RE: Strange behavior w/%editc()

["Joe Pluta" <joepluta@xxxxxxxxxxxxxxxxx>]

> From: rob@dekko.com
>
> Joe,
>
> I didn't see the same thing as you did from Han's point on
> languages.  What
> I saw was that people seem to defend what they know best.  I
> didn't not see
> an attack on RPG as a business language.  Except for his tongue-in-cheek
> reference to Python :-)

I don't know, Rob.  I'm not sure who you're directing that at.  Is it me?
Because I'm not defending MOVE just because I know it so well, but because
there's no logical replacement defined.  At least with indicators, IBM went
way out of their way to make sure there were appropriately named and BIFs,
and even then the %EOF/%FOUND issue arose.  That is the kind of thing that
happens when you design a language without talking to the users - when you
"know better" than the thousands of programmers who use the thing on a
day-to-day basis.

With all due respect to Barbara and Hans, when I'm writing code, I don't
want to have to code

   eval b = %editc( %dec(a : %len(b) : 0) : 'X')

when I simply meant

   MOVE A B

I mean, really, is this readable?  I think to someone just learning RPG, the
latter is far more readable than the former.  I mean, %editc is pretty
tightly bound to some rather arcane RPG terminology, and I don't know many
RPG programmers who know what the heck edit code 'X' is.

If everybody here is willing to say that the former is the right way to do
things, and the latter is just a pile of manure used by old-fashioned
fossils like me, then fine, I'll drop the whole thing.  But I have a funny
twitchy feeling in the back of my head that if you took a poll, there would
be a WHOLE LOT of programmers who think the way I do.

Enough.  The points have been made.  Time will tell.

Joe