Saturday, September 7, 2019

Dave Boyd and the STSS

 

When I worked at IBM's Field Engineering headquarters Information Systems Department (FEIS) in the late 70s, I had the good fortune to collaborate on several projects with David P. Boyd, a truly gifted systems analyst with an impressive résumé of accomplishments, including writing the curriculum for the first Autocoder class.  (I don't know what 'Autocoder' is, by the way.)  Dave was the analyst and the brains behind the Suggestions Tracking and Statistical System (STSS), FE's first IMS DB/DC application.  For reasons that were not initially clear, I, a very junior programmer, was tasked with doing the core programming for this quite-elaborate application.

Management at FEHQ was very averse to slipping a date once the client had signed off on it.  That meant that on occasion programmers (and sometimes the analysts) would work through the night and sometimes through the weekend.

On one such overnight effort, Dave and I were returning from FE's Sterling Forest Data Center in the wee hours of the morning.  I don't recall the project or whether it was a one-day or a multi-day work-through, but I do recall that Dave had driven us there in his Porsche 914.  As soon as we were sure the project would make its deadline, we wrapped up and headed for the parking lot.  It was around 2 AM when we set off for Yorktown Heights where Dave and I both lived.  The only plausible route was Seven Lakes Parkway winding through Harriman State Park, one lane in each direction, heavily forested, and unlit save for the occasional traffic circle.  As we approached Tiorati Circle, Dave casually mentioned that Tiorati was the highest elevation on Seven Lakes Parkway.  "You know what that means, right?" he asked me.  I admitted that I did not know what he was asking.  "It means that we could coast from Tiorati all the way to the Bear Mountain Bridge.  Are you in a hurry to get home?"  I shook my head.  A minute later, Dave negotiated Tiorati Circle at 45mph and as he rejoined the main road, slipped the 914's stick into neutral.  For the next 20-something miles, gravity provided all the motive power to the car, and Dave only put it back into gear when we could see the bridge at the end of a long downslope.  It was the weirdest sensation: racing down a darkened country road at highway speed, with the road ahead illuminated only by the car's headlights, and the Porsche engine purring contentedly.

STSS had been mandated to FEIS by Corporate because the Suggestion Department in Endicott (NY) was nominally run by the Field Engineering Division even though it served all non-plant locations in the United States and thus served several different divisions' employees.  The nature of the project involved a great deal of data entry, so much that it stretched over several years.  Since it was to be a 'flagship application' and involved technologies FEIS was largely unfamiliar with, the project time line was allowed to extend far longer than would otherwise be permitted.  While I did most of the design and coding, several other programmers would now and then be pressed into service to produce parts of the system as well.

When it was complete, the software was cut over to run in parallel test mode.  That is: Endicott considered it to be "running in production" even if everyone in 'production' considered it to be not-yet-installed.  Getting it documented and installed was my job alone since everyone including Dave had been shifted onto other projects leaving me as the last man standing.  After several abortive attempts to get 'the package' accepted by the production side of the house, it became clear that some sort of animus was at work to prevent STSS from becoming 'official', and the animus extended to my own management who steadfastly refused to intervene to get an otherwise perfectly-working software system past a gamut of ever-changing rules.  In August, 1979, with STSS still officially uninstalled, I left IBM for greener pastures.  Within six months, Endicott's management began complaining to Corporate management about the situation, and Corporate told FEHQ to 'get that installed or find a new job'.  Presto!  Installed!

Each of the plant locations also maintained their own Suggestions Departments and each therefore had the same problems associated with tracking and counting submitted suggestions.  Before too long, plants here and there started whining that they didn't have an STSS of their own.  Dave Boyd — as the most knowledgeable person regarding STSS — was pulled back from whatever he was then doing and assigned the task of visiting each plant and installing a copy of STSS for each of them.  Well, plants aren't only in the United States.  There are plants in Milan Italy, Corbeille-Essonnes France, Boeblingen Germany, and several other places.  Dave got The Grand Tour of Europe.

Years later, I would learn that FE was resentful at Corporate HQ 'ramming STSS down their throats' and assigned the newest, greenest programmer they could find to screw up that project to a fare-thee-well, but they never told me about their nefarious plans.  Thinking they meant the project to be a success, I did everything I could to get it to work.  They were very, very upset that I succeeded, and I think they were more surprised than I that it turned out as well as it did.

IBM, I hear, no longer has a Suggestions Program, so FEHQ got their wish after all.

 

No comments:

Post a Comment