l Home l News Archives l Search News & Comment l Email the editor l



Richard Chirgwin

Off-Topic on a Friday Afternoon

It doesn't really matter whether Microsoft money has gone into the latest piece of loopyville from the Alexies de Tocqueville Institution (a hint to their literary prowess: they don't know the difference between an institute and an institution). But surely Microsoft must wonder whether its money is well-spent whatever these guys do?

I'm just barely too young to have been around in the grandly subversive days of John Lions' ground-breaking Unix commentary, to have smelt the gunpowder as police stormed the rioters blocking access to the PDP-11, and so on.

But I can read history, which puts me a step ahead of the ADTI.

What's got me hot under the collar and in a mood for a rant is a dual combination: the ADTI has published a report in which it seeks to debunk the idea that Linus Torvalds invented Linux. Bit of a straw man really, since Linus doesn't think much of being called an “inventor” anyhow, but when you spent your childhood licking at too much lead-based housepaint, facts are malleable anyway.

The second characteristic is simply sloppy journalism.

And here's the sliver I want to shoot down:

“The study also raises the issue that Torvalds saw Unix source code. This was available in annotated source code that John Lions, a professor at the University of New South Wales in Australia, made available to his classes. The notes were widely distributed illegally afterward, and "many suspect that Linus also had the Lions notes," the report said.

(This comes from a report doing the rounds of the CNet sites. I certainly hope that ZDNet Australia likes its history better than its American owners.)

Doesn't anyone at the ADTI know anything? By the time Torvalds got to work, Lions' book wasn't “the” Unix source code; it was “a” Unix source code, and a hoary old one at that.

John Lions' work was about 15 years before Linus Torvalds. When Lions wrote his commentary, there were roughly three versions of Unix (not long forked so not much different); by the time Linux was created, the number was more like 30. And the ADTI author seems to think that a hardcopy of 15-year-old source code for a PDP-11 is just the place to get source code for an Intel-based PC.

Then there's the matter of the words “illegal distribution”. As far as I'm aware, the legality of distribution has never been determined by either legislation or court action, anywhere in the world. Something's not illegal, twenty years in retrospect, just because some shill says it is.

Lions is dead now, and for all practical purposes so is the Unix Time-Share System, Sixth Edition. Let's bury the system, and honour the educator.

Would you like to make a comment to the editor about this story or any other issue? If so click here.