System iNetwork

Thank you! This article has perfect timing. First, as a small consulting firm supporting legacy iSeries application systems, we see significant value (especially now) in legacy systems that were developed with a "smart" core. We add value to the legacy investment by introducing modern add-ons with simple interface models that have immediate ROI.
Second, as the economy recovers, the smart business owners will have a significant jump on the competition emerging as leaders by having streamlined their businesses to maximize the ERP productivity with minimal investment.
Just as is said about real estate, 'Now is an opportune time to invest and hold.'. In the ERP world, now is an opportune time to leverage the investment and streamline the processes. Intelligent, short term solutions will deliver long term success with a minimal investment.

"...but the economic downturn may very well prove useful to the IBM i world: the System i has long done well in total cost of ownership, ..."

as companies reduce headcount, the remaining programmers cant be as specialized as they have been. IBM does not sell a comprehensive OS, runtime framework and suite of programming languages like MSFT does. On IBM POWER system, you need specialists in RPG, AIX, IBM i, CL, Perl, C, PHP, Java, SQL procedures and DB2. Skills in all of these languages and environments are not transferable to any other. Someone who knows Perl knows nothing about CL. An RPG programmer cant get the simplest task done in PHP. This is the result of IBM's underinvestment in their own software technologies.

Yes, IBM i is easy on sys admins. But to program it, you need a team of specialists.

A Sliver of opportunity for the 'i'?

The economic downturn affects all IT platforms, MSFT, SAP, HP, IBM all lose business. But what about the IBM i? Some say the 'i' runs 5 years behind competing platforms for the past 15 years on all fronts like hardware, software and marketing. Ibm 'i' just doesn't sell, the effort in 2007 to recoup share in the SMB space flopped big time. Market share of 'i' in the server market dropped from 15% in 1998 to less than 5% in 2008. IBM is trying to catch up very hard lately, so their babble on modernization gets louder every day, but no one is listening.
And given the current gloomy economic climate, companies are inclined more then ever to let their traditional RPG400 frameworks run unchanged. The sad part is this downturn adds probably 2 more years to the stand-still of 10+ years these applications were already in. Once the economy picks up, these apps will look so old, managers will look so young (having no affinity with the legendary stability of 'i') that the decision to migrate quickly to a modern platform will fall inevitably. Good news for the 'i' migrators and demolition guys. Those who still cling their future with the 'i' platform, if you poll them for their motivation, the answer is invariably 'great platform...', 'good business value...' but at the company restaurant they all confess, well, in silence we all hope to stick it out until our pension.

It could be a silver lining if … If business leaders start looking at value over the long haul instead of mere start up cost and short lived hype … Hmm

To phrase it another way … Mr. CFO, CEO and all the other O’s in the business world would, you like to have your IT staff‘s time busily providing business solutions for you business issues or busy trying to hire another resource so you can buy another server to support another business software that relates to nothing your doing currently all the while tell you how cool and sexy it all is? Please choose the right CHOiCE!

Seriously it could be a resurgence especially combined with the green movement and the i’s ability to consolidate workloads. We would also need an influx of new software development and the right marketing. That a tall order but it is doable if the business leaders at IBM are paying any attention this time. Do not let this be another Window 95 vs OS2!

Speaking of RPG, the C-prototypes were a good start. I am now "Full-bore" learning C#, yes! C# not C++!!!

There are no Global Methods in C# as there are in C++. This utterly forces us into small easily documented chunks of problem-solving. I am a big offender in maintaining some of my old code that exceeds 10,000 statements.

I am grateful for my rebirth in Windows, MS SQL-Server, and C # - land. It means employment for me at near the age of 63. It also
means "Neuroplasticity" in continuous improvement. IBM did not do "Continuous and never-ending Improvement" with the AS/400 in the 90's.

I wish, but this is water over the dam, that IBM had pioneered in rolling out a graphics-based IDE but it was Microsoft that did it with Visual Studio. VS has such fire-power that it has a HUGE community that eagerly awaits improvements and flames M$ when they screw-up.

Others invented good IDE's but M$ slammed the "Home-run". WDSc was too little and too late. It started out as a resource-suck. It compared very unfavorably with Visual Studio.

Let us move on.

I dedicate these comments to Ralph Daugherty, great i-supporter until, he himself, was unemployed from i-Land

Cheers!

--John

John, sorry to hear Ralph D. - my biggest but most respected opponent over the past years - had to leave i-land.
I don't know if his enthousiasm for the 'i' is still standing in these gloomy days, but worse, since he and many other i-fans relegated all MS, Oracle, SAP, Java stuff to 'toy'-land, they probably have no alternative left to search for employment outside i-land.

