Now here’s some data to inspect using IEHIBALL: as a new generation of mainframers arrives around the world, there are many subtle aspects of our culture that they are unlikely to be directly educated and mentored in, but until they have internalized these obscure facts, they won’t be true mainframe initiates.
It’s not like this data is undocumented — great mainframers such as REXX author Mike Cowlishaw have put together lists of our vernacular over the years (for example, see “IBM Jargon and General Computing Dictionary”), but until you’ve heard these terms used in real sentences, you may be tempted to mispronounce or misuse them based on your own assumptions, possibly learned on other platforms.
A great example of this is “SQL,” which distributed people pronounce “sequel,” but established mainframers, especially those who don’t maintain other platforms, are prone to spell, i.e., “ess kew ell,” instead.
Over the years, I’ve put together a few pieces documenting various aspects of our mainframe linguistic culture, such as the SHARE’d Intelligence article “You say ‘CICS,’ I say ‘CICS’” about how we pronounce CICS, and this video about our shibboleths.
Of course, language is constantly changing, and varies across the planet as well, so no comprehensive list of such words can be found in one place. But it’s still fun to take a moment to consider some of our favorite words that make us comfortable and signal that we’re dealing with a fellow mainframer.
One, or rather six, of the words that are most deeply embedded in our culture have a history predating electronic computing. During World War II, when many of the people who would go on to contribute to the development of computing and the mainframe learned their stuff, the standard phonetic words for the letters of the alphabet were mostly different than they are today. Most specifically, once hexadecimal came into common use for representing the binary values of the eight-bit bytes on the mainframe, the phonetic words for the letters A through F became important parts of the culture. As it turns out, the modern standard for these letters has changed for all but “C,” which is still Charlie. But today, we say Alfa, Bravo, Charlie, Delta, Echo, Foxtrot, so a new mainframer listening to their most experienced colleagues may feel the same consternation or confusion that I did at the beginning of my career to hear people using the “wrong words” — especially when it becomes clear that all the people who use the wrong words use the same ones: Able, Baker, Charlie, Dog, Easy, Fox! If you’re A to memorize them, fitting in with other mainframers becomes 2E.
Another area where we have insider words is abbreviations, which seem to require being spelled to be pronounceable. While “CICS” is a special case of this, we also have VTAM, VSAM, DASD, EBCDIC, and, for some people, SNA. But, we also have compound abbreviations where the way we break up the word to pronounce it is the secret, such as ABEND (which may be “a bend” or “eh bend” or “ab end”), PROCLIB, and CLIST.
Then there are the “legacy” terms that we still use to refer to the current version of something — kind of like giving someone directions that include where a business used to be. A great example of this is MVS, which could be said to still be an aspect of z/OS, but more appropriately is its predecessor. Similarly, TSO/E, SMP/E, and ISPF/PDF are often just referred to using the portion that came before the slash was added.
That reminds me of another key differentiator: function keys. When referring to 3270 (which we pronounce like a pair of two-digit numbers), many of us still refer to program function keys, or PFkeys, not function keys, or Fkeys like consumer devices have.
And, for that matter, we have a number of terms for something we do on the mainframe that is similar to but not the same as distributed platforms. So we don’t “boot” z/OS: we “IPL” or Initial Program Load it, or, on a bad day, we “POR” or Power on Reset, both of which are pronounced by spelling them. Likewise, we don’t have subdirectories. Rather, we have catalogs and partitioned datasets, or PDSes, pronounced by spelling the uppercase portion and then using normal English rules for the rest.
Often, the reason for our jargon is to refer to something unique to the mainframe. So, the Principle of Operations, a comprehensive manual describing the hardware architecture, has a couple of nicknames: the POO or the POP for those who feel sensitivity about the former — both pronounced as spelled.
Finally, there’s the fun practice of creating reverse-acronym meanings. So, the old mainframe term MIPS (pronounced as spelled), which originally stood for Millions of Instructions Per Second and has often inaccurately been treated as if the S was a pluralizing letter, has many deliberately false “meanings” such as Meaningless Indicator of Processor Speed. Likely the term that has the largest number of such meanings is “IBM” – but I’ll let you look them up yourself (after all, It’s Better Manually).