The sixtieth anniversary of my first computer program: things have and have not changed.

2022-08-21 01:21:28 By : Ms. Fenny Deng

An entirely improbable confluence of events in my sophomore year in high school in Denver, Colorado, in 1962, sent me off in unpredictable directions.

One of the members of the Math Club at George Washington HS was the son of a sales rep for Control Data Corporation.  The sales office had a CDC 160A demo unit, and if it was not in use for sales (and was actually not down for trouble diagnosis, or raided for spares by the field service folk...) Math Club members could use it for instructional purposes.  This, admirably, was organized as a course in programming using the F0RTRAN language.  (Whoever designed this presentation of the language decided that a circle with a stroke through it would be the letter ‘O’, when for decades it had signified the digit ‘0’  for telegraphy and radio communications.  Consequently, a slashed circle as upper-case ‘o’ became the sign that you really were hip to exotic technological information.)

The CDC 160A was Seymour Cray’s first commercial computer, and was not as massively capable as most of the quarter-acre systems, but being transistor-based it was capable of being miniaturized (the base unit was like a very large office desk).  In overall processing power terms, it was basically the equal of the computer that runs your wall thermostat, or your coffee pot.   In present-day currency, delivered and supported models began at $200K, according to my ability to recover price lists.

We would craft our programs, writing the instructions on coding sheets (a practice that persisted for perhaps twenty more years, in some shops), then use a Friden Flexowriter to punch a paper tape with the program.  When your turn came.  Then, when your turn came, reset the machine, load the Compiler tape, then your program source code tape.  Normally, the machine would then spit out a paper tape containing terse and cryptic diagnostic criticisms of your precociously inept work product, which you must now correct and re-punch.   (Our first lesson in LOOP,  ENDLESS:  SEE ENDLESS LOOP).  Otherwise, it would spit out the Object tape.

Eventually after fussing with the Linker tape and the Library tape and the Object tape, an executable tape would issue forth, which then (when your turn came) could be loaded and run in the machine, usually with results that elicited vocalizations like “Well, bless my sox, whatever does this unexpected result betoken?", and a return to the coding sheets.

In the ensuing 59 years, there were many adventures.  I inadvertently invented the Third Normal Form in 1976, not having the benefit of a formal education in what little was known of computer science in that era.  I restructured the operating system that controlled the engine and powertrain in Ford passenger cars (EEC-IV, 1987 and beyond).  I was the architect of perpetual inventory and stock management systems for a couple of large telecom enterprises.  This stuff has always been in my blood, so to speak.  In fact, after I discovered at age 68 that I am autistic, it became obvious that these things were a Special Interest, and not every other human will have these interests and/or skills.

Today I read the article www.dailykos.com/…

CDC gave permission for the Math Club to allow us to invade, and benefit from, their commercial enterprise.  I don’t doubt that there was some grasp of the fact that, when these machines became ever more affordable, a vast army of nerds with coding sheets would be indispensable. 

From the standpoint of what benefits for-profit corporations, this makes eminent sense in that, like construction workers, you need a lot of ‘em to get things built.   Withal we have for-profit coding academies, as well as some that appear to have objectives other than accumulation of wealth.

From my experiences while jobless and in bankruptcy due to unemployment, I find that most businesses are entirely clueless when it comes to their actual needs, and how to find individual humans who can meet them.  After all, remember that the HR department’s primary function is to identify the persons who should not be hired.   I only got my job at Ford because I was a contractor (therefore a product, handled by Purchasing and not a human, handled by Human Resources).  Even though I didn’t have an engineering degree (thus not eligible for hire through HR), I was at one point assigned to tutor a recent new-hire, who did have a degree (U Mich), in how to do software engineering.

It seems like, daily, I encounter examples of elementary software blunders in software that controls our lives.  These things cannot happen (at least not as oppressively often) if “coders" have correct and unambiguous instructions from designers, whose designs derive from the overall architecture propounded by competent architects.

Our problem, in part, stems from the unfortunate hierarchical distinction between coders, designers, and architects.  The inbuilt assumption is that the coders are of the least value, and the architects are of the most.  This is harmful.  Coders, carpenters, pipefitters, bricklayers — if they don’t love what they do, things will tend to fall apart.  If their value is not perceived and taken into account, there will be issues, short- and long-term.

My late Dad was a pipefitter.  He loved it, was very good at it, and taught me as much as my neurologically deficient self could absorb.   I wound up as a designer and an architect; I’m really a poor coder, but I have enormous respect for those who do it well.

An effective system for finding, cultivating, and deploying people in their promised land must begin by knowing what you are looking for, and what to do with it when you find it. 

The for-profit model only works accidentally.