10 May
2000
10 May
'00
11:20 a.m.
At 01:08 PM 5/10/00 -0700, Fred Cisin (XenoSoft) wrote:
This is THE device that will transfer files to
ANYTHING. IF you can stop
laughing.
Sounds like fun. How could it come close to hitting any key on any
keyboard? Are the solenoids positionable?
Certainly in some situations, a parallel-to-serial conversion box
could capture the bitstream from an archaic computer. That route
would have the benefit of a buffer in most cases. Does this gizmo
have a buffer?
As such, it would be most useful for getting data *into* a system
that didn't have a way to import an ordinary text file. You could
even write filters to massage the stream in order to hit weird
key sequences to reformat the text.
- John
10 May
10 May
12:54 p.m.
New subject: OT: Re: Keyboard actuator
John Foust wrote:
At 01:08 PM 5/10/00 -0700, Fred Cisin (XenoSoft) wrote:
This hits a little too close to home.
Back in '90 I was involved with a project which had to interface
to the Tokyo Stock Exchange. As it turned out there were multiple
means of introducing trades into the TSE, all a function of the
instrument in question. These consisted of:
- Something not completely unlike bisync on a 56K line that could
only be used for a handful of instruments.
- A system which required writing 5'4" 360K floppies using a
proprietary disk format which were then hand carried from
the writing station to a separate reader where the trades were
snarfed off and uploaded to the TSE (again, for a small universe
of instruments)
- An async dialup system which one was absolutly prohibited from
connecting to anything other than an approved terminal.
Naturally this was the system that was used to both input and
capture the vast majority of trades. To make things worse, the
TSE employed a small army of auditors who would make surprise
inspections to assure that you hadn't contaminated their pristine
dial-up lines with your own equipment.
As a consequence of this, most trading floors included a bizarre
piece of hardware which -- you guessed it -- hammered the
keyboard with solenoid actuated fingers along with (then
quite pricy) OCR hardware to make sense of the stuff being puked
back by the exchange.
When I first saw all of this stuff I thought it was one of the
funniest things I'd ever seen. It was when I realized that it
was up to my team to make it work that depression set in...
--
Chris Kennedy
chris(a)mainecoon.com
http://www.mainecoon.com
PGP fingerprint: 4E99 10B6 7253 B048 6685 6CBC 55E1 20A3 108D AB97
This is THE device that will transfer files to
ANYTHING. IF you can stop
laughing.
Sounds like fun. How could it come close to hitting any key on any
keyboard? Are the solenoids positionable? 
2:19 p.m.
New subject: OT: Re: Keyboard actuator
On Wed, May 10, 2000 at 02:54:48PM -0700, Chris Kennedy wrote:
- A system which required writing 5'4" 360K
floppies using a
5 foot 4 inch floppies? :-) You'd think they'd hold a lot more than
360K. LOL
As a consequence of this, most trading floors included
a bizarre
piece of hardware which -- you guessed it -- hammered the
keyboard with solenoid actuated fingers along with (then
quite pricy) OCR hardware to make sense of the stuff being puked
back by the exchange.
Gawd. Sounds like something out of _Brazil_.
I used to wonder when I was a kid if I could turn my Selectric into
a printer... would there be any shortcuts to controlling that ball, or
would it be necessary to use solenoids for every key. This was before
I knew that Selectrics had in fact been used as tty's back before my
time.
--
_______ Shawn T. Rutledge / KB7PWD ecloud(a)bigfoot.com
(_ | |_) http://www.bigfoot.com/~ecloud kb7pwd(a)kb7pwd.ampr.org
__) | | \________________________________________________________________
Get money for spare CPU cycles at http://www.ProcessTree.com/?sponsor=5903
3:01 p.m.
I knew someone who had one of these gadgets back in the '70's. He wore out
a selectric in about two weeks.
