Showing posts with label andy tanenbaum. Show all posts
Showing posts with label andy tanenbaum. Show all posts

13 May 2010

How to Become Linus Torvalds

Most people in the free software world know about the famous “LINUX is obsolete” thread that began on the comp.os.minix newsgroups in January 1992, where Andrew Tanenbaum, creator of the MINIX system that Linus used to learn about operating system design, posted the following rather incendiary comment:

On The H Open.

19 February 2010

Herding the Meta-Cats

In the famous online argument between Linus and Minix creator Andrew Tanenbaum during the very early days of Linux, one of the more memorable statements from the latter was the following:

I think co-ordinating 1000 prima donnas living all over the world will be as easy as herding cats.

On Open Enterprise blog.

21 July 2006

Tanenbaum Rides Again

For younger readers of this blog, the name Andy Tanenbaum may not mean much. But for oldies such as myself, it is highly redolent of those epic days when Linux was but a fledgling kernel, and taunts like "your mother was a hamster" and "Linux is obsolete" were thrown down like gauntlets.

I had the pleasure of interviewing Tanenbaum for my book Rebel Code, and it was fascinating to learn how close he came to creating what we now call GNU/Linux with his Minix. But Tanenbaum failed to do one crucial thing that Linus did almost without thinking: to let go. Understandably, as a professor of computer science Tanenbaum wanted to keep control of his teaching materials. But that one, tiny, reasonable brake was enough to stunt the growth of Minix and lend wings to Linux when it appeared in 1991.

Tanenbaum is still teaching, at the Vrije Universiteit in Amsterdam (another Dutch story, then - must be the Rembrandt Effect), and I was interested to note this piece about some of his recent work on developing an anti-RFID device. Good to see him still moving forward in his work. (Via openspectrum.info.)

29 March 2006

Linus Torvalds' First Usenet Posting

It was 15 years ago today that Linus made his first Usenet posting, to the comp.os.minix newsgroup. This is how it began:

Hello everybody,
I've had minix for a week now, and have upgraded to 386-minix (nice), and duly downloaded gcc for minix. Yes, it works - but ... optimizing isn't working, giving an error message of "floating point stack exceeded" or something. Is this normal?

Minix was the Unix-like operating system devised by Andy Tanenbaum as a teaching aid, and gcc a key hacker program that formed part of Stallman's GNU project. Linus' question was pretty standard beginner's stuff, and yet barely two days later, he answered a fellow-newbie's question as if he were some Minix wizard:

RTFSC (Read the F**ing Source Code :-) - It is heavily commented and the solution should be obvious (take that with a grain of salt, it certainly stumped me for a while :-).

He may have been slightly premature in according himself this elevated status, but it wasn't long before he not only achieved it but went far beyond. For on Sunday, 25 August, 1991, he made another posting to the comp.os.minix newsgroup:

Hello everybody out there using minix -
I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu) for 386(486) AT clones. This has been brewing since april, and is starting to get ready.

The hobby, of course, was Linux, and this was its official announcement to the world.

But imagine, now, that Linus had never made that first posting back in March 1991. It could have happened: as Linus told me in 1996 when I interviewed him for a feature in Wired, back in those days

I was really so shy I didn't want to speak in classes. Even just as a student I didn't want to raise my hand and say anything.

It's easy to imagine him deciding not to “raise his hand” in the comp.os.minix newsgroup for fear of looking stupid in front of all the Minix experts (including the ultimate professor of computing, Tanenbaum himself). And if he'd not plucked up courage to make that first posting, he probably wouldn't have made the others or learned how to hack a simple piece of code he had written for the PC into something that grew into the Linux kernel.

What would the world look like today, had Linux never been written? Would we be using the GNU Hurd – the kernel that Stallman intended to use originally for his GNU operating system, but which was delayed so much that people used Linux instead? Or would one of the BSD derivatives have taken off instead?

Or perhaps there would simply be no serious free alternative to Microsoft Windows, no open source movement, and we would be living in a world where computing was even more under the thumb of Bill Gates. In this alternative reality, there would be no Google either, since it depends on the availability of very low-cost GNU/Linux boxes for the huge server farms that power all its services.

It's amazing how a single post can change the world.