The Early Days of Computing, Part Two

I worked for Sweda Litton designing a POS (Point Of Sale) system for Boots The Chemists. The relevance for my future computing work is that I invented  a table driven methodology for controlling POS terminals. For each part of a transaction sequence there are legitimate and logical states that you can go to next. Depending how you arrived at that state there was another state you must return to if the operator presses cancel. For example, if you press sub-total then you are in the payment state and must enable certain buttons like Cash, Credit, Coupons etc.

I believe I was the first person in the POS industry to propose and produce fully programmable POS terminals.

Mid 1970’s we were in the age of distributed computing. Smarts held on a central computer and your work was done via a teletype or VDU (Visual Display Unit).

POS systems were VDU’s with an interface to fire open a till to take the money. I was involved with the first implementation of a system for Argos late 1970’s and another for MFI.

Today we take networking for granted but in the late 1970’s we used IBM Token Ring to chain together POS terminal. With centralised computing you needed to know which terminal was requesting a price lookup or customer account reference.

Token ring local area network technology is a communications protocol for local area networks. It uses a special three-byte frame called a “token” that travels around a logical “ring” of workstations or servers. This token passing is a channel access method providing fair access for all stations, and eliminating the collisions of contention-based access methods.”

Terminals communicated with their own language by transmitting data with reserved header bytes:-

  • SYN
  • POL
  • ACK
  • NAK
  • ENQ

(Not a complete list)

  • SYN – Synchronise was a request for all stations to reset their communications
  • POL – Polling all stations “Who’s there?”
  • ENQ – “I’ve got some data”
  • ACK – “I got the data packet”
  • NAK – “I didn’t get that data packet”

Lugosi, Going Postal

Lugosi, Going Postal
Late 1970’s, Commodore PET (Personal Electronic Transactor)

A basic POS terminal output its transactions to magnetic tape. The PET could read the contents of a tape. So, “Can you write a program to analyse the data?” So I did!

Computing was about to enter its “Personal Phase”. I presented a POS terminal I had programmed at an exhibition and I was stolen, with an offer I couldn’t refuse, to develop my own terminal. Based on Intel 8080 I had to write in assembler and needed a development machine.

Lugosi, Going Postal
Xerox 820. MSDOS with an assembler compiler, twin 8 inch floppies

The disk controller amplified the “clink” “clunk” of the disks loading and being accessed.

I was probably the first person in the world to make MSDOS multi-user. I needed the primary display to represent the POS register screen but also wanted a second screen that acted as the Managers terminal. I found a document describing MSDOS interrupt handling and replaced the keyboard interrupt handler to poll which keyboard was being pressed. Then I had to write a CLI (Command Line Interpreter) to work out what was being requested and which screen should reflect what was typed. I bet you didn’t know that a keyboard has to be “debounced”.

debouncing is any kind of hardware device or software that ensures that only a single signal will be acted upon for a single opening or closing of a contact .”When you press a key on your computer keyboard  you expect a single contact to be recorded by your computer.”

You have to work out, by a timer, whether someone wanted to type “mm” or they only meant one “m”. I had to write that in code.

My sponsor wanted me to be able to take my software to the hardware builder and his home to discuss the project. Enter “Luggable Computing”.

Lugosi, Going Postal
Osborne One with twin floppies

The Osborne One had no battery but there was an after-market external battery option. My problem was that all my development was on 8 ½ inch floppies and I had to get down to 5 ¼. I seem to remember the solution was RS-232 transfer using X-Modem controls.

We built the POS terminal and showed it at a major UK POS exhibition. My sponsor, a well-known, at the time, electronics, audio, art and music producer genius, deepened into his bi-polar depression and was committed for pulling an imaginary Smith & Wesson on the police who were climbing the long staircase to his top-floor, Mayfair flat. Our second sponsor, a major retail giant couldn’t sustain his interest. I wasn’t getting paid so we had to part ways. I joined Mitsubishi Electric as European POS Consultant and worked closely with the Japanese.

And that’s a whole new story!

The Early Days of Computing, Part One

© Lugosi 2019

Audio file