My one and only encounter with Wordstar...
...was back in the very early 90's (around 1990-1991).
As part of of thesis, I had the task of building a plotter, but it had to have a function and the one chosen was to read a barcode and reprint it. Oh I had to design and build the barcode wand, as well.
I mean... really... of all the use cases, this was effectively an overcomplicated photocopier, but I digress.
The kit we were programming on was some esoteric all in one device that had a hexadecimal keypad, a series of 7 segment displays and a few limited ports - a serial and parallel if my memory serves, as well as an edge connector that exposed one of the buses.
We did have the luxury of being able to write the code on a PC which allowed us to use assembly language, then generate the hex output, and print it, but then manually input it on the keypad.
My first hurdle was the sheer size of the program itself. It wouldn't fit - the lookup table alone ran to something like 4x the available RAM on the board, so I had to design and build a memory expansion. That went onto the edge connector. I seem to recall from memory (no pun intended) that the system itself could recognise up to an additional 32kB of RAM as long as you got things like the timing and trigger circuits correct.
The next issue was that the program wouldn't fit in the application we used to code on the PC (I forget what it was - it was over 30 years ago!). So... I was given a copy of Wordstar and told to write it in that... which was kind of useless given it had none of the compile-to-hex capabilities I needed.
But given that I now had a ton of RAM spare, I decided in my infinite youthful wisdom that it would be easier to transfer the program straight from the PC rather than coding it in manually every time so I reverse engineered the original program and made it read in the Wordstar file, strip out the bits it didn't need and pipe the output into the board via serial cable. Which also meant of course I had to write the program to transfer it. I think that I burned that to an EPROM in the end to save having to type it in manually each time, which meant I kind of had my own kit.
Every character from A-Z (upper and lower case) and the numbers 0-9 was in a lookup table that was simply a series of commands for driving the stepper motors in the plotter (which was very simply an x/y and a relay to lift and lower the pen).
It all worked and it was an excellent learning experience for projects in later life and how they have a tendency to grow out of all proportion if you don't get the details right from off the bat.
I learned more about barcodes than anyone needed to know. It was actually fun for the most part - just a series of ever expanding challenges to get it working but other than being a learning exercise it was otherwise utterly pointless in practice. :-)
Biting the hand that feeds IT
