Within the past decade, the rate of technological change in the IT industry has accelerated exponentially. Many technologies used by midrange enterprises today didn’t even exist 10 years ago. Back then, midrange professionals could rely on the simple, core technology that comprised the infrastructure of a typical midrange enterprise: an AS/400 running integrated enterprise applications, usually written in RPG.

Networking was an internal affair, limited to connecting an AS/400 within the enterprise or to building PC LANs with Novell Netware. The character-based, green- screen OS/400 operating system was a refinement of its predecessors, the System/38 and the System/36. The Internet was not a focus of business as it is today, and data exchange was conducted through proprietary electronic data interchange (EDI) channels. PCs were on the periphery, used primarily for office productivity or for workgroup applications; they were cranky low-powered devices best suited for light word processing and spreadsheets. Microsoft Windows was merely a GUI floating on top of a very creaky DOS-based operating system. Enterprises that outgrew their AS/400 could either buy additional AS/400s or move up to an IBM mainframe.

The Good Old Days

In those simpler times, a midrange professional plotting his career course could choose a well-defined path in management, application development, or systems administration. Regardless of which path he chose, the list of core infrastructure technologies was small: OS/400, green-screen RPG application development, and basic PC networking and office productivity applications. After acquiring core competencies with these technologies, a midrange professional was a portable commodity and could “plug in” to a new enterprise easily.

The AS/400 itself was marketed as a secure, evolved enterprise platform, offering compatibility with the past and continuity into the future. IBM actively promoted the notion that the AS/400 would remain the stable centerpiece of midrange enterprises for many years to come.

COMMON meetings had a quasi-religious tone, with much time devoted to sessions reaffirming the decision to stand by the stable, reliable IBM midrange platforms. Midrange professionals were a happy, contented clan.

Sailing Charted Waters

The stability and continuity promised by the IBM midrange platform provided midrange professionals with a high degree of marketability and job security. As a result, many midrange professionals focused on refining their skills, using the proprietary IBM midrange technologies. Loyalty to the midrange platform and technologies ran high; despite the rise of PCs as a force in computing, many midrange professionals either dismissed or avoided them because of their technological immaturity. Why dabble (and struggle) with PCs and learn new software development technologies if the IBM midrange platform will remain the centerpiece and platform of choice for “serious” application development and mission-critical applications? Why devote time to immature and potentially transitory technologies when one could leverage his or her midrange knowledge and experience for greater reward?

Foresight or Foolishness?

Changing course to become a PC-literate developer or devoting significant time to learning new PC development technologies in the early 90s may have seemed like a giant step backward for many midrange professionals. After all, PC programmers got paid less, worked on lower-profile projects, had higher frustration levels, and often were forced to support a hodge-podge of immature and incompatible hardware and software applications. PC applications were then ancillary to the meat-and-potatoes, AS/400-based technology core. In many enterprises, the PC programmer was a second-class citizen. He was not next in line for the vice presidential slot; the seasoned AS/400 professional was. The decision to stand by the AS/400 seemed a logical career choice.

The Rise of the PC Platform

With little warning, and at an accelerated pace, the PC platform began to mature. Microsoft Windows matured and became a dominant force on the desktop. PC server technologies, including distributed SQL databases, became powerful when distributed. PC application development tools and programming languages became increasingly powerful, robust, and easy to use. The PC hardware itself became exponentially more powerful, and hardware prices plummeted. Networking, standardized by the TCP/IP protocol, was popularized by the recently commercialized Internet. Microsoft and others refined visual development environments, such as Visual Basic.

These changes made it possible to create more robust applications using cheaper hardware in less time. And the new technologies enabled companies to construct more powerful, GUI-based applications much faster and at a far lower cost than using midrange or mainframe tools. PC technologies also permitted developers to make truly interactive applications instead of green-screen or batch-processing systems. Developers flocked to the simplicity and power of the new tools and applications. Companies reveled in development-cost savings, shorter development cycles, and ease of maintenance.

The PC was no longer a peripheral player; it was becoming a platform for high- value applications (including custom executive information systems [EIS]) and increasingly powerful office and analysis tools, as well as becoming an active client in client/server applications.

The Internet Technology Tidal Wave

The rise of the commercial Internet and the new e-business mentality has brought greater tidal waves of change, such as redefining the accepted frameworks of client/server

computing. Seemingly overnight, new PC-based technologies have emerged, promising increased power and ease-of-use, cross-platform compatibility, and rapid development.

These technologies have evolved so quickly that an entire section of the publishing industry has arisen to feed the need to learn about them. And numerous startups have become billion-dollar industry leaders without ever having used a midrange product.

The focus has moved from stable midrange technologies to constantly evolving PC- based technologies. The advent of open-source software and operating systems such as Linux has attacked an assumption that was the foundation of computing, the long-held belief that one vendor’s technology should be the cornerstone of any enterprise.

