MathGroup Archive 2005

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Types in Mathematica

John Doty wrote:

> No, we're only suggesting that you don't take their analogy too far.
> Rather, match the capabilities of Mathematica to to the needs of your
> application without injecting irrelevant ideology.

Analogy? Who said anything about analogy?  I was talking about how these
authors describe Mathematica directly.  I will grant that Dr. Trott has not
made much use of terms such as "data type" when discussing Mathematica. 
Nonetheless, he has discussed the behavior of expressions which differ
according to the characteristics (i.e., types) of their arguments. Dr.
Wolfram does however specifically state that "[a]ll expressions in
Mathematica are ultimately made up from a small number of basic or atomic
types of objects."  And Dr. Mäder says much about the implementation of
abstract data types in Mathematica.

> Sure. Forth software development is a good model for Mathematica
> software development (and *very* different from C++ development), but
> the languages themselves are utterly different, so formal concepts in
> one map not at all into formal concepts in the other. Still, the process
> of development "feels" very similar, and experience in one is therefore
> useful for the other. Just don't go looking for a parameter stack in
> Mathematica or DownValues in Forth. Use the tool, don't fight with it.

Fine.  So what about the idea of expressing data types as I had been
describing before we took this tangent?  Did anybody even bother to try and
understand what I was saying?  The only thing I said regarding C++ was that
what I was describing is not something I would do in C++.
The Mathematica Wiki:
Math for Comp Sci
Math for the WWW:

  • Prev by Date: Re: Confusing results with N[expr]?
  • Next by Date: Re: Confusing results with N[expr]?
  • Previous by thread: Re: Types in Mathematica
  • Next by thread: Re: Types in Mathematica