SCOUG-Programming Mailing List Archives
Return to [ 12 |
January |
2004 ]
>> Next Message >>
Content Type: text/plain
"No. I'm having trouble grasping why you aren't making any
progress.
I'm not trying to dissuade you on your quest. But if you'd
spent the past five years coding we'd all be farther along. ..."
Peter,
A quest? I don't think so. Maybe that explains the difference
we have on the matter of coding or finding ourselves "farther
along". I find this "we" thing interesting, basing it on what "I"
have or have not done.
"We" seem to have forgotten how "we" got into this in the
first place. It began when IBM formally announced its
withdrawal from the retail OS market and lowered its sights
relative to OS/2. We had much whining, gnashing of teeth,
frustration, and anger. The OS/2 community felt threatened
and in need of a solution.
Many offered solutions from begging IBM to change its mind to
projects like ODIN to enhance Windows product support under
OS/2. When I got my turn to speak up and out I offered the
Warpicity Proposal. It contained three parts, one relating to a
fee-based organization, one relating to its staffing, and then
one related to a methodology.
Now of the three I felt most strongly about the need for the
first, the organization. Its lack meant that we had no means
of negotiating with IBM as a business partner. Its lack meant
that we had no financial means to operate as a business, no
means of making community decisions, and no means of
implementing decisions once made.
Unfortunately the community's focus bypassed the
organizational and staffing parts, choosing instead to focus on
the methodology and its two tools, a specification language
and its implementation in the Developer's Assistant. Both
relied on existing software technology. I didn't have to invent
a thing. We had the "what". We simply needed to change the
"how". In IT (and elsewhere) we call it process improvement.
If "we" agree on it, then "we" can do it. That "we", however,
says that we have the first part, the organization, as our
means to our end.
I still feel that way. I certainly feel no obligation to prove
something based on what already exists. I know what has to
go into its development. I don't have anything to prove. So if
I seem somewhat laggard in my coding efforts, chalk it up to
my feeling that until we have a "we" thing I feel no obligation
to make it a "me" thing.
"...The PL/I Level F compiler compiled straight to machine
code, and had an option to disassemble that code into
Assembler. I did that a number of times. Guess what? The
generated machine code didn't look any different from what
you might get from any other language compiler (with the
exception of unique library calls, of course). I too love PL/I,
but I realize that it is just one particular organized syntax. ..."
Somewhere along your line of reasoning you forgot that it
generated code that other languages could not, because they
had no means of expressing it. It didn't lack data types, i.e.
operands. It didn't lack operators. It didn't require additional
libraries. It didn't require resorting to assembly language. It
had a complete exception handling facility, allowing even
programmer-defined exceptions. it had a simple syntax. It
had complete record (sequential, direct, and indexed) and
stream i-o support.
That makes it fairly specious to say its code looks like what
other compilers produced when you have massive amounts
missing.
"...The compiler ain't gonna figure it out by itself, Lynn. And
if you keep adding specs, well, all you are doing is writing
your algorithm in a different particular organized syntax."
I said you didn't get and you still don't. No software figures it
out by itself without someone figuring it out for it. All that
figuring amounts to writing specifications. The argument
comes down to how many you want to write before you get
to the final one you compile. If you can do it writing once,
why burden yourself with writing multiple?
The difference lies in logic programming, not in the specific
language you choose. The write once occurs there. How
much you have to write lies in the syntax. That depends on
the language you use. So why not use logic programming
coupled with a minimal writing syntax?
Nevertheless in a programming language you can only write
specifications. You can in a suitable language write those
specifications in the same language. In short you only need
one language. Whatever language you choose must
incorporate all of formal logic, all its data types, all its
operators, all its rules. All that you can reduce to machine
instruction form.
It's good that you can drop in from time to time. We sorely
miss you from our meetings. Meanwhile "we" remain content
to fill in the gaps in your logic.
=====================================================
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".
=====================================================
>> Next Message >>
Return to [ 12 |
January |
2004 ]
The Southern California OS/2 User Group
P.O. Box 26904
Santa Ana, CA 92799-6904, USA
Copyright 2001 the Southern California OS/2 User Group. ALL RIGHTS
RESERVED.
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.
| 
