Kelly
I have taking a look at your company homepage just to understand what line
of business you are in.
One thing I would have expected to find in a logistic and redistribution
company was a strong emphasize on EDI communication (X.12/EDIfact/XML) and
track and trace to support the physical movement of goods, but there is no
apparent reference to it on your homepage it is kind of hidden in PDF’s?
You have further more described that you have a Cobol and a .NET team of
programmers. My guess is that the IBM I runs your core administrative
application and that it is either legacy or semi legacy software. Semi
legacy should in this context be understood that the company may have
started out with some ‘standard’ system that during the years of refinery
has become legacy.
Btw, does your Cobol team use SQL in their code or do their code depend on
RLA? SQL is a must know technology in whatever you do in direction to web!
It would also be interesting to know what team that handles the EDI and
what formats are used. This may seems like an odd question but tells
something about which team is used to handle/have skills to handle
hierarchical text based data structures.
If my guess I right your cobol programmers are likely guy’s that besides
programming in cobol has a deep knowledge of how your business is conducted
while the .NET guy’s may have more technical skills but less business
knowledge.
What I also am missing in your description is what the current status is
and what the goal of the technical exercise is e.g. run your web
technologies primarily on .NET with IBM I as a backend delivering raw data
or let the IBM I have a more active role in the overall web application
delivering runnable components to the overall web application.
As a general remark to node.js I can tell you that it is very hard to move
procedural programmers without any web skills into node.js. You will simply
overload them with new technologies that has to be learned such as HTML,
CSS, Javascript, jQuery, angular/bootstrap, HTTP, Websockets, PASE, JSON,
how to connect to DB2, SQL, stored procedures BUT most important OO
javascript that is used by node.js.
OO javascript is one of the hardest languages to learn. Even OO C and java
programmer struggle because it may be very different from traditional OO
languages since it is event driven, has a non-blocking I/O model and may be
used in the traditional way with classes and methods but also can be
prototyped and injected in several ways.
In other words, node.js is as raw as it can get and very similar to the JVM
and it doesn’t come with any standard UI framework in its basic form, it is
just a virtual machine and a programming language like JVM/Java.
So the question isn’t really where you position your IBM I since it can be
positioned everywhere, but where do you position your Cobol team. Throwing
basic/low level ‘on edge’ web server technology in their face is IMO a
recipe for disaster or at least a showstopper while moving them step by
step is a feasible roadmap.
Remember also that many of those you discuss with in this forum take many
things for granted and has completely forgotten that they have learned
their own skills in an ongoing process that has been going on for decades.
They may assume and discuss from the thesis that your Cobol team is at the
same level as themselves, but frankly if they were you wouldn’t be in this
forum but would attend internal meeting were you would be lectured in their
strategies.
This is however not the ‘normal’ case for many programmers that have been
sitting in their shop focusing on maintaining a running business system and
their current IT infrastructure.
On Sat, Oct 3, 2015 at 5:19 PM, Nathan Andelin <nandelin@xxxxxxxxx> wrote:
Buck,
You make a good point about a lack of tooling in regards to DB event
procedures. It would be nice if developers could hover the mouse over any
physical or logical file reference in any source member and be prompted to
"open" any DB event source members pertaining to each file reference, for
example.
Such tooling would also ameliorate concerns about the "visibility" of code
pertaining to cascading events.
But I wouldn't delay the adoption of DB event handlers, waiting for tooling
to make them more visible. After all, it would be nice if one could hover
their mouse over any procedure reference in any source member and be
prompted to open the source member where each procedure is defined.
Soon after the adoption of DB event handlers, developers begin to view
write, update, and delete statements the same way they view procedure
calls. The benefits of implementing ILE concepts out-weighs the
"visibility" that was available in the old days "monolithic" source
members.
Regarding the question of how broadly or narrowly to scope DB event
handlers so as to avoid unintended consequences of cascading "calls", that
question applies to standard procedures too.
One might begin by writing DB event handlers for data validation purposes
which are called before write and update operations. Once comfortable with
the interface, branch out to business rules which are incidental to
successfully completed updates (after triggers).
--
This is the Web Enabling the IBM i (AS/400 and iSeries) (WEB400) mailing
list
To post a message email: WEB400@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/web400
or email: WEB400-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/web400.
midrange.com
