[TUHS] Spacewar at Bell Labs [ really paper tape readers and tangentially related things ]

[Jon Steinhart]

Earl Baugh writes:
> Why not build a variation of this with an Arduino?
>  https://www.instructables.com/id/DIY-Paper-TapePunch-Card-Maker-and-Reader/.
> You could use cardboard rather than wood if it’s just a one time job. ( or scan
> the tape into files and process digitally ?) 
> 
> Earl 

I thought that I said earlier that I had a paper tape reader here but don't
remember much about it or if it ever worked.  If I didn't have a huge project
list and it wasn't ski season I could hook it up to a pi.  More likely that
I'll get to a computer museum sooner.

Just to keep this UNIX-related so that Warren doesn't get cranky, I believe
that this reader came out of some sort of experimental telephone exchange
in our group that was decommissioned.  Dave Weller was very supportive of my
interests and somehow arranged for me to take much of it home as surplus
equipment.  Kept me in 7400-series parts and Augat wire-wrap boards for a
long time.  While there was no way that I could have kept the thing, I wish
that I had the magnetic drum memory because it was so cool from an industrial
art perspective.

Heinz may remember more about this than I do because he actually worked on
this project, but our department developed what I believe was the first
all-digital telephone exchange that used digital filtering (Hal Alles and
Jim Kaiser were in the group).  I think that it had a pair of PDP-11/10s
in it, and a bigger PDP-11 as a supervisory machine that ran UNIX.  I have
vague memories of Heinz and Carl poring over huge C program listings.  I
also remember that there was a bug in the long-distance code where it wasn't
sending out the ST tone that ended up taking all of the key pulse senders in
the Berkeley Heights telephone exchange that provided the trunk line to our
lab off line as they didn't have timeouts and needed to be manually reset.
But hey, we were the phone company too so what could they do about it?

Oh, I think that the PDP-11/10s were used because we tried to use LSI-11s
but those turned out to be useless because of the way that DEC did the DRAM
refresh; it wasn't interleaved, they just stopped everything every so many
ms and refreshed everything.  Non-starter for real-time systems.

And another thought, this machine may have been why Heinz wrote MERT.

I was gone before this system was completed so I have no idea how it fared
and how many of the ideas were incorporated into production systems.  Oh,
yeah, I think that it was called the SS1 for Slave Switch 1.

Jon