Re: Moving from Green screen app to GUI because of push back

I can attest to that! One customer I have has been trying to "migrate".
Well, 2 years and literally millions of dollars later with little to show.

Original project was done over 10 ago (which was a web application using
CGIDEV2). Total for that was under 40k. That's 40k over 10 years. Ground
up application (minus DB which was there), support, updates, etc. That
doesn't include the green screen apps and full time programmer who handles
that, but still... It was making a web interface for existing green screen
applications.. both to run side by side.

I'm sure all the new management wanted is something that "looks nicer", and
given a couple months I know that would be simple to do with what they
have. 10 years ago there just wasn't the technology to do it. HTML5, CSS,
jQuery.. very good things to have in your toolkit. 10 years from now it
will probably be something else that takes their place.

Higher ups (and mostly non tech) see these websites using responsive
design, etc and want that and think the IBM i can't do it. That just shows
the lack of understanding of how all this web stuff works.

Another reason they want to move off is they feel it will be cheaper than
bringing in someone to train under the retiring single programmer. I'd say
that's been proven wrong.

But each instance is different. I guess.. lol. But this has been my
experience... zero speculation. Actual experience.

Bradley V. Stone
www.bvstools.com
MAILTOOL Benefit #15 <https://www.bvstools.com/mailtool.html>: The ability
to add a Footer to each email sent using an IFS stream file.

On Wed, Dec 13, 2017 at 2:54 PM, Jon Paris <jon.paris@xxxxxxxxxxxxxx> wrote:

+1 Nathan.

Migration is expen$ive - very very expen$ive. And few shops seem to get
the promised benefits.

That's why I mentioned the idea that more IBM i shops - when threatened -
should ask to make a competitive bid. If nothing else it forces the shop to
look at the value they have and the true costs that are likely to occur
when transitioning.


Jon Paris

www.partner400.com
www.SystemiDeveloper.com

On Dec 13, 2017, at 2:35 PM, Nathan Andelin <nandelin@xxxxxxxxx> wrote:

Mark,

The part of your argument that I'm having difficulty reconciling, is the
assertion that some of your clients have moved to another platform

because

providing an IBM i GUI was too costly.

The cost of migrating databases and applications from IBM i to Windows

and

Linux is so high in my experience, that I question the validity of the
assertion that the cost of an IBM i GUI was really a factor, or that IBM
could have averted migrations by bundling a GUI solution with the OS.

One group asserts that they do provide GUI applications on IBM i. This
group tends to delineate how they do it. Another group blames IBM for not
bundling a GUI with the OS, but refrain from delineating what they mean

by

a GUI solution.

Most people who provide a GUI under Linux, do it with HTML, CSS, and
JavaScript on the front-end; paired with PHP, Python, Ruby, Node.js,

Perl,

or Java language environment on the back-end. Why don't they ever

complain

about the lack of native GUI support, bundled with the OS?

How should we reconcile the opposing perspectives?






On Wed, Dec 13, 2017 at 10:53 AM, mlazarus <mlazarus@xxxxxxxxxxxx>

wrote:

Nathan,

You're only partially correct. The application would need to be
rewritten, to some degree. But, the additional cost of the add-on

product

plus the additional development and testing overhead made it too

expensive

an endeavor to undertake, from the clients' perspective.

I went through this at several medium to small clients. Had IBM
included and integrated this functionality into the OS I would have

created

some standalone mini applications, or a parallel function to an existing
application, and let them get used to having *their* data displayed in a
real GUI, with additional functionality.

There is no way that they would shell out for an expensive tool (plus

the

yearly maintenance!) just to try it out to see if they want to go in

that

direction. There are other details in play, but that was a big part of

the

decision.

The fact that we can get a GUI to display does not mean that this is an
inherently a GUI OS. It's not. IMHO, until IBM decides that it's
worthwhile investing in creating a fully integrated, modern interface as
part of the OS, this box will be perceived as old and dated. That's a
shame, because it's a real workhorse with many innovations and

capabilities.

-mark