but frankly it has been tried and has not > born out its promise. By all means keep trying, but I'm a little > discouraged... Agreed. The simple kernel/language embellished (even supplanted) by a rich set of libraries/packages is the model of choice. Keep the base language simple, and expand it as necessary with libraries and packages. For instance, there's been mention of natural languages in this discussion. They should be implemented in the universal language in such a manner as to embellish the language, but they should not be part of the language's definition -- they're just too darned expensive. I could continue with a wish list of the features that the Ultimate Language should have in mine eyes, but such discussions often turn into jihads and tend to be nonproductive. (Wanna have one anyway?, hee hee!) We could continue kicking around the same old arguments and discussions. They'd benefit muchly those who haven't participated in them before, and perhaps some new and germane issues would be discussed. But is it possible that we could end up with an acceptable Universal Language? Not without a *lot* of concerted effort. It sure would be a pretty sight, though! Regards, [Ag]