History of the Harbison Lineprinter Posters
Douglas H. Quebbeman
Without further comment, here is what Dr. Sam Harbison had to say:
=======================================================================
Here's the history of my line printer posters as best I can remember: I
was an undergrad at Princeton 1970-1974, and worked both in the main
university computer center and for Prof. Robert Langridge's lab in
Biochemistry. The lab used x-ray diffraction experiments to derive
protein structure. The x-ray transparency film produced (maybe 5" x
7"?) was put on a film scanner (drum style) which recorded 8-bit
grayscale data to a 9-track magnetic tape. I can't remember how long
the scan took--long enough that you didn't want to watch it. The tape
was then carried to a computer in another building for analysis.
My work with Langridge was on his new Evans & Sutherland vector graphics
processor, but I got interested in the film scanner somehow. This was
probably 1971-2. I had seen computer-generated pictures made from many
small graphic symbols. I can't remember if I saw any done with
overstruck line printer characters. I decided to try to produce them
myself. I scanned a few photographs: the moon, an astronaut on the
moon, my girlfriend's dog, some Playboy playmates, etc, and took them to
the computer center. I printed page-size tests of various combinations
of overstruck line printer characters, laid them out across the floor,
and chose a set that appeared to offer a smooth gray scale. The
lines/inch setting on the printer had to be increased so the lines were
closer together. I then wrote a program (probably Fortran, certainly on
cards) which did rudimentary resizing, cropping, scaling, and filtering
of the input from the tape, and mapped the resulting values to the
overstruck character combinations. I would make several runs late at
night to tweak the parameters used to process each picture, until I
liked the results. I then produced a new dataset on 9-track tape for
each picture that could be sent to the printer directly.
The computer was an IBM 360/91 and the program was run as a batch job
during the "white paper" times when the line printer was loaded with
plain white paper for students writing theses in Roff. (Usually it was
loaded with the alternating green and white stripe paper.) You could
always tell when the poster jobs ran because the line printer made a
characteristic high-pitched noise as it overstruck the characters. The
operators, I recall, were annoyed because the overstriking occasionally
weakened the paper and jammed the printer.
In any case, the posters became popular, and after I graduated they got
to various places with few people remembering from whence they came. I
no longer have any copies, paper or electronic, of any of the posters.
I'd be glad to receive one or two if they still look good! I am happy
to give my permission to use them for any noncommercial purpose. No one
ever asked before!
Thanks for the chance to recall the good ol' days.
With best regards,
Sam
=======================================================================
--
-Douglas Hurst Quebbeman (DougQ at ixnayamspayIgLou.com) [Call me "Doug"]
Surgically excise the pig-latin from my e-mail address in order to reply
"The large print giveth, and the small print taketh away." -Tom Waits
Charles Richmond
>
> Followup to the thread on the 'Cyndi Wood' Cyber 173 poster thread:
> Without further comment, here is what Dr. Sam Harbison had to say:
>
>
How did you ever dig up Mr. Harbison's email address so you
could extract this valuable piece of computer folklore???
--
+-------------------------------------------------------------+
| Charles and Francis Richmond <rich...@plano.net> |
+-------------------------------------------------------------+
Douglas H. Quebbeman
> "Douglas H. Quebbeman" wrote:
> >
> > Followup to the thread on the 'Cyndi Wood' Cyber 173 poster thread:
> > Without further comment, here is what Dr. Sam Harbison had to say:
> >
> > [full story snipped...]
> >
> How did you ever dig up Mr. Harbison's email address so you
> could extract this valuable piece of computer folklore???
Oh, I fancy myself a half-way decent researcher... available
for hire, I might add...
But it was either in USENET or the web. Dang. Citations,
they really *are* just as important as the information itself.
-dq