THE REXX STANDARD IN CONTEXT
 
When Rony suggested I might present here, I wrote: I don't think there is enough material of interest to justify calling it a proper presentation but if a few minutes of my standing up and talking about the development of the Standard will suffice then please schedule it.
 
That morphed somehow into this agenda item. Not worth your hour.  But extra time allows us to look more broadly – Rexx, RexxLA, Standard, and IBM history intertwine.  Maybe some remember it differently.  
 
Celebrate Rexx.  Could be more to celebrate?  (Competitors. Hand-helds)
 
The Product Lottery  and how does anybody wind up writing Standards?
Couple of examples from my own history:
 
1963   BLM to Hursley
SRPG Non-programmers designing reports. 4000 punched cards written in US. Hardware 1440 variable field based. 12 pennies to shilling.  Extended, made maintanable.  Technical achievement irrelevant – UK goes metric.
PL/I Rexx influence.  PL/I Design by informal committee.   (including users) 
1966 PL/I out of Hursley
One of the second wave PL/I projects, Checker, BLM manages.
Entirely favourable context.  Good people wanted to work here.  20% recruitment per year.  IBM business model - Monopoly, software leased.
Used up-to-date tools, most reliable software from Hursley to that date.  Did what was intended to well.  Good for staff & for me - youngest Senior Programmer.
1971  PL/I Checker out  
 
Wanted to tell of those two because they illustrate how we techies can be at the mercy of wider issues. I am going to suggest Rexx and the Standard buffeted in the same way.
 
Wider issue – IBM management in a rut.  Couple of decades.
Not discourteous to our host to say that now.
Likely to happen to any very successful company – management tuned to what used to be needed.
Hursley – developing hardware that never got out.
Technical v Management -  Santa Theresa, Hursley new PL/I compiler.
BLM outspoken on management misdirection.
 
1979 Mike starts on Rex. (MFC also some manager problems.)
 
MFC Oxford English Dictionary machine readable, also Jargon Dictionary (humour like Dilbert)  Definition of “Penalty box.”   
BLM Staff to software manager, on Standards (internal to start with).
 
Rexx Peak Eagerness at this time
1987 Rexx SAA Procedures Language. Idea is IBM/360 for software.
1989 9 Rexx sessions at SHARE. (41 by 1992)
1990  SLAC’s Cathie Dager – First Symposium. (See website list)
Not 100% support.  Pretty forceful case that MS Basic would prevail.  Warp later confirms?
1990 November – X3 announces new technical committee X3J18 to develop American National Standard for Rexx.
(Never could make worldwide case)
Committee members have to commit resource.  ANSI training required for some.
1991 Asilomar, California.  
Linda Green, SAA Procedures Language Interface Owner.
BLM “retires”, becomes subcontractor. 
1992 Annapolis, Maryland.
1993 La Jolla, California.
1994 Boston
Easy job as chairman of X3J18.
“Better wrong than different” versus “opportunity to improve”.
Rexx defined by Rexx.
Not much turnover of committee membership.
Complexity, e.g.  Latching conditions....  Literal Labels.
Austrian chocolates.
Format of examples a struggle.
 
Rexx Peak Plans at this time
1995 Stanford, California.
1996 Austin Texas.  IBM backs down from commitments given to SHARE.
Undercutting of the rationale for the Standard.
Standard & Copyright
Debugging Standard piecemeal.
Standard out.
Object Rexx out.
1997 Heidelberg  
1998 Research Triangle Park
1998 Dallas draft – Programming Language Extended Rexx, the attempted 
Oryx standard, did not get far.
1999 Jacksonville – first J18 meeting in maintenance mode.
Validate Standard as a whole.  (Tiny Rexx, 31KB)
2000 Phoenix Arizona 
2002 Reseach Triangle Park, North Carolina
And more.
2006 10 years of Rexx standard – refresh justification. (Sold how many at $125?)
2009 Celebrate 
 
The Standard gave Rexx status, and resolved some contentions.  But by-and-large falls in the large bucket of technically good projects that lost their impetus through events outside of the project’s control.