RE: Why we don't use procedures more (was MOVE opcode in freeform / strange behavior w/%editc)

Rick,

>>But I've just had a real block on the ile stuff.  I bought a book on
it a
>>couple years ago - but I just couldn't get through it.  There are now
100
>>ways to set up one of these beasts (service pgm, bound, by reference,
by
>>copy, prototyping, procedures, sub-procedures, bifs, activation
groups).
>>I've got nothing to base any decision on. I've nothing to compare it
to.


This is normal. ILE was never implemented the way that Paul Remptima
originally designed it or intended it to be. It supposed to be elegant
and simply.

I sat in a room at IBM years ago when they said they were going to do
ILE, The called it "NPM" at the time ("New Program Model") and started
referring to what was then "the" program model, as they "Old Program
Model" or OPM. I of course, after listening to the implementation
details of the NPM started to refer to the Old Program Model as "Opium",
not OPM.

Needless to say the others in that room I think were just as perplexed
about activation groups as I was. I guess I equated them to a sort of
"Group Job within a Job". That way I could visualize the resource
allocation capability more easily.

Today, I think the reason ILE is hard is because we continue to
misrepresent what ILE is.

ILE is not a version of the RPG language. This is IBM's own fault and
the fault of a lot of writers for both News/400 and Midrange Computing.
They constantly referred to RPG IV as ILE RPG. So we now had IT Managers
allowing programmers to use RPG IV, but not ILE RPG. In practice, "ILE
RPG" became a dialect of RPG IV. And that just confused things more.

Think of it this way: If you want to use a Bound Call (CALLB or CALLP)
or do dynamic memory allocation, you cannot run the RPG IV program in
the default activation group.

If you do not use the Bound Call nor dynamic memory allocation, the
program can be compiled to run in the default activation group.

I refer to the Default Activation Group as "RPGIII-compatibility mode".
If you create an RPG IV program that needs to be RPG III environment
compatible, it can run in the default activation group, but it does not
have to. If you create a program that uses bound calls or dynamic memory
(and no there's a couple of other things that restrict *DFTACTGRP usage)
then it needs to run in the native environment, which today is an
Activation Group.

If you want, for performance or resource allocation reasons, the
activation group to stick around after your program ends (that is, until
the Job ends), specify that it runs in a NAMED activation group. If not,
it can run in a dynamically named activation group referred to as a *NEW
activation group; and in that situation, the activation group is
automatically ended when the program ends. *CALLER is just a convenience
thing. It allows you to create additional programs or service programs
that inherit their callers activation group attributes. So if the caller
is running in a *NEW activation group, the called program or loaded
service program also runs in that same dynamically created activation
group. If the caller is a named activation group it runs in there.

At the end of the day, I just "accept" the aggravation group technology
as it is, and move on. I've stopped trying to comprehend the logic for
it, (not that it's bad, it is just a little too complex to worry about).

Bob Cozzi
cozzi@rpgiv.com
Visit the on-line  Midrange  Developer  forum at: http://www.rpgiv.com


> -----Original Message-----
> From: rpg400-l-admin@midrange.com [mailto:rpg400-l-admin@midrange.com]
On
> Behalf Of Richard B Baird
> Sent: Thursday, February 28, 2002 4:10 PM
> To: rpg400-l@midrange.com
> Subject: RE: Why we don't use procedures more (was MOVE opcode in
freeform /
> strange behavior w/%editc)
>
>
> Buck and Bob,
>
> I agree that the onus is on me to train myself.  I have never blamed
the
> boss for not learning new things - I always just did it myself.  I too
> never went to college.  I took a cobol class at a community college
and a 6
> month trade school gig (about two months of which was actual coding -
the
> rest was accounting and weak 'related studies'.  I've been to company
paid
> training 4 times in 19 years.  Everything else, I've learned on my own
> (through here, and other on-line venues or just getting my hands dirty
and
> doing it).  I visit the midrange archives 3 or 4 times a week, the ibm
> manuals website (NOT infocenter - it sux) twice or three times that.
I'm
> not lazy (as long as you don't ask my wife).
>
> I've had 3 IT jobs in 19 years, two of them in non-hourly 'project'
> consulting and my current one in hourly contracting.  I've been the
boss,
> I've been the employer, and I've been the peon (now, i'm the peon :)
but
> I'd never made less money one year than the last until last year.
>
> All my new development is in RPGIV, I've personally written three
> 'real-world' RPG-CGI web apps and designed and project-managed another
(all
> are still being use and continue to provide value to the customers).
That
> puts me in the company of about 5% of the of the AS/400
programmer/analyst
> world.
>
> I'm as good or better P/A as I've ever worked with, and that's no
brag,
> it's because I work harder, longer AND I have the desire and ability
to
> grasp new things and keep up with the times.
>
> But I've just had a real block on the ile stuff.  I bought a book on
it a
> couple years ago - but I just couldn't get through it.  There are now
100
> ways to set up one of these beasts (service pgm, bound, by reference,
by
> copy, prototyping, procedures, sub-procedures, bifs, activation
groups).
> I've got nothing to base any decision on. I've nothing to compare it
to.
>
> Just take a look at the few posts to my questions about activation
groups
> today... a couple were from people who tried to help me understand,
and the
> rest were people correcting what someone else had said that was wrong.
>
> What does that say to me?  Few people truely know what they're doing
and
> why, and of those that do know, few can articulate it well enough to
> explain it to someone who doesn't.
>
> anyway, I've had my rant,  I really do appreciate the help I get here.
I'd
> go nuts if I didn't have it.
>
> ttfn,
>
> rick
>
> ----original message----
> Bob said:
>
> >This is stuff you should be doing on your own,
> >even with your own money.
> >
> >It is the right thing to do.
>
> AMEN!
>
> Most certainly I never had a job where the boss said "Hey, don't worry
> about
> the deadlines, take a week or so and go tinker."  It took me 23 years
to
> make it to Common!
>
> I am... let's politely say 'well seasoned,' starting on card
equipment.  I
> have no university education.  I learnt literally  e v e r y  thing I
know
> by myself, and virtually all on my own time.  I have a library of
books
> bought with my own after tax dollars - and not all AS/400 books
either.
> K&R, Knuth, etc. as well as our friends in the midrange world.  I have
an
> account on one of the timeshare machines to play on.  I use Code/400
and
> before that Picante's Flex and before that Brief.  All paid for by me.
>
> If I don't keep my skills current, what choices will I have for work
next
> year?  In 1978, I could run a 1403 blindfolded (I was GOOD!)  If I
merely
> maintained my awesome 1403 talent I'd be out of work, as the 1403's
are
> almost all gone.
>
> I have a family, house, etc., etc., etc.  I value my ability to feed
my
> family, so I make time to keep up.  If I can do it, anybody can.  If
> there's
> one wish I have, it's that I could really, truly convince midrange
> programmers that THEY CAN DO IT!
>
> GO GO GO GO GO GO GO GO GO!
>   --buck
>
> _______________________________________________
> This is the RPG programming on the AS400 / iSeries (RPG400-L) mailing
list
> To post a message email: RPG400-L@midrange.com
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/cgi-bin/listinfo/rpg400-l
> or email: RPG400-L-request@midrange.com
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/rpg400-l.