Reader Feedback on The X Factor: Head in the Clouds
February 16, 2009 Timothy Prickett Morgan
Good, thoughtful comments from readers of The Four Hundred are always welcome. My suggestion a few weeks ago, in an article entitled The X Factor: Head in the Clouds, that someone–meaning you people, certainly not me–should get RPG running out there on cloud infrastructure like Amazon‘s EC2 cloud, got one reader thinking about what the real issue with clouds and open source RPG compilers is: the apps, not the tools to run them.
Check it out.
Why would someone programming for a cloud use RPG?
I would agree that some people are stubborn and their preference for a programming language has prevailed. Indeed, it’s this stubbornness and a somewhat child-like blind faith in their vendor that has led a vast number of System i companies up the creek without much of a paddle.
Over the last year, I have been bringing together a disparate group of companies and individuals, with a view to providing some very specific tooling for System i development. To achieve this, we must use a model that captures the open source community model, some serious AS/400 pundits, and one or two existing vendors in this space. Your own publication ran an article about the early stages of this last year: Databorough Teams with Genuitec to Push Alternative Eclipse IDE.
This whole thing is jollying along at a nice pace, with some interesting and unpredicted deliverables now being planned, as a direct result of bringing in the open source/pundit community or iTeam (an embarrassingly corny name dreamed up five minutes before the first meeting we had last year). The iTeam’s enthusiasm for a new business model, and the voracity to use it effectively against the main protagonist, has been surprising and also very encouraging. Stick it to the man themes always get some support, but this one seems to have a rare combination of will, vehicle, and timing all in simultaneous alignment. It might actually succeed!
Anyway your idea of slapping down a million bucks to seed an open source version of RPG is right in keeping with type of discussions that take place between members of the iTeam and others. I would however like to add my 5 cents worth of why my colleagues and I don’t think an open source RPG compiler would happen, or even should.
I realize that IT Jungle is passionate about its support of the RPG ideal. Databorough and the iTeam are, too. If RPG magically ceased to exist tomorrow, we would all be out of work, as the majority of our revenue depends on the fact that the System i community has lots of messy old RPG to deal with on a daily basis.
For at least a decade now, we have been looking at what might happen to the RPG community and ways to capitalize on those events commercially. On the one hand, we invest significant percentages of revenue into analysis and research of new technologies, and on the other we are forced to deal with the real RPG world, in terms of the language itself (and all its vintages), how it has been used by the conservative, adventurous, 4GLs, and hackers, in corporate systems across the globe. So when we assess a new technology in the context of its relevance to the System i world, we are afforded a fairly unique perspective.
Tackling the pure language question first. RPG has been evolving steadily over the last 30 years or so. Unfortunately, it hasn’t evolved quickly or far enough and, if I am to believe people I have worked with that were on the original ILE team at IBM, that could go down to a single management decision back in the mid-1990s on how far to take the language. If that evolution were given a shot in the arm and a strategic future by initiating an open source community to develop it, as you are suggesting, what would it end up like? Probably like C# or Java in all likelihood. Of course Java has the benefit from an 18-million strong active community, and even if only 1 percent are actively involved in some way in evolving the language via open source, that’s still at least 180,000 brainy people making it more useful. RPG on the other hand probably has a conservative estimate of 150,000 developers actually using it, so the statistical probability of it achieving similar success via this model is unlikely. You could perhaps argue that many vendors and System i companies would have an incentive to see this open source RPG venture succeed. Why? Because they could port all their existing code into it in a flash and off we go again? Unfortunately, therein lays the biggest hoax of the 21st century so far.
Existing RPG applications are messy, inconsistently coded, and have vast amounts of duplication and redundancy built in. Some 20 or 30 years of efficient coding by conservative, adventurous, 4GLs, and hackers will do that for you. Natural architectural erosion aside, these legacy applications are very risky and costly applications to maintain and develop. Taking the same code base “as is” into a new development and deployment model, while simultaneously rapidly evolving the actual language itself, would be like inventing bungee jumping while you still had rope and hoping the correct elasticized composites will be developed and in production by the time you hit the end of the rope. I would even go so far as to say that a more binding and long term aspect of applications written on System i is not what language they are written in, but what they do and how they do it–that is, their design.
If a company is going to take that much risk they might as well achieve their real objective, and that is to clean up the application by rewriting it or reengineering it.
So if you do get anyone actually offering you that million dollar check, I would suggest investing it in these growth areas:
If, however, you do succeed and RPG becomes a mainstream language again, many, many thanks in advance from all of us at Databorough!
–Stuart Milligan, vice president of business development, Databorough