Dick
----- Original Message -----
From: John Foust <jfoust(a)threedee.com>
To: <classiccmp(a)classiccmp.org>
Sent: Wednesday, May 10, 2000 2:20 PM
Subject: Re: Keyboard actuator
At 01:08 PM 5/10/00 -0700, Fred Cisin (XenoSoft)
wrote:
This is THE device that will transfer files to
ANYTHING. IF you can stop
laughing.
Sounds like fun. How could it come close to hitting any key on any
keyboard? Are the solenoids positionable?
Certainly in some situations, a parallel-to-serial conversion box
could capture the bitstream from an archaic computer. That route
would have the benefit of a buffer in most cases. Does this gizmo
have a buffer?
As such, it would be most useful for getting data *into* a system
that didn't have a way to import an ordinary text file. You could
even write filters to massage the stream in order to hit weird
key sequences to reformat the text.
- John
3:56 p.m.
On Wed, 10 May 2000, Richard Erlacher wrote:
I knew someone who had one of these gadgets back in
the '70's. He wore out
a selectric in about two weeks.
I know a HUMAN typist who wore one out in ONE.
(she could AVERAGE 150 words per minute over 8 hour days! Not the
world's fastest, but close. At the end of the day, she barely had any
remembrance of what she had typed.)
It is vitally important that you not exceed 14.8 characters per second on
Selectrics.
The cheap home Selectrics are way too flimsy and short-lived for ANYTHING.
Always get the "heavy duty" models.
--
Grumpy Ol' Fred cisin(a)xenosoft.com
7:45 a.m.
Depending on the number of bits per character, start bits, stop bits,
etc., 14.8 or 15 characters per second is APPROXIMATELY 150 bits per
second.
Depending on how you count words, spaces, etc., 14.8 or 15 characters per
second is APPROXIMATELY 150 words per minute.
30 years ago, when I worked at Goddard Space Flight Center, there were
some holes in the wall, and discussion of the record "driving" distance
for selectric type balls flying when they came off. But I suspect that
that was mostly due to sloppy locking down of the ball on insertion.
(Some people there kept putting NON-APL type balls in the terminals.)
BTW, I'm reminded that there were two more versions of the device. There
was a guy in Walnut Creek that made one specifically for selectrics that
mounted UNDER the keyboard in place of the stock base. And there was one
shown at the 5th? West Coast Computer Faire that used strings for moving
the carriage, and worked on MANUAL typewriters (it drew a big cheering
crowd).
Now, after all of this discussion of them, am I going to have to go to
e-bay just to find a buyer??
--
Grumpy Ol' Fred cisin(a)xenosoft.com
On Thu, 11 May 2000 allisonp(a)world.std.com wrote:
Fred Cisin
wrote...
Like most typewriters it's mechancical and if driven to fast it will jam
and do bad things.
allison It is vitally important that you not exceed 14.8
characters per second on
Selectrics.
... or what? Does the Balls-O-Meter wear out? :)
11:47 a.m.
Yes, that's the "rate" for the selectric. However, there were at least two
selectric mechanisms, and the 134.5 baud rate was for the ruggedized printer
for the 1130 system. Those would last over a year under pretty hard useage
(I was the user and watched the same mechanism run for about a year).
However, the normal desktop selectric typewriter would fall apart within a
very short time with a mechanical actuator. I tried this myself once, but
was dissuaded from continuing it when a friend's selectric broke under the
actuator he'd bought through an ad in BYTE.
The selectric mechanism is a VERY complicated one in terms of the linkage
between the typeball and the keys. The keyboardless 1131 printer was much
more rugged, owing to the fact that it didn't have that complicated
mechanism of bails and cams that was the "keyboard encoder" and set the
rotation of the typeball.
There was also a selectric model that had a data cable that went to/from a
computer system. I never had the opportunity to use one of those, but I
recall reading that it had the separate mechanism of the 1131 printer but an
electronic keyboard encoder, hence, relied on the computer to tell it what
to type. In that sense it was full duplex. When the cable was detached, it
would echo directly to the typeball.
