---- Henrik said ----
Aaron has btw like a reed been swaying in the wind, he introduced an open
source project using EXT JS years ago, then he went for powerRuby and now
for node.js. This is very typical for young geeks that doesn’t has a
continuous strategy, I call them “technology jumpers” who leaves a trial of
what they did yesterday for others to thought out after they have left the
building and moved forward to other new technologies.
---------------------
It's not often a person gets a raw outside-looking-in opinion of themselves
so I suppose this is a great opportunity to lay out my decisions over the
past years. It's entirely possible my decisions are derived foolishly, so
I thought I'd digress to maybe save others from hitting some of the same
pot holes I did. Warning, this is a winded response so feel free to hit
delete.
I was fortunate to start my career with one of the pioneers of RPG CGI
programming, Brad Stone. He literally "wrote the book"** back in 1999. I
learned a lot from Brad about RPG CGI during our time at Taylor Corp and
continued forward with it when I began my employment with KrengelTech
(current employer). It was then I started writing articles and giving
presentations with the mantra of "there's very little you can't do with RPG
and a whole lot you should". It was an understatement to call me an RPG
bigot (these forums remind me of those unfortunate days when I tore people
down in reckless fashion). To back up my assertions I published open
source projects on mowyourlawn.com showing how RPG can spread your
horizons. OpenRPGUI was one such project that took a unique approach to
web development and we still have customers that use it today, though we no
longer support it. As you'll see later in this story, I couldn't continue
to recommend a technology that I didn't believe had a sustainable future
(i.e. no community contributions behind it).
**
http://bit.ly/brad-stone-erpg
It was around this time Mel Beckman made the proclamation in an
iProDeveloper.com article that "RPG is Dead". That caused me to break a
seal and I, in juvenile fashion, unleashed all manner of public slander of
Mel. After a time I realized how foolish it was that I was so consumed by
a technology (RPG) that I'd put down my fellow man. Long story short, I
had a call with Mel and asked for forgiveness and he forgave. If any of
you know Mel you already know what a gem he is to our community.
That whole experience of "worshiping" a technology threw me into disarray
and caused me to instead look at technology based on merit vs. emotion (a
first for me). Around that same time Don Denoncourt (another gem in our
community) approached me with a RubyOnRails project for a tech startup. I
said yes. We completed that project and I learned some about Ruby/Rails
(still a newbie). Quickly thereafter another RubyOnRails project emerged
where we were tasked with writing an entire online school from scratch
(course catalog, student registration, payments, grade book, report cards,
live classroom, etc). The school already existed but they were using many
silos of technology and many manual processes of copying info from one silo
to the other - hence the need for a rewrite.
Here's the kicker. We developed the aforementioned features in 9 months
with 3 developers**. It was then I realized there was no possible way we
could have done that in RPG-CGI (I had over a decade of RPG-CGI under my
belt at this point and about 2 yrs of RubyOnRails). That spoke volumes to
me. When I did things in RPG-CGI I was doing A LOT of plumbing code. When
I was writing RubyOnRails I did Google searches to find a Ruby Gem and
would have a need met in a few hours - I rarely write plumbing code for
RubyOnRails. I can actually focus on business requirements.
** Didn't really have a choice given school had a hard start date for 4k to
5k students :-)
This made me realize that technology is only as good as its ability to meet
business needs. Doesn't matter how much I like xyz language or xyz
platform, if I can't produce results to meet market deadlines, and if the
competition can, then I've chosen the wrong technology.
That online school RubyOnRails project was hosted on Amazon EC2 servers.
There wasn't yet a formal port of Ruby to IBM i that worked in all the ways
it needed to. That's when Anthony Avison, Tim Rowe, and I connected to
bring about PowerRuby. As I understand it the PowerRuby folks have some
exciting stuff coming, so stay tuned.
Now to finally address Henrik's critique that I've moved on from RPG to
Ruby to Node.js and his statement that I am "a reed in the wind"... I am
still significantly involved in Ruby on IBM i. For example, here's some
extensive debugging Tony Cairns (IBM) and I did to uncover/fix a database
adapter issue:
http://bit.ly/ruby-fetch-column-metadata-error Not only
that but I also highly encourage RPG shops to NOT throw away their RPG code
as they adopt new languages and instead revitalize it with XMLSERIVCE.
This aligns with my mantra of meeting business deadlines to adequately
compete in the market place. You will lose if you try to rewrite
everything Ruby/Node.js/PHP/Python/etc. That's why I've devoted a lot of
open source time to the Ruby XMLSERVICE interface:
https://bitbucket.org/litmis/xmlservice.
My day-to-day programming is currently split between Ruby and Node.js,
though we currently have more Ruby contract work coming in the door. I
imagine that has to do with the newness of Node.js on IBM i and expect more
of those projects in the coming months.
Henrik, if you ever have questions of me you can feel free to ask before
taking the route you did here. My email is aaronbartell@xxxxxxxxx. FWIW,
I hope you have great success with your PowerEXT.
Aaron Bartell
litmis.com - Services for open source on IBM i
midrange.com
