I think IBM needs to get back in touch with its customers and it needs to ride herd on some of its renegade ISVs...
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
if it can.
I have seen more ISVs moving their wares to Windows because IBM's i5 is an unknown and the low lying fruit does not get picked by the i5 ISVs anymore. Rather fight than switch? No! Many ISVs have switched without calling mother IBM and they recommend their newer "Windows-based" software rather than their traditional iSeries wares.
In local government and education, for example, my clients see little ISV presence to combat the incessant marketing of all the little Microsoft sellers. I see ISVs joining the Windows crowd to sell their Windows versions if they've got 'em or packing it in when Great Plains or one of the other imitation ERP packages are under consideration.
ISVs are not IBM and they have their own business model to run after. IBM does not sell solutions itself and in many cases it does not provide enough reason and enough help for an ISV to want to push an i5 vs. a Windows solution.
As Windows gets better and better, the i5 will be left in the dust because there is little to no inertia. IBM chooses to implicitly suggest today that the system does not make a difference, the solution does. And, that is a losing strategy if you make systems. And, the i5 is losing turf as far as I see and as far as I read.
Is a system at the top of its game when articles about it are titled "Can IBM keep the iSeries alive?" If you are the product manager for such a system and people are wondering if your product is dead, isn't that a clue that something is wrong?
IBM does not seem to care which way its customers go. In my experience the ones who get away from an i5 solution don't go to a P5 or an xSeries. They got to Dell or HP and IBM gets nothing. Having four servers, none of which the company can "push" because it wants them all to do OK, means that IBM is easy pickings for its competition - especially in the small server area. If it really tried to differentiate its i5, and it spent the marketing dollars to do so, it would blow away the competition and also, of course its other server lines.
Re: Customer IT shops
My customers feel abandoned by IBM regarding the most important aspect of computing today, the Internet. Until IBM gets away from its WebSphere and Java push, Big Blue will continue to lose shops to the Windows and Unix crowd.
IBM's solution on the i5 is really a Unix solution, WebSphere and Java. The best people to implement that on an i5 are the Unix people and if they choose an IBM server it will be a p5, not an i5. Why would a Unix person want to run Unix on an i5 when there is a p5? Though the i5 may run Unix and Linux, few Unix shops would pick an i5 because it runs Unix and Linux unless they already had a need for i5/OS.
The i5 is staged as the right machine for everybody but i5 traditional customers. Yet those who it is staged for have other ways to accomplish their missions with PCs for Linux and p5s for Unix. They don't need the i5. And the i5 constituency does not typically care about Unix or Linux unless for some strange reason they want to partition and use Linux in a firewall or some unique application. But, from where I stand, there is little motivation for an i5 Shop to want multi-OS partitioning to make life more cumbersome for them. Most would opt out of using all of the latest and greatest IBM offerings because they are not made for them.
Just because IBM chooses not to admit that most of its i5 sales go to shops that use green screen processing does not mean that is not the reality. You and I know that PDM is still the main AD tool on the AS/400. IBM has abandoned this style of computing while its customers still depend on it. Isn't that ironic?
The thing that keeps an i5 shop tied to an i5 is the thing that IBM suggests its customers move away from. To what--Java and WebSphere? Who needs an i5 for Java and WebSphere? Isn't that a valid question? So i5 shops let the Windows and Unix folks run the most salient part of the business today - the Internet. Every now and then the other IT people (Unix and Windows) gain more respect from the organization than the "legacy" crew and eventually, the i5 system is either undermined or displaced.
IBM permits its i5 to be branded a legacy solution. Actually, in many ways it is legacy because IBM has not enhanced its "integrated" operating system with the integrated facilities it needs to help its customer IT shops fight for their existence in today's Internet world. IBM has left its i5 constituency without an integrated Internet solution. That makes the i5/OS operating system a legacy operating system. That makes the i5 a legacy machine. IBM's real AS/400 and iSeries and i5 customers feel isolated and abandoned because IBM has isolated them and abandoned them.
The i5 needs to be strategically positioned to win the Internet battle. The tools are not going to be WebSphere and Java. The OS needs to be usable as easily through a Web browser as through a 5250 terminal. Considering that IBM prematurely discontinued 5250s while its customers were still using them shows the rush the company has to make the paradigm change to eBusiness style computing. Yet, the only real tool (besides WebSphere and Java) that IBM gave its shops to make the move to better technology was iSeries Access(PC5250) capability. So instead of running real terminals because you can't really get them anymore, shops had to buy Windows PCs to pretend to be using dumb green screen terminals. What else is there?
If IBM were going to force its i5 shops to PCs, then why not change the 30-year old green screen command interface to browser based? How about a real Internet machine? Of course I would not rid the machine of its 5250 heritage while most of the applications continue to run this way, nor would I call attention to their age by calling them "legacy." I might call them proven but not "legacy."
Why not have every program and panel that is written for a 5250 automatically be callable from a browser without needing iSeries access for the Web (two-day installation) or any other piece parts solution? And for heaven's sake, don't make us drop our panels to a PC to have them converted so they can be re-uploaded to be used for the Web (WebFacing). That's not integrated.
How about modifying SDA and DSPF files so that keywords can call out specific functions available for tweaking browser pages? Sure, put them in WDSc editors also, but don't make your loyal constituency go there as the only source for the Internet. Convince the iSeries shops to modernize by giving them tools that fit, not by suggesting that Unix tools are the best in this particular situation.
How about having RPG natural operations such as READ, WRITE and EXFMT, and CL operations such as SNDRCVF and COBOL WRITES and READS be able to send out Web pages to Web browsers? Web pages remain foreign to pure iSeries developers because IBM has not integrated that notion within the constructs of the i5/OS.
The next step of course is to free the i5 shops from isolation. Who in IBM has decided that RPG applications should not run on Windows or Linux or Unix? Somehow it is important that Java applications which have just arrived in the last seven years must run on all platforms but device integrated RPG and COBOL applications that have been around for 40 years (long before S/38) should now run only on just one IBM platform.
If an ISV develops a real i5 application, that ISV has one heck of an effort porting to anything else, unless they write in Java and WebSphere. Why not do the right thing now that all IBM boxes are heading to Power architecture, bring the notion of DDS and RPG and COBOL natural database and interactive facilities to the other platforms? Then i5 applications could be transportable and the i5 ISVs and i5 shops would not suffer from application code isolation.
That's the roadmap to get the i5 back on track. A map heading to Java and WebSphere will move the shop to Unix, Linux, and maybe Windows.
Brian Kelly retired as a 30-year IBM Midrange Systems Engineer in 1999, having cut his eye teeth in 1969 on the System/3 and later with CCP. While with IBM, he was also a Certified Instructor and a Mid-Atlantic Area Designated Specialist. He has written twenty-three books and numerous magazine articles about current IT topics. His new book was just released titled, The All-Everything Machine.
Let us know what you think about the story; e-mail: Matt Stansberry, News Editor