Tomorrowland: Optimism, Risk, and Preparation For IBM i App Dev
June 28, 2017 Dan Burger
The rules of application development were never carved in stone. They’ve always been subject to change, improvements, enhancements and even replacements. Uncertainty and risk – do you stay or do you go – torture the brains of long-range planners who recognize the importance of preparation, but are caught in the whirlpool of innovation and hyperbole. Predicting future skills and tools is not an exact science. But with a high degree of certainty, we do know that sitting on your hands is no way to prepare for the future.
I asked a group of skillful IBM i application development monitors and mentors for advice on making app dev preparations for the future. They share a deep understanding of IBM i application development technical details and a recognition of the importance of developer productivity and organizational investment in technology and workforce skills. They are also connected as part of the team of instructors assembled for the RPG & DB2 Summit conferences that have occurred during the past 11 years.
Alan Seiden, a PHP on IBM i consultant, open source advocate, facilitator, and problem solver, sought input from folks on his Club Seiden crew. They provided the following insights about where we are with IBM i app dev and where we’re headed, touching on the topics of sharing resources, the use of containers, modernizing and refactoring, and new talent.
“Copying and pasting of RPG code is a time-honored practice within IBM i shops, but by limiting ourselves to copying past patterns used in our company, we won’t advance,” Seiden tells IT Jungle. “I see a trend toward community-‘curated’ code resources, snippets and functions that would help to make IBM i programming become as accessible as on other platforms. RPG, CL, and SQL snippets and functions are available online today, but in a somewhat scattered fashion. The ideal would be a clearinghouse of vetted, reliable code snippets and functions in all styles of RPG from old to modern, for example–call it CopyPasteOfTheFuture. SQL snippets are key also.
“Young open-source developers, especially on Linux, have become accustomed to developing within ‘containers’ that provide a consistent, self-contained environment for every developer. An example is Docker. While IBM i has chroot today to provide separate environments in PASE, and doubtless there would be technical challenges, five years from now true container capability would help make the new generation feel at home on IBM i.
“Some say IBM i modernization is the new Y2K, an imperative to keep IBM i applications relevant and IT responsive. Every company should have a strategy to keep their code fresh and modular, and therefore agile, to respond to new business requirements and new UIs. Smaller shops have a bigger challenge from a manpower standpoint, but the key is to have a strategy, even if it’s a gradual one.
“The IBM i community has already woken up to the fact that we cannot keep recycling the same people in IBM i roles. We need a blend of experienced and new talent. At every conference, companies are discussing how they take responsibility for creating and training the next generation. IBM has a role here, but it’s time for private companies to realize they must reclaim their traditional role of training new people on the platform. I predict that a set of best practices will evolve to create new IBM i talent. In my experience, millennials often find IBM i very interesting and can work very well when they have the resources, training and mentoring they need.”
Liam Allan is one of the bright young and innovative talents in the IBM i community. He’s an IBM i Innovation Award winner, a Power Systems Champion and an open source on IBM i advocate. His view of what will become important for future app dev on IBM i was briefly stated as the use of containers (used for packaging applications) and the incorporation of DevOps (collaboration that involves blowing up individual siloes and integrating tools and processes).
Most IT Jungle readers recognize Ted Holt as the senior technical editor of Four Hundred Guru. He’s a published author of books and articles, taught at the university, community college and vocational-technical school levels, and worked in the information industry for more than 30 years. For years, he has said developers have no choice but to learn programming beyond RPG, CL, DDS, PDM, SEU, and SDA.
“The need for programmers in the industries that the IBM midrange serves has been gradually declining over the past decades,” Holt says. “A few decades ago, users had to ask a programmer to write a program in order to view their data in a new way. Nowadays, we give the users access to the data and they use Excel to format it as they wish.
“Another reason for the decline is that the quality of packaged software has improved. Many organizations that run packaged software have less need for customizations and add-ons. The fact that users are less reliant on programmers has freed programmers for more important work, notably modernization of applications and integration with other systems. As much work remains to be done in these areas, the demand for programmers will continue to be steady for some time.
“Eventually computers will program themselves or each other, but that’s still many years into the future. By then I will be dead or, if I’m lucky, working in the garden department at Walmart.”
One of the most highly regarded educator/trainers in the IBM i community, Paul Tuohy has helped people around the world with application modernization and development technologies. He’s the author of books and articles and heads up a consultancy firm based in Dublin, Ireland. In his opinion, the two biggest influencers on the future of IBM i development will be young people coming to the platform and open source development.