Dick
----- Original Message -----
From: <allisonp(a)world.std.com>
To: <classiccmp(a)classiccmp.org>
Sent: Thursday, May 11, 2000 2:13 PM
Subject: Re: Keyboard actuator
> Depending on how you count words, spaces, etc.,
14.8 or 15 characters
per
second is
APPROXIMATELY 150 words per minute.
Well back then if someone said 134.5 for the baud rate you could bet the
printer was a selectric.
Allison
12:43 p.m.
On Thu, 11 May 2000, Richard Erlacher wrote:
Yes, that's the "rate" for the
selectric. However, there were at least two
selectric mechanisms, and the 134.5 baud rate was for the ruggedized printer
for the 1130 system. Those would last over a year under pretty hard useage
(I was the user and watched the same mechanism run for about a year).
However, the normal desktop selectric typewriter would
fall apart within a
^^^^^^
very short time with a mechanical actuator. I tried
this myself once, but
was dissuaded from continuing it when a friend's selectric broke under the
actuator he'd bought through an ad in BYTE.
A properly setup mechanical actuator is quite a bit LESS stressfull to the
keyboard than a human. You are probably talking about the POS "home"
model, that would fall apart within a very short time if anybody typed on
it. It was nice to have for home use, but not suitable for office usage.
Using the keyboard actuator and a selectric as a substitute for a normal
printer seems a pretty BAD idea.
--
Grumpy Ol' Fred cisin(a)xenosoft.com
<A HREF= "http://www.xenosoft.com/dogears" >DogEars</A>
12:59 p.m.
I have to jump in since someone brought up the Selectric.
Has any-one here ever dismantled a Selectric? Did once
and though I thought knew mechanisms at the time it turns
out I sure didn't. The central mechanism seemed like a cross
between Dr. Nim (On Topic) and DigiComp-1.
The key presses were relayed by an almost binary series of
pushrods to a cascaded teeter-totter matrix.
The damndest mechanism I've ever seen. Anybody else have
this reaction?
Seems like a pretty good topic so I'm chiming in with this...
John A.
P.S. Don't go breaking up working ones to find out though.
1:43 p.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1889 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20000511/5d3bbf98/at…
3:50 p.m.
ard(a)p850ug1.demon.co.uk (Tony Duell) wrote:
I have to jump in since someone brought up the Selectric.
Has any-one here ever dismantled a Selectric? Did once
Never done a Selectric. Worked on teleprinters, though, and some of those
have some _strange_ mechanisms. Olivetti had a binary->position decoder
that consists of a series of metal cylinders each with an eccentric hole
such that each one fitted over the next smaller. These were rotated into
one of 2 positions according to the state of a particular bit (one bit
for each cylinder, all cylinders could move separately), so the
overall displacement of a follower on the largest cylinder was determined
by the total binary value of all the bits.
Creed made a paper tape reader where the holes in the tape for a
particular characeter were sampled one at a time by a series of metal
'peckers' and the state transmitted serially by a single contact assembly
operated by a link from the peckers.
Creed also had an amazing carriage feed mechanism consisting of a pair of
pawls opeated by a camshaft (half a turn per character IIRC) that 'walk'
along a rack fixed to the carriage. That one is almost amusing to watch.
and though I thought knew mechanisms at the time
it turns
out I sure didn't. The central mechanism seemed like a cross
between Dr. Nim (On Topic) and DigiComp-1.
The key presses were relayed by an almost binary series of
pushrods to a cascaded teeter-totter matrix.
FWIW, I believe the service manuals for Selectrics were available from
IBM, and maybe they still are.
The damndest mechanism I've ever seen.
Anybody else have
this reaction?
Seems like a pretty good topic so I'm chiming in with this...
John A.
P.S. Don't go breaking up working ones to find out though.
Dismantling a machine is not equivalent to breaking it up :-). It can't
be that hard to get them back together again...
-tony
3:52 p.m.
ard(a)p850ug1.demon.co.uk (Tony Duell) wrote:
Dismantling a machine is not equivalent to breaking it up :-). It can't
be that hard to get them back together again...
