RE: Compile error when using field in a subfield data structure

> With a
> fixed-form system, there's only one way to code, and so it's hard to
get
> it wrong. With free-form coding (as is the case with practically every
> other programming language out there), you have to apply some
discipline
> to ensure your code is readable.

It's no harder "to get it wrong" with fixed format code than with free
form. Other than indenting, there's no difference between the languages,
except for those that the RPG team has artificially contrived.  In fact,
according to your later statement, it's actually EASIER to hide bad code
in fixed format than in free format.


> As I said, this isn't a problem if
> you're used to coding in another language and you already know how to
> format code. But unfortunately, it seems to be a new concept for some
> number of RPG coders.

Uh, Hans?  For some RPG coders, it IS a new concept.  Duh.  But
"formatting code" is a lot more than indenting.  There are multiple
schools of thought on nearly every decision in the process, from number
of spaces to where to hang your end opcodes.


> For those who need some guidelines, the most important thing about
> free-form coding is consistent indenting:
> 1) Start at column 8, and for each new nesting level, indent 3
> characters. That is, the code within a IF, DOW, DOU, or subroutine
block
> should be indented 3 characters in from the IF, DOW, DOU, or BEGSR
> statement.
> 2) The ENDxx statement should be indented the same amount as the
> corresponding IF, DOW, DOU, etc.
> 3) Comments should be indented at the same level as code.

This is one man's opinion, and ignores the hard stuff, including how to
properly break multi-line statements, handling selects, and a number of
other topics.


> Another issue may well be that proper indentation of free-form code
may
> well magnify and emphasize a design structure that's poor to begin
with.
> It's easy to hide a poor code structure in a "straight-line" fixed
> coding style!

Hey Hans!  Flip!  Flop!  Flip!  Flop!  Earlier you said it's easier to
code properly in fixed format!  Now you say it's easier to code bad code
in fixed format!

Hey, maybe IT'S JUST EASIER TO CODE IN FIXED FORMAT!  Wow, what a novel
concept!

Actually, the truth is that choice is personal, like choosing an IDE,
and comparing the two on any sort of absolute basis is just bombast.
Poor code lives in any syntax.  As an example, one of the poorest coding
techniques ever devised can only be accomplished in C and its
derivatives: the misuse of the "for" construct as a generic looping
mechanism.  But your implication here is that somehow programmers stick
to fixed format to hide bad code.  I suggest you correct that
perception, Hans, unless you wish to cultivate a reputation as an
elitist.


> Personally, though, it just boggles my mind that so many RPG
programmers
> have such a strong attachment to the old fixed-form style. Sometimes
> this debate reminds me of the following quote: "If you've been
pounding
> nails with your forehead for years, it may feel strange the first time
> somebody hands you a hammer. But that doesn't mean that you should
strap
> the hammer to a headband just to give your skull that old familiar
> jolt." --Wayne Throop. It would appear that some RPG programmers still
> prefer using their forehead.  :-(

There is no debate, Hans.  Some people simply prefer fixed format.  They
have as much right to their opinion as you do to yours.

But frankly, many of us have been slower in the uptake of free format
than we would have otherwise because you decided to remove the MOVE
opcode.  Because there is no MOVE, there is no clean way to mechanically
translate to entirely free format code, as there was from RPG III to RPG
IV.  This single act of ... what?  Arrogance?  Misguided elitism?
Whatever the reasoning, that decision has slowed down the uptake of
/free more than perhaps any other single thing.

So when you want to blame someone for the lack of acceptance of /free,
look in the mirror.

Joe