RE: Many RPG programmers would like to be able to port their apps to other

> From: Jon Paris
>
>
> First of all none of you have seen the requests that continue to flow into
> the Labs.  I have.  For years one of _the_ most frequent requests has been
> "Make RPG available on other platforms".

Not by any application vendor I know, Jon.  I can only guess that these are
people who somehow think that running RPG on some PC-based SQL database
makes sense.  You well know what my opinion of that particular line of
thought is.


> For goodness sake we've seen requests on this very forum
> for PC and Linux versions of the RPG compiler!! how can you claim that no
> one wants to do it?  Maybe few people want to do it to so that
> they can move off the box, but that's not the only reason.

We were talking about porting applications.  Nobody I know wants to.  And I
certainly have no desire to see RPG on a PC, either.


> Maybe you guys don't want to port your apps - but there are many who do.
> Are all of you seriously telling me that if you were a BP you
> wouldn't like to be able to sell a version of your app that ran on
someone's
> Unix box?  or the NT network they already have?

Absolutely, undeniably NO.  Capital letters.  Absolutely the LAST thing I
want is for NT to be seen as a "viable platform" for mission-critical
application development.  The single worst thing that could happen to our
community would be to place the iSeries in a position where it's an also-ran
with SQL server or Linux.  Since we as a community seem to be completely
incapable of selling TCO as a strength, by making our applications available
to NT platforms, we'd just be cutting the iSeries' throat.  Every CEO would
be buffaloed by the fact that an application could "run" on a low-priced NT
network, only to be put out of business as his IS requirements grow past the
ability of NT to scale.  You might argue that this would be a good point to
then "scale up" to the iSeries, but my argument is that NT should never be a
point in that spectrum.  By putting a simple, powerful browser interface on
applications running on a low-cost low-interactive iSeries, you make it
unnecessary to ever use the NT box.  There is NO reason to have RPG on
another platform.

Jon, I would never sell an NT version of an ERP application.  Nor would I
sell a Linux version.  As you know, I'm intimately familiar with the concept
of porting a non-trivial application from OS/400 to Unix.  Neither platform
can even hold a candle to the iSeries in terms of TCO, backward
compatibility, reliability and scalability.  If an application doesn't need
these things, then it's a desktop application and by all means run it on
whatever OS you happen to have lying around.  In my mind, industrial
strength applications and desktop applications are completely different
animals.  If an application is small enough not to need an iSeries, then it
probably has no business there in the first place.  But if the application
does need an iSeries, it should never run on NT or Linux.  I can see no
applications that would reaonably support both an industrial strength
iSeries version and a "don't care if it crashes" NT version.

That's just my opinion, but I'm pretty adamant about it.


> The fact that you think the iSeries is the best dang box in the
> universe and
> that RPG is a great language for business apps doesn't mean that you
> wouldn't like to run on other platforms.

Yes it does.  I feel it's a matter of professional integrity.  I wouldn't
sell you a lemon of a car, nor would I sell you an inadequate operating
system.  Neither NT or Linux belongs as your mission critical business logic
server.  Unix might, but OS/400 applications don't port to Unix, and it has
nothing to do with RPG but instead with emulating the OS/400 support, as has
been discussed.

In any event, what I want to see is the iSeries community quit complaining
about how they can't compete with Windows and Linux, and instead see us get
up off our collective butts and start writing killer applications like we
did in the 70's, 80's and early 90's, before we became so darned complacent.

Unfortunately, those of us who wrote those first message-based apps in the
early 80's or those great windowed data entry applications in the late 80's
or those powerful client/server applications of the early 90's seem to have
grown weary, and are now waiting for IBM to fix everything.  As I said in my
previous post, it's up to us.  The tools are there.  The tools are there,
honest.  Just pick them up and use them.

Joe