-tony
Hi Tony
I don't know about that. I remember this alarm clock.
I just took out a few screws and .........
Dwight
12 May
12 May
9:03 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 1410 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/20000512/ab774bd7/at…
11 May
11 May
2:36 p.m.
"John Allain" <John.Allain(a)donnelley.infousa.com> wrote:
I have to jump in since someone brought up the Selectric.
Has any-one here ever dismantled a Selectric? Did once
and though I thought knew mechanisms at the time it turns
out I sure didn't. The central mechanism seemed like a cross
between Dr. Nim (On Topic) and DigiComp-1.
The key presses were relayed by an almost binary series of
pushrods to a cascaded teeter-totter matrix.
The damndest mechanism I've ever seen. Anybody else have
this reaction?
Seems like a pretty good topic so I'm chiming in with this...
Hi
I used to work on teletypes with moving drums and type boxes.
These are alway complicate machines with decoding and encoding
bars that slide this way and that. The only type that is
really simple is the more standard rotating drum that
drives a single print hammer for each key.
Think about making mechanical software that first
takes a single input port status and then finds the
codes to send to a couple of ports. One port describes
rotation and the other lift ( or tilt ). Now think
how you'd do this mechanically. Add the fact that
you need to do all this in sequence so that it was
completed by the time the ball hit the ribbon. It
is not a simple operation. I marvel at the fact that
anyone even thought it was possible to make it work.
Then, I've always been fascinated by a sewing machine!
Dwight
7 p.m.
New subject: Selectric II or III ribbons
Since we are talking Selectric, I have a box of new ribbons. At least one
carton of 6 or 8 ribbons is for the Selectric II or III, some are for Qume,
some I have no clue. If these sound like $5 of adventure plus shipping to a
list member, tell me soon, otherwise I plan to rack in the big bucks on
eBay.
3:20 p.m.
See embedded remarks, plz.
Dick
----- Original Message -----
From: Fred Cisin (XenoSoft) <cisin(a)xenosoft.com>
To: <classiccmp(a)classiccmp.org>
Sent: Thursday, May 11, 2000 3:43 PM
Subject: Re: Keyboard actuator
a
I was not aware that there was a
"falls apart by itself" model from IBM.
The ones I had, and which I'd split with a number of friends was extracted
On Thu, 11 May 2000, Richard Erlacher wrote:
> Yes, that's the "rate" for the selectric. However, there were at
least
two
> selectric mechanisms, and the 134.5 baud rate was
for the ruggedized
printer
> for the 1130 system. Those would last over a
year under pretty hard
useage
(I was the
user and watched the same mechanism run for about a year).
> However, the normal desktop selectric typewriter would fall apart within ^^^^^^
> very short time with a mechanical actuator. I tried this myself once,
but
> was dissuaded from continuing it when a
friend's selectric broke under
the
actuator
he'd bought through an ad in BYTE.
from a building about to be domolished but belonging to
the "Phone Company"
(before the breakup) and the three that I got were
still in the sealed box.
These were definitely not "home" typewriters.
There were also numerous Teletype terminals (glass TTY) with band printers
and several different modems, not to mention quite a number of other fancy
phone hardware items, e.g. 6-button repertory-dialing phones, etc. This was
during the '70's when such things were still considered VERY fancy.
A properly setup mechanical actuator is quite a bit LESS stressfull to the
keyboard than a human. You are probably talking about the POS "home"
model, that would fall apart within a very short time if anybody typed on
it. It was nice to have for home use, but not suitable for office usage.
Using the keyboard actuator and a selectric as a substitute for a normal
printer seems a pretty BAD idea.
--
Grumpy Ol' Fred cisin(a)xenosoft.com
<A HREF= "http://www.xenosoft.com/dogears" >DogEars</A>
3:41 p.m.
New subject: PDP8 found and available
This would certainly require some cash, but Ed at CPB mentioned he had
found a PDP 8 this week.
