RE: VRPG And/Or Java -- MIDRANGE-L

Chris,

All good points, but... <g>


> In that light, spending extra effort to put things in RPG that are 
already
> in Java would seem to be spending money on keeping people from moving
> where you want them to go.

Java is still new. It doesn't do stuff that RPG does. I like Java and the 
idea of Java. And, I don't not use SEU or SDA for development. Heck, I 
wrote my own windows-based development tools!

What I believe is that AS/400 developers do not want to do it all 
themselves. They want IBM to build it into RPG, DDS and CL, and then give 
it to them for free. While Java is "free" it does not do stuff without the 
AWT. Each AWT has or will have subtle differences. I hope the AWT toolbox 
for the AS/400 does GUI.  I don't believe AS/400 developers want to build 
routines and tools that they are used to having IBM give them with their 
products.

In my experience, existing AS/400 customers do not want to buy stuff that 
IBM doesn't "give" them. Case in point, IBM just can't seem to move 
CODE/400. And my product, CodeStudio was priced at about 1/3 the price of 
Code/400 and I shipped just under 1,000 copies. I don't have the numbers on 
similar products by other vendors, but I know they have similar issues.


> While bytecode generating ILE compilers should be very handy (and I would
> thing relatively easy to produce by simply treating the byte codes as a
> processor op code set) they only address part of the issue. Java 
bytecodes
> are what allows for the process to be platform independent, so the RPG
> apps become independent.

RPG is only part of the answer. There is DDS, there is CL, and there is 
database. Where do these languages fit in? Is there a 100% conversion to 
Java for these?


> But Java is also faster to code in. Object oriented languages are faster
> to program in and easier to maintain.

Well, I suppose German is faster to talk in than English, unless of course, 
you don't know German.

> Also, there is so much broader
> support for Java that Java programmers will become easier to find than 
RPG
> programmers are.

Perhaps, if Java "sticks". But the skill of the business developer (i.e., 
the RPG programmer) will continue to be easier to find than a Java business 
developer for quite some time.


>Third party applications are allowing for Java add ons or
> plug ins. This means that a Java based AS/400 shop could broaden it's
> influence because the same coders that put the applications on the AS/400
> could be writing the front end plug in parts to go into documents or web
> pages or whatever.
>
> What I am saying is that what I see is that IBM can either try to keep 
RPG
> up with Java by establishing two development efforts, or they can simply
> throw their weight behind Java. Since Java won't be GA for six months, I
> suspect it will be a year or so before we see visible signs that RPG is
> falling behind, but how many new projects can we expect IBM to be
> launching in the mean time?


Right now, everybody at IBM is involved in Java. And if they're not, they 
want to be. I think this is largely due to boredom on their part. I mean, 
how interesting is SEU to someone with a BS in computer science?  New stuff 
is usually of interest to true systems programmers (like the folks in 
Toronto) and system programmer wannabes.


> >I mean, it is 1997 isn't it? While text-mode application do have their
> >merits, they are NOT the only applications in the world.
>
> I think IBM has been trying to tell us that very loudly. VA C++, VA
> Smalltalk, VA Java, CA toolkit, etc. IBM has been trying for years to get
> us to adopt a solution that allows for GUIs on the 400 and we have been
> dense. Toronto gives in by trying to stuff more features into RPG, and
> when you step back and take a look you realize that RPG is evolving into
> C.

Well, I agree with your thinking, but I don't agree with the premise. I 
think IBM has, in the past, simply reacted to what the others in the 
industry were doing. I believe they were doing a lot of "me too" and 
"anything but what the other guy is doing" rather than leading. Today 
there's still some of that, but no where near what it was.


> But the problem is that the solutions we have rejected on the AS/400 have
> been widely accepted elsewhere. As a result, the AS/400 loses ground. Now
> we are at a make or break point. If the AS/400 programmer cannot finish a
> project because he cannot produce the front end required, then the AS/400
> will lose sales to technologies that DO go end to end.



> Bob, in a way you are acting like the IBM of a few years ago. The market,
> and the world, are changing around you. You are still demanding to do it
> your way with the same tools you have always had.

On the contrary, the "tools" are SEU/PDM/SDA, etc. I have moved on from 
those legacy things years ago. The Language, is just a language. It is not 
the tool. A "move" is a move is a move is a move. Just because I can code 
it as

EVAL X = Y
or
X = Y;
or
Compute  X = Y.
or
MOVE X   Y

doesn't make it a tool. It is a cultural thing, I think.

> Wouldn't it be nicer if you could pick and choose the tools you wanted
> from any of a number of vendors instead of always going back to the same
> vendor and complaining that they need to add the things that all other
> market segments get?


Well, whether I complain to IBM about the AS/400 of which many of us have 
extremely good incite into it's capabilities, or I complain to IBM and 
Microsoft, and only get results from IBM isn't really any different to me. 
IBM does respond. Microsoft doesn't. Microsoft make much more attractive 
and often (not always, but often) much easier to use products. IBM usually 
(not always, but usually) makes stuff that works. And as you say in your 
tag line... "You have to ask yourself, How often can I afford to be 
unexpectedly out of business?"


Bob Cozzi
Any typos are purely there for your entertainment.


>> Chris Rehm
>> Mr.AS400@ibm.net
>> You have to ask yourself, "How often can I afford to be unexpectedly out 
of business?"
>> Get an AS/400.
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to "MIDRANGE-L@midrange.com".
| To unsubscribe from this list send email to MAJORDOMO@midrange.com
|    and specify 'unsubscribe MIDRANGE-L' in the body of your message.
| Questions should be directed to the list owner/operator: david@midrange.com
+---