From its conception, the IBM midrange platform has provided us with an unparalleled ability to do our jobs without a lot of folderol. No smoke and mirrors, no rewriting applications with every release of the operating system. Just a solid, no-nonsense machine that contains all the tools we need to write powerful business applications. The box is as near-perfect a business application development platform as has ever been introduced, yet people both inside and outside of IBM insist on "fixing" it.
I'm Not Just Drinking the Kool-Aid
Let's take a look at some of the things you get with an IBM midrange. One of the major breakthroughs is the concept of the integrated database. You plug in the machine, and there's a database already in place. In fact, with UPDDTA and RUNQRY, you arguably have all you need to create an entire business system (heck, that's basically what you have with some of the less-advanced SQL-based "applications" on other platforms).
Another breakthrough is integrated work management. Try to explain the concept of batch subsystems (much less routing entries) to UNIX developers. Watch their eyes dart around as they edge toward the door, while telling you how "cron" does all of that for them. Scheduling is something you do on your calendar, not on your production machine. And what in the world is "print spooling"?
So what's my point? Well, only that the idea behind the midrange—even before the (over)use of the lowercase "i"—is that it's integrated. You get a box, and just about everything is in there. A couple of configuration options and away you go. IBM has not only embraced this idea, but extended it. The Express machines provide great bundled packages, while the venerable 5722WDS licensed product has evolved into a one-stop shop for development tools. With just a couple of basic decisions, you have a great machine for running a small-to-medium business.
And that's important, because in these days of the great internecine rivalry between the Windows server farms and the System i, it's become increasingly difficult to get extra money to spend on the System i. Instead, an ever-larger part of the IT budget goes toward Windows licenses or anti-virus protection or spam firewalls. And while that's starting to backlash—the ever-growing burden of Microsoft licensing is managing to do something the entire open-source world could not: get Linux onto corporate desktops—it's still an uphill battle to get additional line items in the budget for the System i.
So the integration of the platform is one of its most powerful features, and the pinnacle of this integration is the WDSC tooling: a complete, self-contained suite that can not only maintain legacy applications, but also extend them with new technologies. WDSC boasts an unequaled multi-tier debugging environment that supports Java, JavaScript, and RPG in the same debugging window (that's application, UI, and business logic layers!). And now, WDSC brings us EGL, the next-generation business language that can be used to quickly build standalone standards-based Web applications but at the same time is flexible enough to reuse existing legacy code.
IBM stands poised to take the best business application development language (RPG) and marry it to the most powerful user interface technology (JSP Model 2) using a seamless interface (EGL) developed on a tool platform that provides unrivaled debugging capabilities (WDSC), all to deliver killer applications on the most stable, flexible, and scalable platform ever designed (the System i).
They Shoot Their Own Feet, Don't They?
So how will they screw it up? Because it doesn't matter how good a product is; IBM can still figure out how to mismanage it. Remember, this is the company that brought you OS/2 and let it founder. OS/2, a product so good it manages to still be used despite the almost complete lack of support from IBM—does that remind you of anything? Or perhaps you remember the unconscionable bait-and-switch that accompanied WebSphere 4.0 and nearly killed the product in the System i community. Given that history of Custer-like marketing brilliance, it's clearly not out of the realm of possibility for IBM to grasp defeat from the jaws of victory here.
And there are already rumors that IBM stands poised to execute the single most effective way to do just that: namely, "unbundling." "Unbundling" means "we gave it to you for free, but now that you're using it, we're going to charge extra for it." As far as I can tell, this move is being driven by the Rational team and springs from a mentality of charging you for every piece of the puzzle. This may work in PC shops, but it's diametrically opposed to the System i development paradigm of integration. It might not even work in the PC market; charging for formerly free stuff is one of those "greed is good" Microsoft strategies that works only when you're a monopoly and even then sometimes causes a revolt. In other vernaculars, it's known as "the first one is free."
The rationale (no pun intended) seems to be that bundling the tool in with the price of the System i is somehow losing money for IBM. Well, that's hardly the case; if anything, by providing these tools free to iSeries developers, IBM is getting a lot of extra press that it would never have gotten from the midrange world (I've probably written more articles on EGL than anybody else in the entire IT industry), while at the same time providing the System i community with the tools needed to build the applications that will keep the platform viable in the changing technology market.
Whether it's unbundling WDSC from the System i or unbundling EGL from WDSC, this sort of move will not garner extra money for the product. Instead, it will simply bring to a screeching halt the penetration of the unbundled tool into the System i marketplace. I've already talked to people who are interested in EGL but are worried about paying an extra price; they won't be able to get the cost past C-level management if the language is unbundled.
What's the Answer?
Well, the answer is simple and makes common sense, which means it's probably doomed. But I'll give it a shot anyway: If the issue is getting revenue for the Rational tooling from the System i community, then go ahead and put a per-user price tag on the tool, but bundle a reasonable number of licenses into the 5722WDS product. What's reasonable? I'm not sure, but I don't think you need 50 licenses for a P05. And if a company does need extra licenses, it can buy them. But this way, the product gets in the door of every System i shop, and I can tell you that if a tool is free, it will get used. Look at all the Net.Data and RPG CGI out there. If EGL is free and it makes it easy to get RPG logic behind Web applications, you can be pretty sure it will be embraced.
Or you can charge extra for it, make a couple of extra bucks in the short run, and then watch it dwindle away.
Can you say "kill the goose"?
Joe Pluta is the founder and chief architect of Pluta Brothers Design, Inc. He has been working in the field since the late 1970s and has made a career of extending the IBM midrange, starting back in the days of the IBM System/3. Joe has used WebSphere extensively, especially as the base for PSC/400, the only product that can move your legacy systems to the Web using simple green-screen commands. Joe is also the author of E-Deployment: The Fastest Path to the Web, Eclipse: Step by Step, and WDSc: Step by Step. You can reach him at
LATEST COMMENTS
MC Press Online