Have you heard about the article indicating salaries for IBM i types are beginning to fall behind their Windows/Linux brethren? Apparently, it's true. And it makes me wonder why.
A recently published article indicated that, compared to those of Windows and Linux platform professionals, IBM i salaries are lagging behind. Gad!
I certainly have no problem believing Alex Woodie, the author of the that article.
One reason it is easy to believe this is that, in many ways, the i is the second "other brother Darryl," always semi invisible to the tech world.
The Robert Half Salary Review for 2017, for example, breaks out all kinds of web- and hardware-oriented jobs but does not mention the IBM i at all. In fact, the data on IBM i salaries in Woodie's article was provided by Bob Langieri of Excel Systems from Excel's own company data.
So if we accept the proposition that i salaries are falling behind those of Windows-oriented jobs (which I do), the next question is, why? After all, there have been plenty of articles indicating how concerned i managers are about the supposed scarcity of RPG resources.
Personally, I always thought price was related to supply and demand. As the supply gets smaller, the price goes up. That's the way it works with lemon pie. If you are the only place within 50 miles that has some, then you can charge $6.60 or even $8.80 for a great piece of lemon pie. But if the guy next door has it too, then all you can get is $3.59.
But that doesn't seem to be the way this is working. For IBM i talent, the opposite seems true; as the supply goes down, so does the price people are willing to pay for that talent. What's up with that?
In Woodie's article, Langieri indicates the reason for the disparity in salaries is a lack of demand for i skills. Specifically, he referred to the fact that in Southern California "at any given time there are fewer than five openings for people with IBM i-RPG skills." I don't doubt that, but...
Does That Really Cause Salaries to Stagnate?
I have to ask the question. Is that really the reason salaries are lower? I have seen only a few openings for a Senior VP of Marketing over the past few months, but I am pretty sure they are getting top dollar. In the end, the question of why a salary is placed at a particular level is a complicated one, but at the core is how important that position is to the future of the company.
Unfortunately, Langieri was unable to talk with me for this article, so I don't know what his opinion on this is. But it does seem logical to ask...
Is There a Wholesale Shortage of RPG Talent?
Let's start by looking at this whole question of whether there is or isn't a shortage of RPG people.
I think that when we talk about a talent shortage, we tend to think about RPG programmers across the board. I do believe, based on personal and anecdotal evidence, that many people who try to hire an RPG person do have some problems finding someone. And I think there are a couple of reasons for that.
First, yes, there are not a lot of RPG people out there. And, frankly, they don't tend to move around a whole lot. At the same time, there are not a lot of RPG job openings, so it should all balance out. And if you do have trouble finding someone, that shouldn't push you to offer less money. That is, what you basically have is a pool of experienced RPG people who are already somewhere. You don't have a lot of young, new, unemployed people entering the arena. So, to entice someone to move, you have to offer them a real incentive. And generally, that is money.
Second, there seems to be a certain lack of imagination in how companies search. Most of the RPG hiring attempts I have seen are pretty traditional. Put the notice out on a job board or give it to a headhunter and wait for the resumes to start pouring in. And, if you don't get many responses, then there must be a shortage of i people. Right? One company I know tried this approach, and it took a fair number of months before they found someone.
But there are a few ways to find RPG talent beyond just waiting for someone to swim by. Strangely enough, the company I just mentioned was located only 20 miles from Char Parker's RPG program at Muskegon Community College. Parker has a full curriculum that teaches not only RPG but the i as well. Wouldn't it have been easier for this company to check out the college for a potential hire? Now you can say maybe they didn't know about it, and that's probably true, but my response is that ignorance of the resources within reaching distance is no excuse.
Nor is Parker's program the only one of its kind. Everyone in the i world is, or should be, acquainted with Jim Buck and the work he has done with RPG education. Or Bryan Meyers.
Granted, Muskegon Community College is probably the exception, not the rule, but I am also sure it is not the only two-year school dabbling in the i. Do you know if one near you has such a program?
Plus, online courses are available from a number of outlets, and the fact that we don't use those as a resource leads into another reason that it seems harder than it should be to find RPG talent.
Third, failure to develop resources the way we used to.
You know how most of the current old RPG hands were found? They were hanging tapes on the mainframe or coding COBOL, and some farsighted manager walked up to them and said, "You're going to learn RPG. I need another programmer."