Assessing the Storm Damage

These tidal waves of change have blindsided many busy midrange professionals, and, some would argue, IBM itself. Within the past five years, IBM has been racing to integrate new PC-based technologies into the midrange environment with varying degrees of success. IBM has adopted the e-business moniker, and it is racing to reposition and rebuild its midrange hardware to be Internet-centric; it has even renamed the line of servers. IBM is also rushing to support PC-based development and deployment technologies on new servers in the midrange product line, hoping to share in the new wealth. Absent are the long-familiar assurances of continuity and compatibility that came with the old legacy technologies. The new message from IBM is that it will provide stability and continuity by embracing the new technologies, which IBM plans to port to the midrange world. Despite this belated but ambitious plan, companies like Sun Microsystems, Oracle, and Microsoft are the undisputed leaders of the Internet age.

What Heading, Captain?

The tidal waves of change have left many experienced midrange professionals feeling like they are floating in a sea of uncertainty, without a clear direction. The insatiable demand for skilled IT professionals who are fluent in new Internet technologies—Linux, Java, C++, Microsoft technologies—greatly outstrips the demand for midrange professionals. Midrange professionals have watched young dot-com IT professionals become millionaires. They have also watched major application solution providers shift their applications to PC- or Internet-based technologies, away from legacy platforms like the AS/400. And they have watched as new, popular ERP applications have been developed that are not available on IBM hardware.

Older midrange professionals may be justifiably concerned about their long-term career decisions. Many are unwilling to continue relying on the assumption that dedication to IBM midrange technologies alone will provide marketability and job security, especially when IBM is playing catch up implementing third-party technology into a renamed but aging midrange hardware platform. But which technologies should a midrange professional learn? IBM’s new “embrace all” philosophy doesn’t provide guidance.

Midrange Computing recently received a letter from a reader posing these very questions, and I suspect he speaks for many of you:

What about the people and the skills needed to support all this new and wonderful technology? What do you do about the human element? I live with change, but I have a hard time dealing with change for the sake of change. I’m an IS manager. I’m 45 years of age. I work in a one-man shop (me) for a small business in the Midwest. We are running two AS/400s, one for J.D. Edwards OneWorld software and the other for Lotus Domino and Web serving.

I’m busting my butt trying to keep up with all new features IBM is throwing at us—and keeping everything working! I realize I have to learn or die, but what do I learn? I was slow to pick up ILE RPG (big mistake) and ended up taking a class at the local community college. After ILE RPG, I realized it was a short hop from CGI [Common

Gateway Interface] programming, and that made sense because I could get legacy data onto a Web browser.

A friend and I are always pushing each other to stay one step ahead, to learn the next new thing. In the past, we attended COMMON as a learning experience. But you can’t learn anything in depth in five days.

My question is, what would we learn if we went to COMMON? WebSphere, LotusScript, JavaScript, VisualAge for Java, Visual RPG?

I’m getting older, and I don’t learn a quickly as I once did. IBM can roll out new hardware models quarterly, but what about the people who have to support all this new technology? I’d like to see what you have to say about that. It’s easy to rattle off the specs of some new box, but let’s talk about the people who need to support them. What skills are we going to need for the future?

—John Q. Programmer

The following were some of the suggestions, commentary, and responses the MC editorial team gave Mr. Programmer:

The best way for legacy AS/400 programmers to improve their marketability is to learn how to make the best use of their legacy knowledge in Web applications. The Internet world, for the last couple of years, has ignored legacy coders and legacy applications. No more. Companies now realize that, to compete in this new, networked economy, they must deploy legacy applications to the Web. Companies have also discovered that, while these young bucks out of school can quickly slap together a cool Web site, they don’t have the experience to develop real-world business applications.

What works best is separating the coding responsibilities of the Web front-end and the back-office. But, to do that well, the legacy applications have to be modular or use asynchronous message-based processing (such as with data queues or MQSeries).

You should gear your education in the following ways:

• Learn modular programming. ILE RPG is now a modern language. The power of ILE is not all those new op codes and free-format expressions; it’s the modular capabilities of subprocedures and service programs.

• Learn HTML, sticking mostly to knowledge of HTML forms. You don’t have to learn how to make a Web site flashy; all you need to know is how to replace your 5250 screens.

• Learn to use the CGI with RPG (or eRPG, as Brad Stone calls it). Read Brad’s book eRPG: Building Web Applications with RPG. This book has a great tutorial for configuring the HTTP Server for AS/400 and covers HTML basics.

• Learn JavaScript. JavaScript is all but required for browser-based editing of fields. Field editing with JavaScript is usually the last thing done to a Web application, but the language is easy to learn and prepares you for Java.

