Last updated Feb 10 2013. (C)Copyright 2013 Herb Johnson.
In Feb 2013, Charles Murphy sent me these recollections of his work with Friden in the 1960's. They are here on my site with his permission. Links to more conversations about Flexowriters and to some Friden documents, is on my Friden Web page. - Herb Johnson
I have read your correspondence with Bob Behr and found it very interesting. As he was an employee of Commercial Controls and a sales representative his recollections may be more accurate than mine but they differ.
When I went to work for Friden, sometime mid year of 1959 and shortly after being hired, I went to Rochester N.Y. for training, it was my understanding that Friden already owned Commercial Controls.
At any rate the training center was, I believe, on Prince St. and the Commercial Controls Factory was close enough such that we could go to the factory for lunch if we cared to. The cafeteria lunch line was alongside part of the Flexowriter assembly line, separated by a windowed wall, and we could observe some of the assembly process.
On one such occasion, as I remember, and would like to know if Bob Behr could confirm it, we learned that one of the assemblers of the Translator was blind. I found that to be amazing as I found the Translator to more be one of the more difficult assemblies to get right during training because it's input was from the reader and it's associated contact problems. And, my experience confirmed the difficulty of trouble shooting problems, when the Translator was the real source of a difficult problem.
By the time I was hired the initial training in Rochester N.Y. was on Flexowriter model SPS, which denoted as System Programmatic Single Case. There was also Model SPD which denoted as System Programmatic Double Case. That was identical in every way except for the addition of keyboard shift keys for Upper and Lower case character type printing but were much less popular than the SPS.
Understandably, the initial assignments upon returning from training were directed to the model FL which were simply letter writing applications used by companies as diverse as local stock brokerages to large international companies such as United Fruit Co., to send personalized letters to individual clients. There were also companies that had many model FL machines that wrote letters from client lists for large companies.
To my surprise there were Flexowriter models that I had no idea existed such as model FG80 which was used by the largest department store in Boston, the Jordon Marsh Co. It created a paper tape which was then used as the input medium for an automatic embossing machine to create a Credit Card for their customers.
Another application was at the New England Motor Freight Bureau which published the allowable legal freight rate charges for the various types of motor freight carriers in, at least, New England. The differing rates were constantly changing and the rate books were so dense with information that they used a tiny type face called Micro Gothic to conserve the space required for publishing all the information in them.
I eventually encountered Flexowriter model FPC8 which was used by the Digital Equipment Corporation (DEC) to create an paper tape for input to their PDP line of computers. The FPC8 was an 8 channel (code) double case type of Flexowriter. It was used as both an input and output device by many other developing computer companies in Eastern Massachusetts such as Data General which was a company started by someone who left DEC to start his own company; and Wang Labs, a startup computer company
Educational institution MIT had several laboratories that used Flexowriters as input output devices. MIT also had a large off campus site, MITRE, (Mass Inst of Technology Research and Development center).
The SPS Flexowriter was used for billing by companies such as Ford Motor Co's. parts distribution, GMC parts distribution and General Electric's Jet Engine plant for parts billing which created an output tape for input to their computer systems for inventory control.
Raytheon Co. used the Friden Teledata which transmitted the data encoded on paper tape generated by Flexowriters, much the way a Teletype machine works, except that it was a two step process instead of a single step process via Teletype.
Another product was the Friden Collectadata which was a series/group of units spread around a manufacturing plant such as Texas Instruments Co's assembly lines at locations where various components of assemblies were completed and the quantities and employee time consumed would be entered and transmitted to a collector/receiver which produced a tape of the accumulated data for input to a computer system.
A feature of a lot of these processes was the inclusion of utilizing the output ability of these Flexowriters to connect to an IBM 026 or IBM 047 which simultaneously generated punched cards for further processing on IBM computers.
Another use of the Flexowriter was generating an output tape for input to Numerically Controlled (NC) machine tools, essentially robots, used to manufacture a variety of parts for assemblies etc,
When the printing industry was going through a huge transition from the old hot-metal process of print (such as linotype machines) to the modern method of offset printing, companies such as Compugraphic Corp. and others utilized the output tape from a Flexowriter as input to their computerized photo offset typesetting machines.