Yes, back in the olden days - before people lived in houses and when dinosaurs ruled the earth - we promoted and trained from within. Of course, today we don't have anyone hanging tapes (or coding COBOL), but every company has people who are quick learners and would be happy to break into IT. Or maybe they are already there and want to learn something different.
Let's face it. It doesn't take that long to learn RPG, even RPG IV. Couple of months and you're solid. It's not nearly as hard as Java or even JavaScript, where it can take up to a year to get good. And training is so much easier today with some of the online courses available.
Web shops do that all the time. They bring in someone they like, and he doesn't know Angular (maybe he's a React guy). No problem. Give him a little time and some mentorship, and he will learn it.
But we don't do that. It's like we need to have someone who has been somewhere else so that we are sure they are the real deal. Is that a tacit admission that current management doesn't trust their own judgment? I say start with someone you know you can count on and train them to do the job.
Fourth, failure to embrace remote resources.
IBM i shops seem to stand alone in their reluctance to use remote resources. Once more, the web world makes extensive use of remote workers. You can go out to Remotely Awesome Jobs or Working Nomads any day of the week and find a dozen remote jobs for front-end developers, JS coders, Ruby freaks, etc.
But almost without exception, i shops say no to picking up experienced programmers who don't want to move but are willing to work remotely. What's the lesson here? That web programmers can be trusted to work without supervision while RPGers can't? Oh, now you're just trying to make me mad. Most of those web people are millennials, you know.
In the end, it seems that even if there isn't a scarcity of RPG talent, many i shops take a route that may make it seem that way. Yes, it is probably harder to find a bucket of RPG people than a bucket of JavaScript people, but it's definitely not impossible to find or make one of your own. And seriously, how many do you need?
But are sheer numbers what we're talking about when we say "shortage"? Maybe not, and maybe it is not just "an RPG programmer" that is really in short supply.
BOP Programmers
The sad fact is that there are many, many i shops where no new development is going on for RPG. At best, people are taking the existing RPG programs and redoing them on the i, probably with Java.
Most of the activity for RPG is centered on maintaining the existing RPG programs. And when we say "existing RPG programs," that means BOPs: Big Ol' Programs. I'm talking about fixing or enhancing large programs that have been there for 30 years, where 43 percent of the lines are commented out, where every variable is global so that the subroutines just make more spots you have to look at to track something down, and where the program interacts with or feeds into another 20 programs for a massive batch run glop.
Currently, this work is carried on deep in the bowels of these i companies by wizened old geezers who are one cigarette away from a massive stroke. To them, the BOPs are not monster programs, but rather a crazy cousin everyone can joke about at family reunions. And that is what is in short supply. Someone who can walk in and take over from somebody who has been taking care of a BOP for 30 years without missing a beat.
Obviously, no one but the person you have now is going to be intimately familiar with a given BOP from day one. But you can teach people the skills they need to deal with things like this. And there are software packages that can help you untangle the BOP.
There is one thing that management has to keep in mind, though. Instead of maintaining BOPs, we should be breaking them up into modular, free-format ILE programs. In the end, you can't blame people for not applying for jobs that use obsolete skills. Any job seeker tempted to do that would be better off to transition to a new language than to put his job hopes firmly in the past.
And, once again, lack of manpower should translate to high salaries. Yet it doesn't. I think there are a couple of reasons for this.
The Status of the i in General
I believe the i is done as a development machine. And I believe that a computer only gets street cred for its stripes as a development platform.
Certainly, IBM seems to be spending most of their time pushing it as a master server, an alternative to the Windows/Intel crowd, a replacement for the 50 servers you currently have in your backroom.
I know that a number of languages have been added to the i repertoire recently, but I question whether that is enough to reposition it as a major development platform.
The sad truth is that it just took IBM too long to get free-format, ILE, and RDi fully sweet, and a lot of companies gave up and moved on. That, along with the fact that there is still not a cheap entry machine for the i, means it will not get serious consideration from anyone looking to expand their development work. I know all about TCO, but we all also know that upfront cost is king, whether it makes sense or not.
Of course, when we talk about the i, we are talking about not just hardware but the operating system and development tools. And, if the industry sees this as stagnant or outdated (I know it's not, but perception is everything), then it is going to be increasingly difficult to ask for money to staff it.
