3.1291 portability of languages (35)

Willard McCarty (MCCARTY@vm.epas.utoronto.ca)
Tue, 10 Apr 90 19:55:12 EDT

Humanist Discussion Group, Vol. 3, No. 1291. Tuesday, 10 Apr 1990.

Date: Tue, 10 Apr 90 10:26:00 EDT
From: DEL2@phoenix.cambridge.ac.uk
Subject: Re: [3.1243 programming languages, cont. (297)]

Anent the debate on portablilty of languages:
Don't forget that whereas a *language* may be protable, any specific
*implementation* may have non-portable elements; so your *program*
may well turn out to be quite non-portable if you use them. I consulted
our 'help' facility about ICON, and although it is a pretty protable
language, note the following comments:

>However, users should be aware that the operating-system
>interface and (especially) the way the package is run both assume a
>UNIX-like operating system (which Phoenix/MVS is not)...
>Consequently, a few of Icon's facilities do not make sense under MVS...
>...Icon's user interface (as a language system) has needed extensive
>modification for Phoenix. While most of the UNIX-style options have
>actually been implemented under Phoenix, they do not all work in quite the
>same way as UNIX programmers would expect (or as documented in the manual)...
>Users should note that Icon's diagnostics are UNIX-like (i.e. very brief
>and with no variable data), and that some ingenuity may be needed to
>deduce their cause...

Caveat emptor!

Douglas de Lacey.