[Tutor] Sidetrack history: Re: cases with single if and an else clause

[Dennis Lee Bieber]

On Wed, 6 Oct 2021 08:08:56 +1300, dn via Tutor <tutor at python.org>
declaimed the following:


>
>Those of us who learned "Structured Programming" when it was 'new'
>probably still think of it as a 'thing'. (I'm not sure how often such is
>mentioned in 'today's training', and haven't seen anyone under thirty
>using a Nassi-Schneiderman chart for ages) We also take quite-naturally
>to the SRP (Single Responsibility Principle)!
>
	I don't think I've touched one since around 1985. At the time, the
company tended to load up on new-hires, and /then/ shop them around to
department managers to determine final placement^1.  In the meantime we
were all in an "icebox" and provided general (ie: unclassified) work
assignments. ^2

	Company was trying to develop a suite of programs for software
engineering (N-square charts^3, the "Chapin" variant of N-S charts,
requirements traceability^4, etc.). An effort with continuously changing
environments -- it started with non-graphical systems, then moved to
high-end graphical documentation workstations, then started adding OOAD
features, and maybe ended up with COTS software.


*1	I was shanghaied -- had two department interviews, told I would have a
few more before having to choose, was one of a small group sent to a 4-day
course in Ada [mil-std 1815 had only been released a few weeks before and
the Ada/Ed translator even more recently], returned and was told management
decided I'd be in the second interview (the first one had been summarized
by the manager as "after a long study, the Navy had concluded high-speed
aircraft posed a threat" -- I think that was supposed to turn into some
scheme with satellites and/or loitering AWACS descendents relaying
targetting information to ships).

*2	Even after being assigned to a department, one still had to await
proper clearances. My DoD Secret came through in something like two months
(at the time, the average time for a Secret ran 6+ months -- I used to joke
that my paperwork must have gotten stuck into the stack for Reagan's new
staff). My black program access, though, did take over a year.

*3	On an early 48K CP/M machine using M$ FORTRAN (F80). The machine was
limited to 48K as it had some weird ROM and a custom CP/M (needed as the
normal interrupt vector used for OS services was buried in the ROM and
couldn't be modified)

*4	With some effort, I always thought the "Automated Requirements
Traceability System" could have been developed into an RDBM using
relational algebra. It already had operations similar to those of SELECT
and PROJECT -- the weakness was that one had to predefine the "table"
layouts, and specify the format (by numeric ID) on most operations; rather
than incorporating some means of having PROJECT extract the field
specifications and dynamically generating a new format definition. Porting
that application to a CDC MP-60 [in a militarized "drop container" -- the
paint was sturdy enough to chip concrete] was "fun". Minimal FORTRAN-IV
meant all the subroutine calls that incorporated operations
			call xyz(n-1, n+1)
had to be changed to 
			jinx = n-1
			minx = n+1
			call xyq(jinx, minx)
AND, since it lacked direct access I/O via F-IV statements, we had to use
an OS system service call to seek file sectors. Only to discover that while
a sector was 512 bytes, some sort of overhead only allowed F-IV to
read/write 480 bytes.


-- 
	Wulfraed                 Dennis Lee Bieber         AF6VN
	wlfraed at ix.netcom.com    http://wlfraed.microdiversity.freeddns.org/