On Wed, 8 May 2019 15:39:49 +0000, Seymour J Metz wrote:
>I don't know where to begin, but it reads like buzzword bingo. How many of the
>errors can you spot?
>
(link de-cisco-ized):
https://www.computerworld.com/article/3393220/take-cover-it-s-a-student-programmer.html
... Those header cards are punched for four-way symmetry so they can run in
any direction,
whether backwards or upside down. But they have so many punches that
they’re almost
lacy in appearance. ...
Ummm ... 2-way symmetry (1<=>80) is easy; punch just 40 columns starting at
each end.
(12<=>9) harder in EBCDIC -- need to decode binary images (in an exit entered
for each
card). And if decoding binary, why punch the card symetrically -- let software
do it.
I once used a site where the job separator cards were brown stock with no
corner cut;
easily spotted by the operator. No symmetry; needed to be correctly oriented
with the
vendor's EOF code in column 1.
Visual inspection by the operator may have had an adverse consequence. I once
got
punched output with *one* card punched in wrong orientation. Best explanation:
That card was backward in the hopper and so punched. Operator separating jobs
noticed the wrong corner cut and "fixed" it. I fixed it further by duplicating
it in
correct orientation on a keypunch.
I once made an 81-column card by making minuscule adjustments at the punch
station. Reader read 80 columns (FSVO) correctly and ignored the 81st with no
error indication.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
