Re: OOP in Mathematica
- To: mathgroup at smc.vnet.net
- Subject: [mg28938] Re: OOP in Mathematica
- From: "Orestis Vantzos" <atelesforos at hotmail.com>
- Date: Sat, 19 May 2001 22:27:55 -0400 (EDT)
- Organization: National Technical University of Athens, Greece
- References: <9e02ol$30e@smc.vnet.net> <9e2bm5$63u@smc.vnet.net>
- Sender: owner-wri-mathgroup at wolfram.com
Wow! OK, I (in general) agree with you on the subject of Maeder's package, but I have to disagree on just about everything else. Mathematica IS a wonderfull language indeed, but when one's code gets bigger than about 100 lines, thinks DO get ugly in terms of reusability, clarity and organisation of code. Packages are the proposed solution, but am I the only one out there that considers them a half-hearted attempt towards OOP? Encapsulation of symbols, both in terms of contexts and packages, is a first (important) step towards OOP (Inheritance being the next). Your preference of Functional over OOP is not relevant here, since the methods of a Mathematica Object (being symbols themselves) would probably be "programmed" via functional programming. It's simply a different level; OOP working on context level and Functional programming working on Symbol level. I am knee-deep now into defining inheritance in context level and hope to have a working beta uploaded in MathSource fairly soon. I expect it to offer all the essentials of OOP, without limiting the user (so no more pure functions as methods!) Thanx for the comments... Orestis Vantzos PS. I respect Maeder for showing yet another direction of development for the Mathematica language. I just think people took an educational package and demanded that it perform as a full application. No one promised it could do so...