RE: Using RPG /Free how do I get an ISO variable into a non-date 8,0 variable in *USA format

> From: Mike Haston ** Data
>
> Unless of course they take about 15 minutes to soak in what's
> going on with the free form code.  I've run into too many
> situation of programmers throwing their hands in the air and not
> touching a program because:
>
> if  a = b
>
> didn't look like
>
> a  ifeq  b

It's not these trivial differences I'm talking about Mike, it's primarily
the issues surrounding the MOVE opcode and the many BIFs needed to replace
the simple MOVE.  A legacy program will need %dec, %char, %subst, %len and
%editc just to replace a couple of simple MOVE instructions.  All for no
added value.

That's stupid.

Now, if your position is that I can use /endfree to work around it, then I
guess I'll live with it, but rather than have these hybrid language modules
laying about, I think I'll try to segregate my code into /free modules and
/nonfree modules.  That way two programmers can complement one another
rather than having to duplicate each others' skill sets.

Understand, I'm not deadset against /free.  I just think the compiler team
started deciding what was good for us as programmers, and that dog don't
hunt.  Just because you write a given application (be it a game, an MRP
generation or a compiler) doesn't mean you should decide what it does.  In
fact, the people who implement are usually the worst group to decide the
content.

In commercial development shops like SSA in its heyday, there are usually
two groups: development and product management.  Product management
represents the client and decides what goes in the product, development
represents the programmers and calculates how long it will take.  Knock down
drag out brawls ensue as the sides wrangle over what goes in the next
release.  It may be ugly, but it's a necessary evil.  If you let development
decide all by itself, then almost nothing goes in the next release, because
development usually doesn't consider it necessary.  It's product management,
the folks who represent the client, that push the boundary of the product.
(Conversely, if you let PM decide the whole thing, you end up with
unfeasible requirements lists that lead to late releases of partially
implemented and woefully undertested code.)

There's no product management department for the RPG compiler team that I
can see.  That didn't matter until now - there seemed to be a natural
unspoken agreement to make sure that legacy code was brought along.  But
recently it's more as if decisions are made based on what the compiler team
thinks we the clients need.  And from what I've seen, I don't think they're
using the right requirements as the basis for some of their decisions.

Joe