“Young people bring a fresh perspective and any lack of business knowledge is more than compensated by enthusiasm and energy,” Tuohy says. “The challenge is to harness and direct the enthusiasm/energy while passing on the business knowledge. At the same time, us old fuddy-duddies have to embrace their fresh perspectives which provide new ways of looking at old problems.
“I think the IBM i community will be in a state of flux five years from now. It is going to take a long time to modernize all of those existing applications, and I have a funny feeling that some of them will never be modernized. Developers will be working on applications that have RPG/COBOL at their core in conjunction with one or two of the open source languages.
“The traditional RPG development will probably change to work more along the lines of open source (e.g. version control). For the small, one-man shops, this means they have to come to grips with the new environments or the company will have to contract that work to be done by service providers. The bigger shops will combine talents. This is one of the benefits of free form RPG. All the “diverse” programming languages are no longer that far apart.
“The development paradigm will be no more complicated than it is today. The difficulty will be in picking what works best. Just because there are six or seven open source languages does not mean that you have to use them all. You just have to pick the one(s) that work best for you.”
I’d be surprised if anyone reading this doesn’t know Jon Paris. If they haven’t met him, they’ve read what he’s written or what others have written about him. He and his partner Susan Gantner (coming up next) are the most widely recognized educators and trainers in the world. When I asked Jon to look ahead five years, he provided perspectives on what he hoped would be happening and what he feared might be happening.
“I hope that the uptake in open source will continue to add vitality to the platform as younger developers ‘discover’ it and just how nice it is to not have to spend your life load-balancing and rebuilding databases,” Paris says. “The fear that goes with this is that IBM will get ‘bored’ with promoting open source just as it gets close to reaching critical mass.
“I hope that more work will be done on educating folks in ‘why’ open source is important and to providing instructions and education in English (or German or other languages). The accompanying fear is that IBM will continue to document everything in Unix-ese and that only those who already understand the tooling will be able to take advantage of it.
“I hope that more companies will come to realize that they don’t need to try and find RPG programmers to fill all vacancies – that they can hire good business-oriented programmers and teach them RPG. Along with this goes the (less hopeful) hope that IBM will put some effort into this area to help provide training resources for companies seeking to hire just one or two such programmers. The accompanying fear is that hiring will remain in the hands of HR, who will continue to try and find candidates who simply don’t exist – at least not at any salary they have in mind.
“My final hope is that RPG will continue to improve and that along with that will come more affordable and reliable tooling to automate the conversion of old RPG/400 and RPG/36 code, so that those dinosaurs will no longer be a barrier to moving ahead. My fear is that too many managers will continue to thwart the efforts of their progressive staff and stick with the ‘it was good enough for me when I was a programmer’ attitude that has doomed so many shops to old versions of RPG coupled with green screens and SEU (shudder).
“If my hoped-for scenarios come to pass, I see the platform continuing for many more years. If all of my fears are realized, then I suspect a slow and painful death–not perhaps in five years–but by then the writing will certainly be on the wall.”
That brings us to the viewpoint of Susan Gantner, teacher, consultant and advocate for the latest programming and database technologies on IBM i. Everyone in the IBM i community seems to know her name and a great many have benefited from her classes and expertise, particularly Rational Developer for i (RDi). Her perspective on the future of IBM i application development is influenced greatly by her hopes for what will happen.
“Of course, I feel confident that RDi will be the primary development platform for RPG applications in the majority of IBM i shops in five years’ time,” Gantner tells us. “I believe it’s already the primary tool used in most shops using modern RPG coding techniques and practices. Part of this prediction is based on my anticipation that more IBM i and RPG shops will wake up to exploiting what the language and the platform can do to modernize their applications so they become more valuable business assets.
“Five years from now, I’d hope that relative youngsters such as Liam Allan, Kody Robinson and Lynell Constantine will no longer be outliers in IBM i and RPG shops. I’m not just referring to their younger ages – although a significant lowering of the average age in the RPG community is a must if the language is to have a bright future. I’m also referring to the fact that they are comfortable with other languages alongside RPG, as well as environments and application styles such as Web and mobile. And just as importantly, they help to nudge those of us decades their senior to expand our own thoughts and attitudes about how development on this platform can and should be done.
“I believe we’ll begin to see more blurring of development roles in IBM i shops—for instance, there will be fewer shops where an RPG developer group is separate from the Web, mobile and other modern stuff group. RPGers will not be just RPGers because they will also be equally comfortable doing Web, mobile and other modern interfaces, while taking advantage of open source tools, languages and code where it makes sense. Not to replace RPG, but to augment it.
“I do think the pace of change (blurring development roles) is accelerating. And the difference between those shops who are moving and those standing still is becoming greater.”