• Learn basic Java, but don’t bother learning Java GUI programming. Just write to the screen initially; later you’ll be writing to Web pages via HTML. I suggest you read one or two Java introduction books. Then, once you are comfortable with Java syntax, read my book Java Application Strategies for the AS/400.

• Learn JavaServer Pages (JSP). JSP is basically HTML with embedded Java code. JSP allows you to dynamically construct HTML with information obtained with Java, either directly from your application’s database or from interaction with your legacy RPG code.

• Learn Java servlets. Knowing JSP alone won’t cut it. JSP is great to get started, but you need to use JSP only for the user interface and use Java servlets for the complex code. (My book covers both JSP and servlets, as well as the basics of HTML.)

• Learn component programming with Java and more about object-oriented programming (OOP). Take a class at a community college or attend a seminar on pure OO methodologies as soon as you can, and read Object Technology: A Manager’s Guide.

As for learning WebSphere, put that off for a while, at least until IBM gets the kinks out of it. The configuration of WebSphere changed radically from Version 2 to Version 3, and it will change radically once again with Version 4 (due in 2001). Instead, use Tomcat from the Apache Software Foundation (www.apache.org). For more information on Tomcat, read my article that’s coming in the December issue of MC, which has a WebSphere replacement that’ll take you two hours to install on both your PC and your AS/400. Your server-side Java applications will run exactly the same on both your PC and your AS/400. Once you are ready to deploy Web applications and WebSphere becomes easier to work with, move to WebSphere. Your Java applications will run fine in both Tomcat and WebSphere.

—Don Denoncourt Senior Technical Editor

First, build on what you know. Visual programming will be around for a while, so why don’t you start with one of the Visual RPGs? IBM’s new application-development bargain costs $300 and includes VisualAge for RPG. Once you understand it, you’ll have an easier time moving to other visual programming languages if you need to.

Second, you can’t learn everything, so work on whatever people are using, like HTML and JavaScript.

Third, don’t forget that everything eventually goes out of date, including whatever you decide to master now. Prepare to continue learning.

—Ted Holt Senior Technical Editor

[Editor’s Note: For more on RPG, see Ted Holt’s article, “Is RPG Dead?,” on page
103.]

• Become familiar with XML. XML is a standards-based metalanguage that enables data exchange over the Internet. I believe most future B2B application communication will employ XML.

• Learn SQL. Relational databases aren’t going away anytime soon. However, the way you interact with relational databases in a standards-based environment will probably be through SQL.

As you learn more about these newer technologies, you should be able to find ways to apply them to your new projects and possibly some of your existing ones. If you’re a Domino developer, it’s pretty easy to use some of the new technologies. For example, one can choose to use Java instead of LotusScript to write agents (programs in the Domino environment). JavaScript can be used to enhance browser interfaces, and developers can use SQL to retrieve AS/400 data through the LotusScript Data Object (LS:DO). SQL can also be used to retrieve data from Domino through the NotesSQL driver.

—Richard Shaler Associate Technical Editor

Time for a Change

John Q. Programmer may really need a career change or at least a job change. He needs to do some soul-searching to discover the reasons for his resistance to learn new technologies. There can be a lot of practical reasons, such as money, inconvenience, and uncertainty. As you get older, the prospect of sitting in a classroom full of twenty- somethings becomes less attractive. And, eventually, one gets tired of trying to keep up with the next big thing, especially when much of the new stuff either never becomes mainstream or never really works.

I suspect Mr. Programmer wonders whether investing the time at this point in his career is a good decision. Deciding to stay a programmer and a one-man show at 45 may not be a good choice for someone like John. He may be bored and unchallenged. IMHO, John should consider getting out of the one-man shop and taking on a new management or super-visory challenge at another company. That alone is a growing process and will highlight new areas in which he needs to learn, grow, and improve. Alternatively, he should try some consulting, both to grow personally and to learn about what technologies companies in his region are using. That could help him shape his learning and career paths.

I don’t think the answer can be found by selecting specific technologies; it is found by assessing your career path. That path dictates what skills are needed. If John does his homework, he’ll learn what technologies to keep up with.

Embarking Now...

While all professions evolve over time, few have experienced the rapid and radical changes the IT profession has undergone over the past decade. Given the recent shifts and the unique position of the IBM midrange platform, many midrange professionals are attempting to plot a new career course. Some may continue to follow IBM’s lead by embracing the technologies it ports to its midrange systems. Others, like John, may wonder whether they should hedge their bets by carefully choosing technologies that are platform- independent or require skill sets that can be leveraged onto other platforms. Still others may embark on a more detailed review, opting to move to new challenges—perhaps to other platforms, perhaps outside of IT.

Whatever course you choose, Midrange Computing will strive to give objective information about developments in the midrange world. And we wish you fair winds and calm seas during your voyage.