EGL: At Least It’s Not Java, But It Ain’t RPG, Either
September 24, 2007 Brian Kelly
As soon as I read the article, EGL: The Future of Programming for the System i?, from last week, I knew there is nothing I can do but complain. I am upset–very upset, in fact–that I have to use PHP and the other stuff that is used on other platforms in order to get my System i shops to the Web. It did not have to be this way. From years of observation, I know that IBM is very stubborn, often arrogant. And when IBM digs in, it is usually incorrect. In choosing to create a new language for RPG programmers to learn after the sound rejection IBM got trying to push Java, is EGL in my future? No way, Jose!
Again, despite the pleas of its loyal constituencies, Big Blue has chosen a programming language that nobody has requested–other than a few leftovers in its mainframe division who are still trying to prove the mainframe is not dead. We know the mainframe is not dead–I saw it walking yesterday with Abraham Martin & John–and it was guzzling a lot of electricity. It would be better for everybody, with the exception perhaps of mainframers, for IBM to prove something else this time and stop getting System i developers involved in their one-size-fits-all solution mix, led by stale mainframe technology.
When I look at EGL, I see the full hubris of the omniscient and omnipotent IBM. I see the IBM that singularly knows what is right for its customers despite the cards and letters and downright pleas of its loyal base. I’ve seen it before. I saw an IBM that knew that its System/36 customers would abandon SSP in droves to get to OS/400. It took this IBM six years to reverse that guffaw, but at least it did so.
To acquire more market share, I saw an IBM that chose to make the scads of game cards and special device cards for ISA PCs obsolete in 1986 when it introduced its new PS/2 line–a product line that was intentionally compatible with nothing. Additionally at the same time, I saw an IBM that thought its once loyal base would go for its expensive mainframe-oriented MicroChannel Architecture in its PS/2 servers and that, like lemmings, they would just buy their expensive attachment cards all over again. But, wait, I also saw an IBM that did not know that nobody was making those attachment cards. Well who knew that? And then there was OS/2–but I have already made my point. The PC guffaw lasted over 20 years and in the end IBM could not reverse it. (Ironically, Lenovo, having bought IBM’s PC business, might be able to.)
Besides having superior mainframe hardware and software, the most exciting and the most correct thing IBM ever did was build the System/38. Removing performance barriers with faster processors with the AS/400 was another one of IBM’s finest moments. The AS/400 was crowned champion of the minicomputer world and in fact eliminated the minicomputer part of most companies that were in the foray. A regrouped HP with some DEC innards is all that is left of IBM’s minicomputer competition. As the industry shifted to microcomputer technology and industry standard chips, HP was not going to let IBM win again. Using the strength from its Compaq acquisition, HP so far has not only survived, but is beginning to thrive. Moreover, in November, 2006, for the first time, HP passed IBM and became the largest IT company in the world ($91.7 billion vs. $90 billion for IBM). More importantly for its shareholders, HP got to the top while improving profitability across its business lines — including the PC line.
For the last several years, important IBMers have been snookering me about the probability of a natural GUI system interface and a natural Web-oriented RPG language. At one point, they actually told me it would be here in this next release of i5/OS. I believed them, but I shouldn’t have. The handwriting was on the wall when Dave Slater stepped down from IBM’s top System i software post within the last year. There would be no Web GUI for RPG.
Each time IBM announced something, I kept wondering if between the lines this was the natural browser-based interface to RPG? No, Virginia, EGL with RPG hooks is not the natural browser interface for RPG that I have been badgering IBM’s System i and Rational executives to bring to the System i, but it does seem like the start of a repeat performance of the AS/400-Java debacle from which just recently it seemed IBM had recovered.
IBM’s Rational division in Toronto has made it perfectly clear that this often requested, natural RPG Web interface is not coming and that is that. The company will ask RPG shops to write in yet another “new” language, EGL. At least it is not Java. But it is not RPG, either. Maybe there is a book opportunity here for somebody. How about say, EGL for RPG Programmers? There’s a book that would not have to be written if IBM listened to its customers and delivered what they wanted.
For the record, EGL is a warmed-over 1980s mainframe code generator that had once been known as Cross System Product. Since mainframers like COBOL, the language will produce COBOL and it will also produce Java code for those in love with Java. It does not apparently matter to IBM any more what System i programmers like. I would argue that the Java generation output is in place because Big Blue has still not gotten the message about how much System i shops like Java. Java as a Web solution on System i is dead. Get used to it. Another thing is for sure–and I finally believe it–RPG programmers have been forgotten again by IBM. Not only will RPG not ever get to the Web, but there will be no EGL-generated RPG I, II, III, IV, or V. Just COBOL and Java.
In the latest pre-announcement rumblings, the search for an RPG Web solution is over. EGL has been anointed by IBM to be used instead of RPG on the Web. IBM’s says that shouldn’t matter since Rational does not want anybody changing their generated code anyway. If that is really the case, couldn’t they have saved even more development dollars by skipping COBOL and Java and going right to pure machine language?
For the embattled, long-disenfranchised RPG shop looking for a simple way to the Web, the IBM Rational message is immediately understandable. Don’t look to RPG, RPG IV, RPG V, or any RPG language for a lift to the Web. RPG will never be capable of producing modern Web-based applications. After all these years of promises and hints from Big Blue that it was finally going to do the right thing by enhancing RPG with a leading edge GUI facility, the deciders have decided. It is not going to happen.
Thus, to move forward, developers must look outside of IBM for solutions. Thankfully, there are lots of them. One thing is for sure. There is no longer any reason to wait for IBM to announce anything.
Brian Kelly is an assistant professor in the Business Information Technology program at Marywood University, where he also serves as the System i technical advisor to the IT faculty. Kelly has developed and taught a number of college courses in the IT and business areas. He has an active consultancy in the information technology field, Kelly Consulting. He is the author of 27 books, has written numerous articles about current IT topics, and is a frequent speaker at the COMMON and other technical conferences and user group meetings across the United States.