Any interested parties please call ED at 828-274-5963 or E-mail him at
EDCPB(a)email.com.
--
-----------------------------------------------------------
Jeffrey S. Worley
Complete Computer Services
30 Greenwood Rd.
Asheville, NC 28803
828-274-5781 0900-1800 weekdays
Who is General Failure and why is he reading my hard disk?
Technoid(a)Cheta.net
-----------------------------------------------------------
8:50 p.m.
Allison Parent wrote:
Yes, I understand that, but I was also wondering why the precise speed of
14.8 characters per second.
Besides, "Balls-O-Meter" is fun to work into a conversation. I'm not sure
if all Selectrics have them, though (I think the early ones do).
-- Derek
Fred Cisin
wrote...
Like most typewriters it's mechancical and if driven to fast it will jam
and do bad things. It is vitally important that you not exceed 14.8
characters per second on
Selectrics.
... or what? Does the Balls-O-Meter wear out? :)
12 May
12 May
7:48 a.m.
> > It
is vitally important that you not exceed 14.8 characters per second on
> > Selectrics.
Yes, I understand that, but I was also wondering why the
precise speed of
14.8 characters per second. Besides, "Balls-O-Meter" is fun to work into
a conversation. I'm not sure
if all Selectrics have them, though (I think the early ones do).
--
Grumpy Ol' Fred cisin(a)xenosoft.com
10 May
10 May
3:26 p.m.
On Wed, 10 May 2000, John Foust wrote:
Sounds like fun. How could it come close to hitting
any key on any
keyboard? Are the solenoids positionable?
Not quite ANY. It has a little bit of positional adjustment, but not
much, and has provision for different key heights (for "Selectric", etc.).
The good news is that key spacing is, for all practical purposes,
standardized. Different key arrangements can be handled by swapping
solenoids around, or in software. It only does the regular "alphanumeric"
keys, and does not operate function keys, etc., even if they ARE where god
intended.
Certainly in some situations, a parallel-to-serial
conversion box
could capture the bitstream from an archaic computer. That route
would have the benefit of a buffer in most cases. Does this gizmo
have a buffer?
Although it could be used for transferring FROM a machine whose only
external manifestation is a centronics port, for THAT I would agree that a
parallel to serial converter is simpler.
If there is any buffer, it's surely no more than a few characters.
Instead, it is using the paralel port handshaking signals. There are
"inline" parallel printer buffers that would work with it.
Serial port is fine IFF
1) there is one. not on the Merganthaler from hell.
2) there is appropriate software. a VERY common lack.
Cabling is an inconvenience, but not a major obstacle. usually.
At 01:08 PM 5/10/00 -0700, Fred Cisin (XenoSoft)
wrote:
That's right. For routine transfer between relatively ordinary machines
there are other ways. As you know, MY preferred method is reading and
writing alien soft-sectored disk formats. But there are some machines for
which NONE of the "reasonable" methods apply. (such as that Merganthaler
typesetter)
--
Fred Cisin cisin(a)xenosoft.com
XenoSoft http://www.xenosoft.com
2210 Sixth St. (510) 644-9366
Berkeley, CA 94710-2219
This is THE device that will transfer files to
ANYTHING. IF you can stop
laughing.
As such, it would be most useful for getting data *into* a system
that didn't have a way to import an ordinary text file. You could
even write filters to massage the stream in order to hit weird
key sequences to reformat the text.
8832
days inactive
8834
days old
22 comments
12 participants
participants (12)
-
allisonp＠world.std.com -
ard＠p850ug1.demon.co.uk -
chris＠mainecoon.com -
cisin＠xenosoft.com -
dpeschel＠eskimo.com -
elvey＠hal.com -
jfoust＠threedee.com -
John.Allain＠donnelley.infousa.com -
mikeford＠socal.rr.com -
richard＠idcomm.com -
rutledge＠cx47646-a.phnx1.az.home.com -
technoid＠cheta.net