IBM lifts lid on latest bid to halt mainframe skill slips IBM is pinning its hopes on some fresh initiatives – the Mainframe Skills Council and the IBM Z Mainframe Skills Depot – to address a shortage of engineers who have big iron expertise . Big Blue revealed the Mainframe Skills Council at the SHARE conference in Orlando, Florida this week, describing it as a forum to bring …
I'm 84 - last wrote some code around 3 years ago but still love to dabble - I did migrate to PC's & Java partially but never left the big box - being able to write on both sides of the system world was a great advantage - they often left me alone to conjure up solutions - had great fun in my final years ! Fights over who owned what & the snobbery of the PC side was pretty funny, but they all lost their power anyway when the whole company was reshuffled in a takeover - I had just retired so it did not affect me. Moral of the story - never become too 'important' - your job can always disappear tomorrow !
It is highly apparent that, by the necessity of IBM creating the MSC in the first place...
that IBM's management is INCOMPETENT.
During the past number of years IBM has pretty much concentrated its effort in retiring / removing older, skilled workers - remember the "dinobabies" comment? But *now* they realize that the workforce with the required skillset is aging and they must train-up replacements.
If your management *wasn't* incompetent you should have realized that the older workers held decades of accumulated knowledge and created an in-house apprenticeship for the Big Iron workforce, getting those new(er) recruits up to speed with hands-on experience in association with your experienced techs. THEN, after they are trained-up, you consider asking *some* the techs if they would like early retirement, making sure you keep enough of the Old Guard on-hand to assist the newer techs as they, themselves, age into the SuperTechs you need to keep your Big Iron going indefinitely.
But you didn't do that, did you??!! You started pushing your skilled techs out BEFORE YOU REPLACED THEM, YOU IDIOTS.
This is the kind of incompetence that, frankly, should be brought up in stockholder meetings, questioning "Why are we allowing 'managers' to keep their jobs when they apparently can't make wise, foreseeable business decisions?". But I'm sure that won't happen: the mutual funds that hold the greatest sway in stock shares will keep their mouths shut as long as IBM's said management keeps making promises, even if they never actually meet those promises, of quarterly dividends.
And nothing in End-stage Capitalism ever gets fixed.
Welcome to the Wonderful World of Stupidity. You're gonna like it here.
It is quite ironic, isn't it.
The company that's main in the news for laying off staff, at least here, is pursuing an initiative to retain a specific workforce.
Maybe the managers simply don't know the product. Maybe they aren't aware that it's the old iron that earns significant bucks for the company, while they all are using only the cloudy stuff. Or maybe they're just plain incompetent. Apply Hanlon's, Ockham's or any other razors at will.
>Serves IBM right for being so dismissive of any employees with age and experience.
That's the cunning plan.
By re-hiring the experienced engineers they fired made-redundant force choked reduction in force'd as new hires they can be paid new hire rates and new hire benefits while maintaining all the experience and institutional knowledge
Why can't those workers just see the beautiful simplicity of the plan ?
"You started pushing your skilled techs out BEFORE YOU REPLACED THEM"
The concept is that "technology moves really fast" so if it's more than a couple of years old it's "legacy" and irrelevant. Consequently, skills are considered to need constant "updating" (which is quite possibly why, incidentally, software needs constant updating). There are no such things as first principles or theory -- it's all about hands on practice. Funnily enough, this mindset was what primarily destroyed the Soviet industrial base. They sacked (or in that case shot) practically all the experienced engineers and managers and replaced them with others who had zero experience.
"...questioning "Why are we allowing 'managers' to keep their jobs when they apparently can't make wise, foreseeable business decisions?""
To be fair, most managers have zero power, influence, budget or freedom to make decisions of any scope. Executives are the ones with the power, and even then not all are created equal; you have to go pretty high to find somebody who can make strategic decisions versus day-to-day.
Agree having access to an “mainframe”, particularly “free” might encourage more universities to give mainframe computing more coverage, and maybe help rehabilitate (modern) mainframe computing amongst Computing undergrads who may be under the impression that Windows/Linux and x86/Arm are the only system platform options worth working with.
