Brian Reid, the old fuddy-duddy fighting back, turns out to be the Brian Reid, of Scribe fame. Scribe was an early word-processing application that was one of the first practical and public tools for showing that descriptive markup, rather than procedural markup, was workable.

The story is that IBM’s Charles Goldfarb was due to make his big presentation on their GML system to a conference in Switzerland in 1981, but Reid had a paper at the same conference which turned out to present a lot of the same material: stolen thunder! GML morphed into SGML now XML, while Scribe influenced the direction of word processors by showing the practicality of styles (think CSS!).

Reid revisited his 1981 paper at an SGML/XML conference keynote in 1998, which is still online though large (10meg PPT?) The paper includes some interesting thoughts on why markup is wrong-headed.

Reid’s dispute with Google is quite interesting to me. Last year or so, I was looking around to see whether there were any interesting opportunities in Sydney, and Google contacted me to come in for an interview. When I arrived, it turned out to be a rather odd interview for a programming job which was pretty much unconnected with anything I had been doing for the last 20 years, but the questions would have been good for a recent graduate, sort of like interviewing Donald Duck for a position as an egg, if you know what I mean; the people seemed super nice, but I think there was a bit of mutual mystification as to why I was there. The impression I got was very much of a mono-culture: the founders wanted people like the founders. It seemed that standards were not on the company’s horizon at all.

But perhaps this is a new way to break up monopolies: everyone above 35 goes to one company, all the rest go to another!