Has IBM Given Up on the i?
August 24, 2009 Dan Burger
I get the impression there are no night lights in Bob Cancilla’s home. Every switch is wired to a high-intensity bulb. The volume on the TV is always on full blast. I wouldn’t expect him to have 300-thread-count bed sheets. They’d be #50 grade sandpaper. Cancilla, who is one of the many gurus of the AS/400 community, is a likeable enough guy, but some find him nothing less than acerbic, especially when it comes to the IBM i. He says IBM will drop it in the next five to seven years.
He might as well have whacked a hornet’s nest with a baseball bat.
Some say he should just leave well enough alone, but Cancilla is not inclined to when it comes to the current state of affairs at IBM with regard to the i OS. There is no “well enough” in this situation as he sees it. Although Cancilla sometimes sounds like a lone wolf howling at the moon, he unquestionably has supporters who agree with some (but seldom all) of his points. A good example of like-minded thinking can be seen in the iManifest programs that have sprouted in the Asia-Pacific and EMEA markets.
A lot of what Cancilla has to say gets misinterpreted as anti-AS/400, so his forum posts and comments to his blog, called i-nsider and running on Blogspot, and the related Twitter tweets often move the debate to whether the hardware, the operating system, or the RPG language is the greatest invention short of adding bacon to eggs. That’s not, Cancilla says, the target.
His opinions about the demise of the platform are meant to draw attention to an observed disinterest on the part of IBM to view the platform as strategic. With his first swing of the bat, he says IBM’s investment in new software solutions is marginal at best. Enhancements to existing products will continue, but he claims that inside IBM there is “a moratorium on developing new products on i.” No new investments in IBM i software.
Take a look at the reduction in RPG compiler advancements, once a bragging point for the Toronto Software Labs, Cancilla says. (This is an organization for which Cancilla worked for many years until this summer.) It has declined, as he sees it, in proportion to the overall reduced investment in the IBM i platform. “One way or the other, expect support for RPG to eventually be withdrawn,” Cancilla predicts, “and be certain that historical advances in compiler technology are already diminishing.”
I’d say there’s a good chance that IBM will refute any talk of a moratorium, but there’s an equally good chance that nothing definitive will be mentioned in term of i-specific investment and innovation that affect a wide swath of AS/400 shops. That’s been much of my experience.
“I’ll be amazed if there are new apps and tools made specifically for i,” Cancilla says.
To support his opinion about diminished investment and eroded exclusivity of features that the AS/400 once enjoyed, Cancilla pointed out in his blog that many AS/400 features that gave it distinction have been diluted as part of the Power Systems convergence. He specifically mentioned SLIC (System Licensed Internal Code) and TIMI (Technology Independent Machine Interface) because they have become part of the virtualization engine, now known as the PowerVM hypervisor.
Both are highly advanced technologies that separate the IBM i from lesser servers. The SLIC software dictates how the operating system gains access to and interacts with hardware features. The TIMI is a hardware abstraction layer that presents a virtualized implementation of AS/400 hardware that integrates with OS/400.
In this particular instance, Cancilla drew the response of Bill Austin, a technical consultant for a large ERP software company in the AS/400 community. Austin reeled in Cancilla’s comments a bit by noting only a small portion of the SLIC is in the hypervisor and the TIMI is not.
This example doesn’t discredit Cancilla’s overall opinions. However, it adds some degree of grey to a discussion that Cancilla views as pretty much black and white. I realize there is a portion of what he says that can’t be fact checked.
He piles on the criticism by noting the sales and marketing efforts have been demolished, and that many more customers are leaving the platform than are moving to it.
“My points are not about the value or the quality of IBM i–the OS,” he said during a phone conversation last Thursday. “This is about IBM’s commitment to the system. IBM is simply not investing in the OS or products related to or running on the OS. No one is selling IBM i, and a business without new customers is a business that is dying. The IBM i OS is gradually fading away because IBM is not selling it. Since there is no longer an organization to sell IBM i, there can be no turn around or return to prominence. It will simply continue to decline in users and will most definitely be dropped by IBM when the revenue reaches a point where it is no longer feasible to continue supporting it.”
Whack the hornet’s nest again, Bob.
According to Cancilla, there has been “a significant decline” in the number of shops renewing their software maintenance on the OS and “a huge decline” in the number of vendors that were once devoted entirely to OS/400, i5/OS, and IBM i but have moved to Java or other technologies that support multiple platforms.
From what I’ve seen in the comments that followed Cancilla’s blog posting, very few people challenge his charges that IBM has lost its commitment to the AS/400. IBM certainly isn’t crowing about the sales of Power Systems running i OS. The company claims to not even keeping track of such information. No numbers have ever been left unattended within eyeshot that could shed light on the independent software vendors that have moved development from RPG to another language. Ditto for shops that have not renewed software maintenance contracts.
Cancilla says his insights into the comments he makes comes from four years inside IBM as the Rational software marketing manager for the IBM i marketplace. During that time, which ended just months ago, he says various internal reports he had access to, as well as working relationships with IBM i personnel, provided support for his opinions. And in his survey of the Power Systems landscape, he sees it as “devoid of IBM i executives.”
Back in May, during a conversation I had with Ian Jarman at the COMMON 2009 Annual Conference and Expo in Reno, Nevada, Jarman predictably denied there was any lack of IBM i executive input in the Power Systems group. But I’m afraid that more than a few people would side with Cancilla on this.
“When Mark Shearer was general manager of the System i,” Cancilla points out, “he had a project management team, a VP of marketing, and a worldwide sales organization. Now there are no VPs. Ian Jarman is a second-line manager. The Power Systems project management, marketing, and sales teams are all made up of AIX guys.”
Cancilla does not escape without criticism. Some say he is bitter about his experiences inside IBM, where changes come slowly and progress can be a grueling, highly political encounter with frustration. Eating that same sandwich every day would give a lot of people a belly ache. It’s also noted that since leaving IBM, Cancilla has taken a position as chief technology officer with Oxford International, a company that is a Premier Business Partner with IBM Rational. Oxford is in the application modernization business, which ties in nicely with Cancilla’s warnings to companies that are locked into the RPG world that he says IBM will abandon when Big Blue no longer views it as a money maker.
Dire predictions that the AS/400 is approaching the end of the line have surfaced more than once during its mostly illustrious 20-year history. It’s almost like a wound that never heals.
To go into this a little deeper yourself, you can find Cancilla’s blog at http://i-nsider.blogspot.com/. Readers with soapboxes are welcome here at The Four Hundred as well. The airing of various opinions on what IBM is or isn’t doing have been and will continue to be part of what we report. As this story is but one recent example.