Thread regarding IBM layoffs

I remember around 1983 the death of mainframes was announced and customers began offering early retirement packages and incentives for their mainframe programmers to leave.

Then they realized they couldn't run their business - and had to offer nearly 2x (particularly contract) to engage and/or re-hire their.

IRS has about 40 million lines of assembler (from a friend who does conversion work)

Mainframes will be around for quite a while - but also just in an increasingly smaller corner.

I did some mainframe programming on os390 and later z. For a while I did mostly Java and was able to use unix system service which hid most of the mainframe. I needed to be careful and ebcdic when ftping files. There was some weirdness with text files within jar files that were in ascii and caused issues. I also did some C++ and plx programming. This was painful. DLLs needed to be moved into the unit test environment using JCL. Starting and stopping programs and restarting the environment required a lot of learning.
Your choice but staying away is a valid choice

https://www.ibm.com/it-infrastructure/z/education

"this is the year of Linux on the desktop" for two+ decades

IBM-Kyndryl will no longer support Linux on employee workstations.
Windows or Mac upgrades are mandatory.

AS400 is still around although the number of RPG developers has decreased greatly due to offshoring and retirement. Personally, I'd be happy to never look at a green screen again.

Why are you posting on a layoffs site such question? Why are you here for? This is spam posting to divert the focus from IBM's layoffs?! What is or was your job at IBM if at all or are you at an external PR agency working for IBM?....

Demise of mainframes has been announced for at least three decades (just like "this is the year of Linux on the desktop" for two+ decades). All these naysayers and detractors .. why? Truth is: uptime of the COBOL code running on mainframes is measured in decades (as opposed to days on any other system widespread in use), MTBF is many years, IBM gave and gives (AFAIK) unique since 1973 a GUARANTEE that when properly configured and used certain security errors / faults CANNOT happen (system integrity statement https://docslib.org/doc/5974768/ibm-z-os%C2%AE-system-integrity-statement ) and unknown to most IBM z/arch hardware itself is software error resilient (eg it can recuperate from software errors like null ptr derefences in non-COBOL languages).
And I am not an employee, stockholder or any COI, and under 50 y/o, so not a guy from the golden Blue Pension years w an axe to grind. Just a modern security guy who audited the systems a decade ago and was perplexed to see how bullet-proof they were (like DoD's Ada infra which was shamefully abandoned)

truth is mainframes are slowly dying. The only reason they are around is for security purposes and for a legacy code base. Eventually, people will get rid of mainframes all together, as they will offer no distinct advantages over other platforms. They teach special classes at universities for mainframe programmers but not alot of young people seem interested.

lmfao. mainframes are a dying breed, it may take anothe 2-3 decades but they're out. who wants to build their career on this...? fine if you're mid/late 50's...

most growth is mips for existing customers.

When I worked at companies with mainframes, the teams that ran them were usually very experienced with both the systems and software they ran.
They were not the kind of people who would look for answers on StackOverflow or YouTube.

They would also help train new employees, and IBM has good training programs as well.

Copy-and-paste programming from StackOverflow doesn't work well for the kinds of systems that run on mainframes. There are a lot of credit card and banking transactions that need to be written with by people with experience, copy/paste off a website would be risky…

Second, even if IBM is selling more than it used to, that is still a very small part of the computer scene. I am gonna do some estimates here and I will be off, go gently on me… Most businesses don't have more than one (or possibly 2, with one for DR mainframes). So, 80–90% of the Fortune 500 companies are really just 1000 mainframes. If each of the Fortune 500 companies had a pair in each region (roughly the EU, the Americas, and Asia for big companies), that would be six in each region, or three thousand globally.

Some are also in universities, gov labs, and other places. But when compared to standard x86 machines, the numbers are tiny. For example, at one bank, "the mainframe" was actually just one hot server and one DR server. There were also 40,000 Windows servers and 40,000 Linux servers. In the same way, there were thousands of engineers in technology, but only a few on the mainframe team.

Given how much information is out there, even if there were "a lot" of information for some value of "a lot," it would be dwarfed by how much standard tech information there is.