[GreenKeys] Model 20 Teletype available

[Sheldon Daitch]

Dave,

thanks for the interesting additions.

David I. Emery wrote:
> On Sun, Jan 11, 2009 at 10:52:55AM +0300, Sheldon Daitch wrote:
>
>   
>> The UPI in NC was phasing out their M-15s in the mid-1970s and the station I
>> worked at in 1976-1978 or so had a UPI Extel.
>>     
>
> 	Seems to me some UPI 15s were maintained by RCA service... at
> least the one at our college station was...
>
>   
>> IIRC, WRAL-TV was one AP subscriber which had, in addition to their
>> radio wire, a M-20 for the newspaper wire. 
>>     
>
> 	That is my recollection... AP national "A" newspaper low speed
> wire from the start of the VFT mux era onward was TTS only and it took a
> model 20 (or software and a minicomputer) to copy it.
>
> 	Believe it or not, some of the TTS 6 level AP stuff actually
> found its way onto military HF VFT ISB "tone packs" in the 60s and early
> 70s...
>
>   
    I wonder if that was a news distribution feed for the regional 
editions of the Stars And Stripes
newspaper.  Years back, the compsition for the printing was done in the 
regional printing operations, but
some years back, that was all taken back to stateside and the page 
masters were sent electronically
to the several S&S print operations. 
>> In early 1979, the AP in North Carolina was still running the M-20s on 
>> the slow
>> speed newspaper circuit, although there were a handful of newspapers in 
>> the state which
>> did get the slow speed wire fed directly into their computer systems, I 
>> think, or some of
>> them had tape punches and they used an optical reader to input the wire 
>> service material
>> into their computer systems.
>>     
>
> 	I do remember the requirement for slow speed TTS capability in
> early mini systems used with typesetting setups.   This meant odd baud
> rates and 6 level UART capability... (odd baud rates mostly done with
> custom humongous LF timing crystals)...   I do understand that some
> newspapers somehow got a cheaper deal on this service than 1200 baud
> "high speed" and demanded that the SW support it until the end of the
> TTS signals - I asked a couple of times about why in the hell with the
> 1200 baud available anyone would need the other and was told this.
>
>
>   
>>  The optical sensors weren't necessarily an 
>> improvement,
>> since the tape had to have much cleaner punched holes than the 
>> mechanical tape
>> readers.
>>     
>
> 	Optical was universal in the minicomputer world back then, and 
> could read tape very fast... but of course the holes had to be very
> cleanly punched, which some of the older Teletype punches didn't do... I
> remember having to go through tapes and find the holes that weren't
> punched cleanly when some piece of software refused to pass the checksum
> tests on loading (disks were such a Godsend, I can think of almost
> nothing else that was quite such a dramatic improvement except  maybe
> CRT based WYSIWYG software source code editing as compared to TTY based
> software editing).
>
>  
>   
>> There were also a few papers which did have the 1200 baud high speed 
>> circuits. 
>>     
>
> 	I do understand for some silly reason they charged more for this
> even though the audio channels were more or less the same (and same cost)
> as for the VFTG...
>
>   
>> All the AP slow speed circuits used the Lenkurt 25A VFTG system, while 
>> the higher speed
>> circuits used something the AP called a Dataspeed box.  It could be 
>> programmed to allow
>> the subscriber to receive only the circuit the subscriber was paying for.
>>     
>
> 	The actual 1200 baud wire audio signal was standard Bell 202
> tones (FSK) similar to AX25 packet but of course plain old async
> ASCII...  And yes the messages on the wire had various selector codes
> preceding them (a standard was available defining all of this) and I
> suppose they had a microprocessor somewhere that filtered out the
> different services on the wire and only outputted to customer gear those
> the paper subscribed to.
>   
     The modem between the phone line feed and the feed to the 
newspaper's system was the Dataspeed
box.  There was an IC socket for, I guess an EEPROM, chip and that was 
programed for the services
subscribed to by the newspaper.  Again, going back on 30-year old 
memory, I think if the chip wasn't
installed, the box passed everything.  But that wasn't well known 
outside the office.