But long before those companies existed, photo offset printers used the Friden Justowriter as their primary type setting machine. Once again, the Commercial Controls Corporation in Rochester N.Y. was where the Justowriter was manufactured.
Photo-offset was the outgrowth of the IBM proportional spacing typing machine, in the same manner as the Model FL Flexowriter was the outgrowth of their automatic letter writing machine. "Proportional typesetting" means the characters being typed have proportional widths. In the FL, FPC8, SPD etc. machines all letter and number characters are 1/10 of an inch in width.
In proportional type the character width varies and is measured in units. For example, depending upon the Font, (look) of the type, the characters. vary in width. In the Justowriter world, typically lower case character such as i or j are 2 units wide; while an upper case (Capitol) W or M is 5 units wide and most character are 3 or 4 units wide.
There were three models of the Justowriter. Justowriter Model JU generated a paper tape output that provided input to Model (I think) JUA which printed the *justified *newspaper columns, usually for local community weekly newspapers. It was also used to set the type for Flexowriter manuals such as is found [on various Web archives of manuals]. Model JUAA was capable of both generating a tape output and using that same tape as input to itself thereby producing its own justified output.
The Justowriter provides for the right end side (Margin) of a column of newsprint to be Justified. in the same manner as the left. In other words, both sides of the column of print were even. This was accomplished through a simple but ingenious mechanical computer.To this day I marvel at the process. It counted the number of spaces in a line, up to 8, in the line being typed and also calculated the length of the line and then mechanically computed how many units were needed to be added, *(never subtracted)* to make the line the proper length necessary for the perfectly justified line.
The generated tape from the initial typing unit was then placed in the *double reader *of the second machine which produced the justified, photo offset ready, column of type.
For the last word on Justowriters, I get depressed thinking about it. I bought a Model JUAA Justowriter which was little used and hung around my cellar for years, My wife and I had five children and not much room for unused things. I contacted the Smithsonian Institute because I knew they had some similar equipment and offered them the machine for free.
They never responded and I threw the machine away *much to my everlasting regret*. I did, at one time, have all the manuals and schematic diagrams for everything I had ever been trained on at Friden. But they were all totally ruined when my cellar was flooded and all was under water.
A Friden machine I was trained on was the Computyper. It was a Friden STW10 calculator electrically connected to an Flexowriter which performed billing functions. Its memory, about 1012 characters, was accomplished through the use of a of a Piano wire* [as an acoustic memory?]. Those 1012 characters continuously circulated linearly along the piano wire. Amazing to me.
In 1970, the year Friden's System Ten Computer system was introduced, I was sent to California for two 8 week training stints. Thinking back now, on the capabilities of that system, it was a quite remarkable product.
It had 10 memory partitions providing for 200 peripherals including IBM equivalent disc drives, card punches, card readers, tape drives, high speed line printers and operator consoles. That's except for the disc drives which very slightly interrupted other processing when data transfers occurred, all 200 peripherals could be in operation simultaneously.
An oddity, to me was the fact that when inputting from the console to the computer, each keystroke was sent 23 times to check for accuracy no matter how many consoles (terminals) were inputting at the same time.
Then, Singer Corp bought Friden. There were some System Ten problems that were known but not addressed. As time went by, not much, it became obvious that Friden was to be a cash cow for Singer Corp. and not much else. At that point, I left Friden for New England Telephone.
Years later, while conversing with old friends from Friden, I learned that the Friden employees were being forced to leave the company with little in retirement to show for their years of service. Ultimately, the Friden Division of Singer Corp, disappeared.
Charlie Murphy
The Web archive bitsavers.org, and its various mirrors, have a number of PDF's of Friden manuals. Charlie refered to one of them, the Friden Flexowriter Programatic Technical Manual of 1962. Here's a Web link to one archive's copy of that manual. - herb Johnson
http://bitsavers.informatik.uni-tuttgart.de/pdf/friden/SP8673R2_Friden_Flexowriter_Programatic_Tech_Man_1962.pdf
Copyright © 2013 Herb Johnson