Re: Is Ada likely to survive ?

[rav@goanna.cs.rmit.edu.au (robin)]

ok@goanna.cs.rmit.edu.au (Richard A. O'Keefe) writes:

	>rav@goanna.cs.rmit.edu.au (robin) writes:
	>>It's one of the enduring success stories.  The features
	>>in PL/I were so advanced that the language didn't need continual
	>>updating over the years (like some other languages did !).

	>Maybe it didn't _need_ it, but PL/I has certainly _received_ quite
	>a bit of updating.

Please read what I wrote.  I said that it dodn't need continual updating.

The reason that it didn't need continual updating is that it
had -- pretty well from the beginning (except for list processing,
which was added a short time later) --

@ list processing;
@ dynamic arrays;
@ controlled arrays (i.e. user-controlled allocation)
@ commercial I/O (i.e., formatted I/O)
@ direct file access (i.e., random access);
@ data-directed I/O;
@ macro pre-processor;
@ error-handling;

Many of these were important / significant new features, and were new
to any general-purpose language at the time [e.g. list processing,
controlled storage allocation, direct access files, macro-processor).

It was such forward-thinking features as these that made the language great.

They also helped PL/I from becoming obsolete.

	>  Compare modern OS/2 PL/I with the 1976 ANSI PL/I
	>standard.

The ANSI standard was largely irrelevant IMHO.  Better to compare OS/2 PL/I
with IBM's mainframe compilers and the look-alikes.  Those offerings [OS/2,
Windows 95 & NT, and AIX] are relatively recent.
For the first 25 years, there weren't any major changes to the
language.  New features were added to the Workstation versions,
including an improved macroprocessor, and type facilities, not to
mention a swag of new built-in functions [including some for
Year 2000].

	>  Even compare it with the MVS PL/I compiler of 1989 (the
	>last time I saw an MVS PL/I manual) and you'll notice a lot of additions.

You know I know this.  I loaned you the OS/2 PL/I manual -- recall?

	>Richard A. O'Keefe