I hadn’t intended to write even this episode, but was encouraged to do so today by Z – thank you darling! And now I feel like doing another one. If I’ve put any of you on suicide watch, please alert me and I’ll shut up on the subject and write about something else, or nothing, instead.
The first thing I had to do was pass the interview, about which I remember very little except for a question from a madman called Duncan, something like “do you prefer applied or theoretical psychology?” I think I got a left-eye wink from Barrie, the Head of Department, which I somehow took to mean ‘applied’, which turned out to be the right answer. This was the same Duncan who later, unannounced and uninvited, burst into a test session and stormed around the comms control room in Great Tower Street randomly pulling plugs out of wall sockets, in order, as he put it, to test the system to destruction – which he almost managed.
The second thing was to relocate across the river to London Bridge, also not a problem. The tube journey was one stop more; the pubs were better and much cheaper. There were rules about who went to which pubs and when. Computer Department was a world apart from the rest of the bank, and contained cliques within cliques. You only socialised within your own team, which was project-based; and within each team analysts and programmers tended to keep their distance.
Nobody explained any of this to me; indeed I don’t think they were fully aware of it themselves. It was an astonishingly isolated society. I got the hang of it fairly quickly, and joined in, of course. But I think that an injection of influence from the world beyond the castle, in the form of me and a few others like me, helped to break down both the monolithic elitism and the internal factionalism, much like in the former Soviet Union. As in that case, though, we were careful to close ranks when the outside world started to think they might know better than us what they needed. If a ‘user’, as they were disdainfully called, should appear down the Wheatsheaf one Friday, you could hear the air freeze.
SWIFT ruled, and I became Mr SWIFT. The zealots in Brussels were determined that any sphere of business involving communication between two banks was fair game for messaging standardisation, and hence analysis of the actual underlying business structures and transactions. I think it’s fair to say that such an exercise had never been undertaken before – the creation of a set of business rules that would enable pairs or groups of organisations with no prior relationship to conduct trade in a common language. Nor has it been done since. It was a world-changing achievement, and it’s worth briefly digressing to explain what I mean by that.
Nowadays, we take telecommunications for granted, obviously. To be technical, TCP/IP gets the bits and bytes from A to B, http provides an agreed method for unscrambling them, RSA for making sure they’re safe, html for a common grammar in which the content of the message is assumed to be written. But none of these say anything about the content. The ‘l’ in html and gml* is a semantic error, because a language consists not only of phonics, grammar and syntax, but also vocabulary. SWIFT standards were a brave initiative to supply that vocabulary on top of all the rest. A financial Esperanto.
After a lot of faffing about, designing systems that were never going to get built, simply because SWIFT had reached too far beyond the banks’ catch-up capabilities, my next big project came from a completely opposite direction: inside the bank. What was called the Treasury was flexing its muscles, and knew that they would be harnessing computers.
* html – hypertext mark-up language – grew out of an IBM product , gml, generalised mark-up language, which I used to write my reports back then in the 70s. I could probably write this post in gml – :p. means new paragraph, etc. Not html though.