Bryce,
I think that discussing business models is on topic because RPG or any other
language may be used for strategic advantage. And I agree that the model you
delineated has worked in the real world. My main concern with it is in step
four, when your applications need to scale. That's when you necessarily inject
a human element to manage your workloads, as opposed to using something like IBM
i to do it for you. IBM i has lower TCO in that case. That has been documented
over and over.
Right now we're using RPG to develop applications for school teachers, parents,
and students to interact with each other. Homework assignments, instructions,
learning media, tests, scores, grades, questions, messages, class schedules,
attendance calendars, absence, etc.
In the past, schools bought relatively beefy PC's for teachers to run Grade Book
and other applications locally. It's a lot less expensive and much more
efficient to swap out the PC's with iPads and a browser, and to migrate heavy
desktop workloads to servers, which is a growing trend. That's something where
RPG and IBM i makes a lot of sense.
We've posted a quick paced promotional video on our Web site. It might inspire
you or somebody who may not yet see the vision of using RPG and IBM i for
contemporary workloads.
http://www.relational-data.com/promo/promo.php
-Nathan
----- Original Message ----
From: Bryce Martin <BMartin@xxxxxxxxxxxx>
To: RPG programming on the IBM i / System i <rpg400-l@xxxxxxxxxxxx>
Sent: Thu, August 12, 2010 7:26:37 AM
Subject: Re: Future of RPG: What language would you learn?
Nathan,
It *might* be less money to maintain down the road (haven't seen any side
by side comparison costs for a specific project). But here is how the
usual tech startup runs these days.
1. Use your man hours to create a minimum viable product. Low capital
because it can be done on a $500 PC that you have laying around.
2. Use the product to get some market traction. Serving can be as low as
$30/month to start.
3. Develop additional features based on customer feedback.
4. Scale out as needed with commodity hardware or cloud based VMs.
5. ???
6. Profit.
In fact the Profit part on SaaS apps can be as high as #3 in a lot of
scenarios with gross profit margins in the 30% to 60% of revenue range. I
would love to see an outline of an IBM i based app that can follow this
model. When it happens I'm all about it. But we aren't even close yet.
Not until the licensing model gets an overhaul. And it should get an
overhaul, its an old model that is being displaced by more customer
friendly and profitable models.
At any rate, we're a bit off topic here. The truth is that you should be
able to pick up any language without too much trouble. I agree that its
the domain knowledge that really holds the water. And really, these days
domain knowledge will get your a Project Manager job where you can spec
and delegate to less knowlegable code monkeys. Then, through experience,
they gain domain knowledge and the cylce continues.
Thanks
Bryce Martin
Programmer/Analyst I
570-546-4777
midrange.com