IBM did try that in 2001 by donating a mainframe to Warwick University ( https://12ft.io/proxy?q=https://www.timeshighereducation.com/news/ibm-puts-warwick-ahead-in-grid-quest/156954.article ) as part of the Grid initiative. (Did anything ever come of that initiative?)
However, only a few years later the mainframe was decommissioned (there's a photo online somewhere of it being unceremoniously shunted out of the Computer Science building, presumably on its way to the Land Where Legacy Hardware is Eternally Blessed), and my suspicion is that the closed nature of the platform stood no chance in an environment where Linux and open source was exploding in popularity. The mainframe must have appeared to be as incongruous as a headmaster at a school disco.
The truth is it has been very unfashionable to consider mainframes as worthwhile computing systems for several past decades, and because there is such an enormous amount of ancient but still-in-use code out there nobody wants to get involved because it's just not 'glamorous' - bad enough that there are online systems with nasty mainframe front ends when the guy in the next cubicle is having fun with his pretty graphical interface - but if you have to actually learn that stuff to maintain it, there is just not the willingness to 'go there'. I maintained both pc & mainframe code, but when maintenance is needed you have to act in a small company with ancient in-house systems. They generally are trying to move things to new tech but the business has to run first so that is the priority NOT the new fancy systems the programming people want !!!
" if companies valued their over 50's workforce they might stick around for a bit longer"
but unfortunately, they cost more per capita than new youngsters, particularly if they've been on the staff for any significant time (accrued pensions etc.). As everyone is now merely a "human resource", related overheads "must be minimised".
It's also not a good look for younger employees. Why would you invest years in learning certain skill and then you'll be binned at 50. If you don't find alternative job in the field after that, then it is potentially 10 or so years wasted where you will not be having income you should be having.
So if you look at your career and think ok I learn this and I will be pulling at least middle six figures since I am 30 through 40 and 50. If 10 years can potentially fall off the plan then you have to adjust your lifetime earning and then you may realise the offer becomes not viable.
You may find you may be better off doing some less demanding and less paid job and have more time for family and hobbies.
Bingo. My mother wanted / expected me to work for IBM after school but I didn't want to get involved with a mega-corp and end up being just a number.
Turns out I was right -_- I'm in my late-50's and would probably be on the street if I went the IBM path. Sure, sure, I would have earned big bucks for a while, but only through high stress, constant expectation of continued learning (usually on my own dime), and the bullshiate of megacorp office politics.
No thank you.
> Why would you invest years in learning certain skill and then you'll be binned at 50
Well for some reason for several decades people have been learning certain Microsoft skills, only to have them binned with the next release of Windows etc.
Basically, I suggest you should looking to develop skills so you can have a career post 50…
If you go to that site IBM Z Mainframe Skills Depot
and then as a prospective future Mainframe magician you want to see "what's in it for me?" and go to Benefits section you can see that there is sweet f-work nothing.
If they want to attract people into this, they should state if after completing the training person will be:
- hired once passed competency test
- able to afford a house with a garage and garden within a year and close to place of work and comfortable living
- getting stocks
- additional training
- getting generous severance package if company decides to bin the project
Without that, why would you invest your time into it?
* monthly RA's (Resource Actions aka redundancies)
* yearly PIP (Performance Improvement Plan)
* always working in the office
* moving office location every 6 months
* lots of unpaid overtime
* world leading bureaucracy & red tape
* training your replacement in a low wage country
* seeing Ginny's helicopter (no touching)
This post has been deleted by its author
* monthly RA's (Resource Actions aka redundancies)
Yearly
* yearly PIP (Performance Improvement Plan)
Not a regular thing, it might hit you once if you're a fuckup. And you're generally shitcanned after regardless of the outcome, so 'yearly' is bollocks. You might be thinking of PBC.
* always working in the office
Nope. 3 days/week, only in the US. In Europe, they're trying to get it together but nobody's listening and the law doesn't support them.
* moving office location every 6 months
Nah.
* lots of unpaid overtime
Nah. Only if you're stupid. And in the US.
* world leading bureaucracy & red tape
Ok this one's true.