The Status of RPG
Another part of this is the status of RPG itself.
Right now, as stated above, BOPs predominate, and IT work consists of mostly just keeping them running and updating them as business needs change. What should be happening is that the old hands are working on breaking them up into modular programs. There is absolutely no excuse for this not being underway. On the web side, many developers jump all over any new techniques, but on the i side, we are strangely resistant.
The result of not moving, even slowly, to replace these old BOPs is to make the i look even more dilapidated. And, in the end, it exaggerates the question of whether the i is worth saving, given how hard it is to find resources.
But it's not just the programs; it's RPG. I have to admit I came to RPG (from COBOL) kicking and screaming. But after many years, I have learned to appreciate the brevity that RPG offers. And now that free-format is here, it beats the heck out of web languages for readability.
But to many people - many management people - is at the heart of what is wrong with the i. They do not understand RPG, they can't tell you what its strengths and weaknesses are, and generally they don't want to seriously listen to someone else tell them. I believe that in many i shops it is almost dogma: We need to get rid of the RPG programs for our system to be truly modern.
The truth, of course, is that BOPs are what we need to get rid of in order to have a truly modern system. I would put modular RPG up against any web language anytime and anywhere in terms of readability, maintainability, development speed, and execution speed.
The People Who Are Running It
At the same time, I think we need to look at the people who are managing many i installations.
Back in the olden days, which is what most of us IBM i people remember best, the person who ran the show and paid you was probably an IBM mainframe or S/38 guy. He understood that environment and valued what it took to deliver the apps that kept a lot of companies running for decades with minimal modifications.
But over the last few years, a lot of those old hands have retired or passed away. In their places, companies brought in people whose primary background was with...what? Yep, you guessed it. Windows. In some ways, I think this makes sense. The typical i shop today has both technologies (and I give you one guess as to which one is most problematic). But at the same time, I think it is very hard for someone who has grown up in a Windows world to understand or appreciate what the i brings to the party. To them, the i is just a box that runs without much assistance and supports ugly programs that happen to run their business.
And the impact of this on salaries?
Deciding how much you are going to pay people is an emotional issue. Hellfire, folks. Most business decisions are emotional in nature. And who are you likely to view as most deserving of a higher salary? People whose jobs you understand or i freaks?
Do They Really Want to Find What They Are Looking For?
In general, I.m not into conspiracy theories, although I do kind of like the one that claims the top leaders on Earth are all reptiles in disguise. But I don.t see that being a factor here.
Seriously though, there are times when I wonder if some (or even many) i managers don't really want to find RPG people.
Everyone will deny this, call it crazy, but maybe this is one of those things that not even they are aware of.
We have already seen that many current i managers are not i people, but rather Windows people who have a new job and now find themselves in charge of an i. If, as we have already postulated, many new i managers come from the Windows arena, and you were one of those who suddenly found yourself the owner of an i, what would you most want?
If we continue with our train of thought from above, and you have no interest in really understanding the i but just want your days to be pretty much like the days at your previous, non-i job, then the obvious answer is that it would be nice if the i would just disappear.
You are not going to push for that openly. After all, the i is running all the programs that your business runs on, and there is a lukewarm corporate commitment to it. But the simple fact is, if you can't hire any new i talent (and that means RPG), then porting those programs over to another environment would become an option. So maybe subconsciously you don't try so hard to find new RPG folks.
I realize that many people reading this article will think I am crazy. But I am betting there is a solid 40 percent of you who are wordlessly nodding slowly in agreement.
What Does It Mean?
As I said at the beginning, I can easily believe that i salaries are lagging behind comparable Windows positions.
But I am not sure I believe that the reason for this is that there are not many job openings for RPG people. There are other factors at work here. And those other factors might explain why there aren't many RPG job openings and why the pay for those that do occur seems frozen.
Is it the fact that most i shops are still stuck firmly in the past? Or because many people in i management don't value the i compared to Windows? Maybe it's IBM's fault for taking their good old time getting RPG modernized? Or perhaps the conspiracy theory is right and we just aren't trying very hard to uncover new i talent?
It's hard to say, but I do believe the salary gap is a fact. And I believe it is a gap that will continue to widen as the years go by.
LATEST COMMENTS
MC Press Online