Psion Stories #3 - The Sinclair QL

The Gang of Four

Psion was already scaling up it's team to produce a set of business applications when Sir Clive Sinclair approached them to write some for his new home business machine, so it was pretty much inevitable that the two would reach a deal.

The four principal writers for the titles were known as the gang of four, and were (contrary to the labeling in the article that went with the above photo) from left to right; Martin Stamp (Quill), Charles Davies (Archive), Colly Myers (Abacus) and Martin Brown (Easel).

Charles was head of development and along with Martin had worked on Psion's previous business titles for the ZX81 and Spectrum. Colly had recently moved from South Africa where he had been responsible for setting up the division of Psion over there, and Martin was brought on especially to work on Easel.

When the photographer settled on that location for the photo, I started to get out of my chair as my desk was right in shot. Colly stopped me though, stating that my work was far too important to disturb, and that any photographer worth his salt would be more than capable of positioning them all to obscure me. The photographer was in no mood to argue, so I stayed put, the gang of four were duly carefully arranged in front of me, and I got absolutely nothing done, as it is quite impossible to work with a photo shoot taking place right by you!

Developing for the QL

As well as the gang of four, there were 4 more of us working mostly on the QL version (Chris Jakob, Chris Costello, Peter Dain and myself). It would be a while before we had an actual machine though, and then for the majority of the time we'd only have just the 1 prototype machine.

This wasn't a big problem though, as Psion had not long taken delivery of a VAX 11/750 mini computer, and we developed a library of functions called PLIB that sat between the applications and any host operating system we cared to port them to, such that we could all work mainly on the VAX, and just cross compile a version to the QL as and when needed. Thus the QL prototype when it arrived sat on Chris J's table, as he was primarily responsible for the QL specific parts of PLIB, and the rest of us would book time on it as and when we needed it.

Chris J would more often than not get a new version of QDOS from Sinclair's Tony Tebby everyday, and he'd then set about incorporating any new features and changes regarding it into PLIB. One day a version of QDOS arrived where Tony had decided to switch all the graphics functions from passing coordinates in X,Y order, to passing them in Y,X order. Chris J then spent most of the rest of that day going through PLIB switching all the graphics function calls from the old to the new calling convention. The next day Tony had had a change of heart and had switched them all back again!

The QL tea towel

The big secret that all the computing press was talking about at the time was which processor the QL would use. Unfortunately the QL prototype consisted of just 2 boards in a rather large open box, making it all too easy for prying eyes to see the 68008 processor within, and we could get quite a stream of visitors passing through the office. It was only some quick thinking on Chris J's behalf that prevented it's discovery the 1st time, and forever after that, the tea towel that he had grabbed from the kitchen and hastily draped across the front of the box was kept by the prototype ready to be sprung into action.

Sir Clive Sinclair's visit

I lot could be written and tested using the VAX version, but as it was only character based this did throw up some problems, particularly when developing a business graphics application. You could get graphics terminal for the VAX, but these were inherently expensive, so instead Martin B wrote a graphics terminal emulator for another colour machine of the time and we used that. So it was that when Sir Clive came to visit us early on to gauge our progress, and shortly after his infamous fisticuffs with Chris Curry of Acorn Computers at the Baron Of Beef pub in Cambridge, we had to demonstrate QL Easel to him displayed on the screen of a BBC Micro...

More Prototypes

Eventually we asked for and got a 2nd prototype box (and tea towel) as we started work on hand coding PLIB as the applications were getting too big to fit in the QL's memory. Then not long before the launch we finally got our hands on some of the 1st machines off the production line. Unfortunately the power supply circuitry on these was directly behind the microdrives, and as this got hot it warped the case enough to stop the microdrives from working. The solution was to unscrew the case and operate the machines with the keyboard lifted off the top and placed in front.

When it came to the press launch the applications were still not quite fitting in memory, so we planned to use the extension ROM slot just for the launch. Unfortunately QDOS had also outgrown it's space allocation and Tony Tebby had already grabbed this slot, so we had to do a quick rethink and switch to using the cartridge port instead.

Perhaps the saddest tale is what befell those early prototype boxes. In a space saving tidy up awhile afterwards Charles decided to throw loads of old stuff in a big skip, including the two prototypes. Only I seemed interested in preserving them for posterity, and at the time had no space of my own for them either. I deeply regret not grabbing them anyway. I don't even know what happened to the tea towels.