* training your replacement in a low wage country
Also true.
* seeing Ginny's helicopter (no touching)
Also true, although even with her Rhino-thick skin she knew this was a bad idea the moment she did it.
Is this for IBM itself or just an attempt to ward off mainframe abandonment by their customer base? My guess is the latter.
While it has been decades since I wrote any JCL, I have always admired the purity of the batch processing world. The sequencing of steps. The need to know all the files a program would use and describe them. Generational datasets. I remember taking a systems analysis class where the instructor said you were never ready to start writing code until you could define the JCL.
Doesn't really apply in today's interactive, constantly updated world.
There is still an important role for batch processing in today’s business computing world, and mainframes are very good at doing largescale batch.
Plus a Z-Series does pack a lot of processing power (and I/O capability) into its datacentre footprint; something that might be important to cloud data centre operators.
you were never ready to start writing code until you could define the JCL. I never heard that, but I have read it in a book. It's a simple way of ensuring you've got your conceptual shit together before coding. And, it works no matter what your command language is or what it runs on.
Someone said that before Lou Gerstner became CEO, the executive team knew how to spend money - but they didn't know where it came from.
One of Lou's first actions was to ask each division to present on their portfolio and which products made money, and which didn't. The divisions had to scramble to find this info. I heard that the mainframe made the money, and the money was spent on seeding and supporting other projects.
But then I could be biased and wrong.
Disclaimer: worked in Mainframe IT from 1971 to 2019. Was a programmer, systems programmer, 3rd-party software field and phone support (and management), consultant, senior consultant, manager, director, Vice President. Never worked for IBM: worked for local government, that software provider (DB/DC), and a Ocean Transportation company. Wife has a similar career arc.
This is not about IBM lacking people (although they likely do). This is about IBM's cash cow, mainframe sales and support, being in danger because THEIR CUSTOMERS can't fill mainframe positions. IBM is trying to grow the customer mainframe skill set to stop the erosion of mainframe sales (in my opinion).
(Almost) nobody is expanding their mainframe footprint, those that are are doing it due to legacy app lock-in and workload expansion thereof.
Nobody, not IBM, not 3rd-party software providers, not companies, did feck-all to train the next gen of mainframe talent... ever. Oh yeah, CA made a desultory effort for IDMS a decade ago... don't know how successful it was. I do know the User Association was ragging on them to get serious about this.
This situation is akin to climate change, anybody paying attention could see it coming, everybody understood what needed to happen (train or get off the mainframe), but too many people who controlled the money thought it was too expansive so let's just ignore the problem and retire before the house of cards collapses.
Who DID see this coming, and trained staff? Offshore consulting firms, who will happily rent you staff (with varying degrees of competence... some stellar, some not so much).
Retired almost 5 years, and enjoying my popcorn as this plays out.
To me, it's simple really. I would always ask one single question in the many meetings and what have you about "getting off the mainframe" that really ramped up after Y2K.
The one question?
Can this _______ system do 24 hours worth of work in 24 hours?
Yes or No?
Why are there still freight trains, amigos?
Why don't airlines ditch the big expensive planes for Lear jets?
And on and on and on.
Was at a meeting sometime around 2018 where a visiting big shot kicked off the thing by saying:
"I wake up every night in a cold sweat. A real nightmare. And that nightmare is KOBALT!!"
Jeez. Retired since 11/2020 and loving it.
As of circa 2015 x86/Linux can handle the workloads. The hardware will probably cost the same. But you won't need to pay licence fees for
for software that comes bundled with any other OS e.g. IEBCOPY the worse copy utility ever written.
The major stumbling block is the business logic embedded in 49 years of COBOL coding.
But hey COBOL now runs quite well on Linux and there are some nifty CICS emulators out there.
Common misunderstanding. Big Iron is not about processing speed, many mainframes are not that well endowed in that department but what they can do is I/O, lots of simultaneous I/O, very fast I/O.
Most mainframe tasks involving moving data around and not do a lot of processing on the data.
Consider a Telco, they'll have millions, maybe billions of transactions (calls made) a day and they need to record the details of each one and make sure the right person gets billed for each. Not much processing but a huge data volume to move around.
