Re: RPG IV Stories Wanted -- RPG400-L

>  - Did you convert all at once or "as it needed maintenance"?

First we had a company strategy to keep all existing applications in RPG III
and meanwhile start building new versions in RPG IV, using ILE, service
programs and with a new look, based on the GUI elements of the enhanced user
interface.  The idea was to have a clear separation, internally as well as
externally, between the 'old' (CISC, OS/400 V3, RPG III, function keys) and
'new' (RISC, OS/400 V4, RPG IV, pushbuttons) versions.  But while the new
versions were being built, some of our programmers (who? me? <g>) couldn't
wait to make good use of the new possibilities and just started using RPG IV
in existing applications.  I always felt that by combining all these new
things into one giant step the 'newness' of RPG IV was strongly exaggerated.
And indeed, about a year ago the step was made to advise all programmers to
convert to RPG IV whenever a more than trivial change was made to a program.

>  - What were the problems you encountered (if any)?

- Numeric overflow.  We use a homegrown conversion utility that takes the
IBM conversion a step further and converts ADDs and SUBs to EVALs.
Initially it converted MULTs and DIVs as well, but that was changed later.
Although it may be true that in quite a few situations numeric overflow is
unacceptable, we also had situations where numeric overflow apparently went
unnoticed in RPG III programs and led to bombing after conversion to EVAL.
- Detecting changes in programs.  We often take a standard program and
modify it for a customer.  When the original is still in RPG III and the
modified version is in RPG IV, comparing the two is somewhat difficult.
- Copying routines.  When the program you are modifying is in RPG IV and you
need to include a routine from an RPG III program, you have to convert the
RPG III program first.
- Reluctancy of conservative programmers.

>  - What do you see as the major benefits of having converted?

Service programs, BIFs, readability, proper definitions.  Especially the
possibility to replace endlessly copied subroutines by functions in service
programs is great (or was sadly lacking in RPG III, depending on viewpoint).

>  - Have you converted a package yourself because the vendor (JDE or
> whatever) has not done so?

We are a vendor.

>  - Any other comments you may have (e.g. what should IBM be doing to
> encourage laggard BPs to switch, etc.)

I don't think IBM should do more to encourage anyone to switch.  RPG IV is
evolving rapidly while RPG III is standing still.  What can IBM do more?
Make sure RPG III programs don't work anymore?  That would be very unlike
IBM (hm, another company springs to mind;  now what was their name?).

In general:  The complexity of programming in RPG has grown considerably
from OPM RPG III to ILE RPG IV.  In our shop we used to think of RPG as a
tool that could easily be learned in a couple of months (note: most of our
programs are based on templates).  Understanding business problems was
considered to be more important.  There is an ongoing shift from tool (like
the name still suggests) to a full blown programming language.  For the
technically oriented programmers (like myself) this makes programming more
fun; for the less technically oriented it makes it more complicated.

Joep Beckeringh




+---
| This is the RPG/400 Mailing List!
| To submit a new message, send your mail to RPG400-L@midrange.com.
| To subscribe to this list send email to RPG400-L-SUB@midrange.com.
| To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---