Log in

Ritchie, Jobs and hyperbole - Bela Lugosi's Dead, Jim
October 15th, 2011
05:28 pm
[User Picture]


Previous Entry Share Next Entry
Ritchie, Jobs and hyperbole

It’s a shame that Dennis Ritchie and Steve Jobs died at around the same time. I don’t say this out of a belief that the news about the latter somehow eclipsed the former; indeed I’ve seen a number of mainstream obits and no shortage of blog and Usenet traffic about the DMR.

Rather, the reason is the claim that Jobs would have been nothing without DMR (or at least, nothing in computing; for example I saw one suggestion that he might have instead been an especially successful car salesman). I believe this is a ridiculous exaggeration and has more to do with the coincidence of their deaths than reality. While DMR’s contributions were important both in general and to the software underlying modern Apple products, from some of what has been written you might imagine he personally invented the field of computing from the ground up.

Jobs’s career got its start with the Apple 1. It’s hard to understand what DMR’s effect on this is supposed to be. It wasn’t programmed in C and it didn’t run Unix. If you want an innovation that made the Apple 1 possible then Chuck Peddle’s 6502, introduced the previous year at a substantially lower price other comparable microprocessors, seems like a better choice.

Much the same could be said of the Macintosh; as far as I can tell its original system software was written in a combination of assembler and Pascal. (I’m having trouble finding definitive statements on this subject, so I would welcome corrections; very little of that software has, to my knowledge, been published but you can download the source to MacPaint and QuickDraw and they are written in Pascal and 68000 assembler.)

There is a better case to be made for NeXT. The system software for these machines was based on Mach (which at the time, as I understand it, was a research derivative of BSD Unix, though the end goal was something a bit different) and programmed in Objective-C. Both inarguably built upon DMR’s work, and both are still to be found in modern Macintoshes and iPhones, so there is clearly an important technical debt there.

Nevertheless I find it hard to believe that Jobs would just have given up on computing had one or both not been available - the OS research community that produced Mach would surely have still existed in some form and quite plausibly produced something equally suitable, or alternatively NeXT’s engineers might simply have started from scratch. OS/2 and Windows NT date from much the same era, as other examples of private-sector OS innovation. As for the implementation language we might imagine modern Apple computers programmed in some kind of souped-up Pascal dialect.

In summary while DMR’s work has found its way into important roles in Apple’s products (among others), so has that of many other creators, and his contribution was probably not indispensable. This isn’t to underestimate the importance either of C as a portable language suitable for low level tasks; nor of Unix as a portable operating system and, in the long run more importantly, as an operating system interface supporting many implementations.

Tags: ,

(Leave a comment)

I deny everything Powered by LiveJournal.com