SCOUG-Programming Mailing List Archives

>> Next Message >>

Date:	Sun, 13 Feb 2005 03:12:22 PST8
From:	"Lynn H. Maxson" <lmaxson@pacbell.net >
Reply-To:	scoug-programming@scoug.com
To:	< "scoug-programming@scoug.com" > scoug-programming@scoug.com >
Subject:	SCOUG-Programming: Open Source Object Rexx

Content Type: text/plain

Peter,

First off PL/E (Programming Language/Extensible) is SL/I
(Specification Language/One). It was called PL/E early on and
then at the time of the Warpicity proposal changed to SL/I. It
basically reflects the dilemma that every programming
language is a specification language, but not every
specification language is a programming language.

Thus you you have the software development cycle with its
five stages (specification, analysis, design, construction, and
testing) with a different specification language used in each,
but in only one (construction) is the specification language
also a programming language.

Secondly, you have three third generation, programming
languages--APL, LISP, and PL/I--which include everything and
more than in all other third generation languages combined.
So why not combine the features of all three into a single
language. In doing so why would anyone expect that the
operators and data types would behave differently in
expression evaluations and results? That's the way they
work. That's the way they will continue to work. In their
separate languages. In a combined language.

So if I adopt a PL/I syntax for the combined language for its
simplicity in that every program element is a statement and
every statement ends in a semi-colon, it remains the same
even with the additional APL operators and the LISP list
aggregate.

I've no reason to change the syntax in upgrading the language
from third to fourth generation. I just have to have some
means of distinguishing between an assignment and an
assertion statement. That distinction likes in the proof theory
involved. In the assignment statement the proof theory must
produce one and only one "true" result. In an assertion
statement the proof theory may produce 0, 1, or more "true"
results. Just as in SQL, a fourth generation language, a 0
"true" result says the statement "fails" or is "false".

Again it will behave in a combined language exactly as it does
in a separate one. Whatever it does in APL, LISP, or PL/I it
will do in PL/E (SL/I). No more. No less.

The only new thing here lies in the packaging. Even that is
based on something "old". I'm not creating anything as much
as I am repackaging it...in PL/I whole cloth.

Now granted I've incorporated BNF (Backus Naur Form) in the
language. It's not my fault that Lex and Yacc do not look like
C even though they produce C source code. Those functions
in SL/I are written in SL/I. In that manner it is self-defining. If
it is self-defining, it is self-extensible. The one follows the
other.

I would be happy to provide you with examples of SL/I
containing APL operators and list aggregates. The APL
operators present a real challenge to anyone using an ascii
editor. What I may do is attach a file to a message like this
using OpenOffice where I can use the APL font. That says
that you have OpenOffice available to read it. Or alternatively
we could use PDF. Or I could just break down, taking your
suggestion to use function calls instead of symbolic operators.

I think not. I think we will continue to pursue our course in
the Programming SIG and enjoy our second month of looking
at Python. It provides enough food for thought to allow us to
lay the foundation for performing a linguistic comparison of
programming languages. Maybe in doing so we will feel more
comfortable in becoming conversant in multiple languages.

That's not a bad thing, even if it's not the best thing. But
until you get there, until you are fully conscious of the
travails of the journey, you can then decide what direction to
go from that point. I prefer to let people convince
themselves.

=====================================================

To unsubscribe from this list, send an email message
to "steward@scoug.com". In the body of the message,
put the command "unsubscribe scoug-programming".

For problems, contact the list owner at
"rollin@scoug.com".

=====================================================

Return to [ 13 | February | 2005 ]

SCOUG, Warp Expo West, and Warpfest are trademarks of the Southern California OS/2 User Group. OS/2, Workplace Shell, and IBM are registered trademarks of International Business Machines Corporation. All other trademarks remain the property of their respective owners.