Oops. Lets call this phenomenon the 'i-trap'.

Coming back for a moment to the Visual Studio (VS) IDE as mentioned in an earlier post, MS's success didn't come overnight. VS dates from the early '90 ties with its biggest breaktru succes being Visual Studio 6.0 released in 1998. In 2001, when they released version 7.0 (dubbed dot.net), MS made the capital mistake to alter the internal function of some VB language statements and methods so they were no longer compatible with 6.0. Unheard of in the IBM world. But they've learned their lesson and in Visual Studio 8.0 and 9.0 (dubbed 2005 and 2008), they provided the proper import and migration tools.
In a nutshell, MS tries to please the customer, but, sometimes in their enthousiam and lack of corporate experience, they make foolish mistakes. MS learns all the way and continues to try to please and appeal to the customer which is a whole different approach compared to IBM. IBM says, show me the money first, then we'll think over it and then we will teach you what you need. IBM's approach leads to fewer, frustrated and angered customers (just look around in the i world), MS approach leads to ever increasing communities and happier customers. Even the scalability of the MS solutions using its MS SQL server flagship improves constantly, but -pay the king his dues-, hasn't yet toppled the ibm iSeries. On the other hand, for SMB customers (less than 500 users), MS based servers have completely whiped out ibm.

The reports of my leaving i-land are greatly exaggerated. :)

It is true I haven't been here blogging lately. I got a 270 and a Linux server and have lots of plans for them. I got both of them idling on the internet while I ramped up on some software development.

Guess which one crashed and burned until I had to turn it off until I have a chance this weekend to look up how to drop Gnome off of it? (because Gnome is forcing reboots) By the way, was configured and sold by a professional Linux server vendor just last month.

Also true I was unemployed in 2004. Was a nightmare screwing around with "we want someone with this package version x.yy in last year" or "locals only" or "we need 10 contractors for a project to start..." and they never start. Or "we need a contractor for three months, you didn't make the top 10 list for interviews" and the best one of all, "you've been unemployed for a year, we need someone with recent experience".

And also true I gave up and just went into prepping on Java programming although I had no illusions about that being any better. Went through the whole lose your house / file bankruptcy at the head of this curve [correction: death spiral] we're going through. First time I was leading edge ever.

But I moved to Florida to be more local to where I wanted to be and was finally contacted by someone from my jobs board postings and got hired back to RPG programming again in early 2005. So coming up close to four years with this company now.

I laugh and laugh and have always laughed at idiots who talk about "maintenance" programming. Whether new programs or adding to old ones, as long as I've been programming RPG (since '89), there has been a neverending list of new functionality to add to systems. And at least half of it is adding to existing programs.

Business actually needs to perform for customers. They need cost effective, efficient programming and execution of their business systems, and they are required to intensively test and recertify anything that changes. Rewrite systems in some developer's language du jour because he wants to pad his resume? Not with business people doing real work with it, thank you.

Rewrite business system in some developer's language du jour because he claims he will be able to "maintain" it more easily in the future. Only complete imbeciles can make that claim with a strauight face. Even the most business math challenged among us should be able to compute the cost and effort to rewrite and recertify a business system and how many "easy" maintenance projects would be required before allegedly there's a payoff, by which time the developer's language du jour would have changed and the program modified (oh excuse me, refactored is the buzzword du jour) beyond any dream of "easy", and the developer with his resume padded with his favorite language gone anyway.

So yes, I've been doing a hell of a lot of "maintenence" programming all these years, and every modification was new and improved business capability.

But through the years I've interfaced RPG with lots of technologies, and so have most of the rest of us RPG programmers everywhere I've seen. A couple of you here are one note johnny's about this small shop with one programmer who doesn't want to change from RPG III or even RPG II or something like that. I don't know. But it gets old.

I write all /free now but modify existing code in same syntax. I prefer SEU but based on project use WDSc and have used it as long as six months without coding in SEU. I call Java classes from RPG like any other ILE procedure. Or how about IFS? You ought to see the speed RPG has doing file I/O with IFS calls, and with recursive subprocedures dropping down through subdirectories doing it. And try beating RPG native I/O with real business logic. Customers count on the speed and power of our iseries RPG software.

And the 5250 screens still beat web pages by a long shot. I've been doing lots of windowed subfiles and messaging for pop ups. Makes a web page look downright anemic.

We write service programs for both 5250 and web page calls, just like lots of other people who post here. But ultimately, I'd like to see your Windows servers do what our iseries does, in whatever your language du jour is, speaking of anemic.

oh, and Merry Christmas and all that too. :)

rd