> 	The 25A VFTG was 120 Hz tone spacing/ 60 Hz shift  for the most
> part... could be up to 22 or 24 wires on one audio signal... with some
> provisions for using two 120 Hz slots or 3 or 4 of them for a higher
> speed (up to 300 baud) signal (common for some financial tone packs)...
>
> 	I do know some of the Extel printers had built in VFT modems as
> an option, I used to have one grabbed surplus somewhere with that
> feature.  I forget whether it was crystal channelized or synthesized...
> but I got rid of it probably 25 years ago by now, so I can't check.
>   
    The one station I worked for in NC was a UPI station, had an Extel 
with the internal
VFTG modem.  It made a compact package.  I remember one election year 
cycle, when I
was working for this station, and they wanted to have a printer at a 
remote broadcast operation
for election reporting, while doing the broadcast from the county 
elections center.  I inquired with
UPI for the price of a temporary circuit, and perhaps machine rental, 
and the station management
people balked on the price.  I realized the VFTG signal was simply an 
audio circuit, so I had the
telco install a broadcast loop from the station to the elections 
center.  I verified I could back feed the
loop with the UPI TTY audio, it worked, and we were on the air with the 
machine at the remote
location.
>> I don't know when the slow speed wire service via the telco was dropped 
>> for satellite
>> distribution, but I think it was a phased project, state-by-state.  In 
>> the early 1980s, I
>> remember the AP folks I worked with in Raleigh were installing the 
>> satellite dish for
>> wireservice at the Greenville Daily Reflector, and that was probably 
>> about the time
>> the AP dropped the telco distribution.
>>     
>
> 	I do know the UPI switched their radio station stuff to a SCPC
> carrier on a C band satellite for a while in the 70s to mid 80s - with
> the same ole 25A mux signal on a 12 and 16 KHz SSB channel above their
> UPI radio audio... and may have experimented in some places with putting
> this out on FM SCA carriers for distribution...
>
> 	And UPI eventually switched from that to an Equatorial
> Communications spread spectrum box for their wires... using those
> distinctive cute little C band dishes.
>
> 	The AP eventually installed a distribution system using a C
> band based SCPC QPSK digital signal at 256 Kbs... not sure of the
> multiplexing or whether they continued to support any form of low speed
> TTY circuits on it.   The couple of narrow band SCPC signals for the AP
> occupied a whole C band transponder and were rip roaring powerful as a
> result - and the AP  used little 5 foot dishes for this (which is very
> small for C band)....
>
> 	I understand the AP has more recently done a deal with either
> Dish or DirecTV to lease some capacity on their Ku DTH home TV birds...
> no idea what the rump surviving vestigial  remains of the UPI currently
> uses or what Reuters uses... but using an audio channel or two  in the
> MCPC MPEG muxes on the DTH systems makes sense as those signals are
> REALLY strong (thus the little home dishes).   I haven't seen any of the
> AP C band dishes around here in about 5-10 years...
>
>   
    Thanks for those details.
>> When I left the AP in early 1979, the Durham Herald-Sun was still 
>> running hot lead Linotype
>> machines which could read the punched tape from the slow speed wire.
>>     
>
> 	This was a big union issue in some places.  Computers kinda put
> the whole compositor trade out of business... an editor could easily set
> the story in type from his terminal... so newspapers with strong unions
> were stuck with hot metal long after it made no sense whatsoever...
>
>
>   
>> As Dave mentioned, the TTS coding for the Linotype machines had the 
>> coding for the
>> small and large spaces as well as for the variable spacing, the wedge 
>> spaces for hot lead
>> typesetting justification.
>>     
>
> 	A very early computer application was doing all this justification
> for those wires... not sure if it was done this way from the beginning of
> TTS, but way before most newspapers had computerized composition systems
> the wire services had computers to create the TTS copy... especially for
> tabular things like sports scores and stock tables.
>
>