Re: free form RPG reference -- RPG400-L

rpg400-l-request@xxxxxxxxxxxx wrote:

   1. RE: free form RPG reference (Jonathan Mason)

How could this become a "management approval" item? How is it different

from using an API that

became available for V5R1 or a new IBM command? Especially, how is it

different from using a new

%bif or any other feature of a language?


It's very different because it has an impact on the whole RPG programming
department.  Use of a new %BIF or command can have an impact - I've been in
a number of sites where new %BIFs weren't allowed because of variations in

operating system levels.


Jonathan:

As I said, it is indeed a significant difference in magnitude, andmaybe that's the only point. It'd still seem to need someone whodecides whether an item is "big enough" to pass upwards for decision.

And certainly in the case of OS level differences for %bifs, etc.,it shouldn't be much a 'management' decision as a simple fact ofknowing how to program for a platform -- you don't write programsthat can't run on the intended platform if you want to stayemployed. No?

The change to free format RPG requires some degree of a learning curve which
by its nature will reduce the programming efficiency of the programmer
involved and there is also the problem of having to maintain something you
may not be familiar with.  A lot of places will stick with fixed format RPG
purely because that's what their programmers know and is also what their

packaged software is written in.

Yes, there is a learning curve. What... a couple hours? Mostly thedirect learning curve is conquered in the first ten minutes. Then,as individual elements are chosen and used, each one might take tenminutes to look up in the Reference. Many will be understood andused quickly; some will take experimentation. Once a programmer haswritten his/her own line of code, that line becomes the futurecopy/paste guide and future lookups decrease.

In addition, there are a couple particular elements that can addminor, temporary snags. The relationship between the SQLpre-compiler and free-form is an example. Given the total number ofdevelopers, though, I wonder how many adopters of embedded SQL arealso restricted from free-form. But again, once an initial exampleis available, the next ones are dramatically easier especially ifself-written.

As for maintenance, I pretty sure that jumping into maintenance offree-form is easier than starting in writing from scratch; so maybe'learning curve' isn't as important there. And I'm not at all surethat maintenance of a free-form program by an inexperienceddeveloper is slower than maintenance of an equivalent fixed-formprogram, for even fixed-form die-hards. (That's pure speculation. Ireally don't know and I'd be interested in hearing about experiences.)

I've worked at a number of sites in the UK and at one of them, a major
automotive company, a contractor had to ask for permission to code a program
in RPG IV simply to get around a field/record size limit.  That was 5-6
years ago and I doubt if there programming standards have changed much since
then.

BT, DT. Reality is understood. I'm just curious on how it got thatway. I'd think there is some kind of disconnect between managementand staff that doesn't make total sense.


Tom Liotta