I may be a young pup compared to others here, having been in the telco/billing industry since I was a graduate in 2007 (covered telco, energy, utilities billing in that time). I haven't seen anyone use mainframes for billing or handling of calls in that time.
There may well be some legacy mainframe systems knocking around in telco, but none that I've seen (mostly UK and EMEA based telcos).
It was unix based systems, which then migrated to VM and/or linux in my experience.
IBM is pinning its hopes on some fresh initiatives – the Mainframe Skills Council and the IBM Z Mainframe Skills Depot – to address a shortage of engineers a shortage of engineers who are willing to work very cheaply, and who have big iron expertise. FTFY.
Some companies are jumping the other way, converting from mainframes to other systems. A friend of mine who was an AS/400 BOFH for a company which processes a very high transaction volume was not fired, but was retrained by his company as a DBA and then given that role.
Nobody, not IBM, not 3rd-party software providers, not companies, did feck-all to train the next gen of mainframe talent
Micro Focus did have an academic outreach program for a number of years that saw several universities sign up. Exact numbers are hard to find, even within the company, but we did issue quite a number of "personal edition" licenses. It focused primarily on COBOL, though, not the mainframe-environment product features.
There are programs out there teaching mainframe environments; you can find materials from them online if you poke around. They're certainly not marketed well, though.
Wow. I wrote a column for NaSPA Technical Support magazine for over 11 years. Working Smarter.
I have the original drafts of every single column. MS-Word. Also have many other articles from TSO Times, Z/Journal and others. About 50 articles and about 124 columns.
I own these. 85% deal with TSO/ISPF/REXX with lots of examples, code, usage tips and lots more.
Oh yeah, my primary job was a DBA!
But I wrote lots (started in 1982) of ISPF code. In every mainframe language. COBOL, Assembler, REXX and yes, even CLIST.
Would like to host this content somewhere.
Take your skills to the next level
Login below to see all the content
Sign In with IBMid
Log in with Bureaucratic ID™ to see if you like whatever it is we're offering!
I think with that first page alone they've just lost at least half the people who could be interested.
Following that link further it looks like the typical forum sign in/new account form. Though if you use LinkedIn it can bypass some of the needed information inputting steps and there's a bit of automatic data slurping occurring as for me it automatically has Country and State/Provence filled. Then there's the are you a student checkbox which on no brings up a company text field.
For at least two decades the mainframe division was the only one thatmade real money. However this was not s good story to tell on Wall Street. So they employed olympic standard accounting gymnastics to make LUW, Watson, Cloud look profitable.
Moving of the mainframe is now a very profitable niche market. CA in particular have a very nice set of tools for moving CICS/DB2 workloads on to local x86/Linux or the major cloud platforms.
So hey guys come and train as a steam engine driver, get a genuine 1970s salary.
If only IBM didn't sack so many experienced mainframe engineers for getting too old and not thinking about passing on decades of experience, which should be worth millions in revenue if you take the blinkered accountants, short sighted short term profits and senior management out of the equation !
《So hey guys come and train as a steam engine driver, get a genuine 1970s salary.》
You would get a thousandfold more takers. The 70s salary would be icing as just about every trainee would happily pay for the privilege. Who as a kid didn't want to be a locomotive driver? Casey Jones tv series.
Some of the Multicians might be attracted to IBM big iron but I suspect they had a bit more class. ;)
Honestly, mainframe is IBM's only remaining USP. That it's been allowed to rot so much says so much in itself.
I worked on an S/390 arrangement roughly 20 years ago and can safely say it was one of the best thought out systems I've ever been anywhere near in terms of resilience and support chain.
Was that the mainframe itself, or the software environment that you had to operate around it?
Being stuck with Fortran 77 is not an uncommon (it's how I got into it in fact, what being science background and mandatory F77 courses that go with that). Though a genuinely modern mainframe you can more or less pick whatever you like within the constraints of what your sysadmins will allow. If they haven't virtualised it for dev, prod, environments and so forth etc. they are missing half the point in having such a machine.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
