Joined: 13 Jun 2012 Posts: 144 Location: United States
I was a systems programmer for about 30 years. But, when I started (1964) you pretty much had to do it all, because there weren't any application programmers. The mainframe was new and so was Assembler language. COBOL didn't become available until 1967 or 1968.
I recall installing the first Mainframe operating system (DOS), which was expanded to OS/MFT, then OS/MVT. There wasn't a CICS or database for years. There was no JES or SDSF, there was no input or output spooler. The first major software products were BOMP, DBOMP and ARS. So as each of these products got installed, we had to train developers in how to code for them. As other new products (CICS, IMS, DB2, VTAM, COBOL, PL/I, etc.)came along, I had to install them, learn them and teach them.
As a result, I got used to writing code and developing systems and I was good at it. So, I kind of drifted away from systems programming, into application and interface development. I can still read a core dump, though.
Joined: 30 Nov 2013 Posts: 750 Location: The Universe
John Poulakos wrote:
... I can still read a core dump, though.
"core dump" It's been 40 years or so since the last core memory mainframe computer has been retired! I guess "core dump" sounds better than "semi conductor dump!"
My shop had a 370/165, one of the last core memory machines. It had been powered down and wasn't used for some months, replace by /168 machines. Then it had the grossly overpriced DAT box installed and as I understood it had been sold off to somebody. Even with the DAT box the buyer (IMHO) was cheated as the memory - even if maxed out (which it wasn't, as far as I know when it was sold) - was inadequate to future requirements.