My father liked to say that you couldn't call yourself educated unless
you understood quantum mechanics. I have heard the same said about
Latin, and first-order logic, and Hamlet, and DNA. White would add Lisp
to the list.
Fine.
But I don't see what that has to do with Pissarenko's choice of Java,
which seems perfectly sensible. What's more, it's foolish to tell
people who are interested in Maxima that they should only consider using
it if they're willing to convert to Lisp for all their programming.
About the substance of Pissarenko's arguments:
I agree with him completely that it is important to be able to build
systems out of heterogeneous components. If a project needs both
symbolic manipulation and a DBMS, should we wait for a good-quality DBMS
to be written in Lisp, or interface to good existing tools? If a
programmer is familiar with GUI-building in Java (or Visual Basic or
tcl/tk or whatever), it makes perfect sense for him to use that tool.
Oh, and by the way, if it's so easy to build GUIs in Lisp, why doesn't
Maxima have a built-in Lisp-based GUI?
> Not relevant for a project with a small audience -- it is
> possible that 97.5% of all already use LISP.
Hardly seems likely.
> > *) larger market share due to platform independence
> ...Not if your project is too big to
> run on the majority of Java platforms (PDA's, etc.).
Huh? You are confused. The majority of Java platforms are probably
smart cards, just as the majority of electric motors is probably in
wristwatches. But that is utterly irrelevant.
-s