A Short Post in 1991
It is the summer of 1991. At the University of Helsinki, a 21-year-old computer science student named Linus Torvalds is spending most of his time at a PC built around Intel's 386 processor, then new and powerful. He is using Minix, a teaching operating system written by professor Andrew Tanenbaum so students could learn how operating systems work. Minix was great for learning, but limited, and Torvalds wanted to use everything his machine could do.
So he started writing something of his own. At first it was a tiny program that could juggle two tasks at once. Then a disk driver, a file system and other pieces were added. With no grand plan, that small program was turning into a kernel: the part of an operating system that sits between the hardware and the programs and shares out resources.
The post that started everything
On 25 August 1991, Torvalds posted a short message to the Usenet group comp.os.minix, where Minix users gathered. The subject line asked: "What would you like to see most in minix?" He wrote that he was doing a free operating system for 386 and 486 machines and added, modestly, that it was "just a hobby, won't be big and professional like gnu". In the same message he asked readers what features they wanted and what they disliked about Minix.
Those words are quoted endlessly today because what followed was the opposite of what their author predicted. But the real value of the post was not its modesty; it was the asking. Torvalds didn't announce a finished product. He showed unfinished work and invited others to help shape it.
From one person to a crowd
A few weeks later, in September 1991, the first version of the kernel was put on a university file server in Finland. It was raw and barely worked, but it was enough to attract the curious. Programmers in different countries, who had never met Torvalds, tried it on their own machines, reported bugs and sent patches.
What made this possible was the young internet of the day: newsgroups, email and file servers. There was no head office, no contract, no budget. The motives were simple and human: curiosity, the joy of solving problems, and the wish for a tool they could look inside and change. Commercial operating systems were expensive and their code was closed; this project left the kitchen door open to everyone.
Torvalds took on a new role too. He was no longer just the author; he was also the referee. He had to decide which patches to accept, which to reject, and why. That role of maintainer, which we will return to again and again in this series, took shape in those first months.
The turning point
The turning point of this part is the moment a personal project became a collective one without anyone deciding it should. Torvalds owned the code, but the project's value no longer lay only in his code; it lay in the people gathering around it. Fourteen years later, in July 2005, he joked that "this thing ended up being a bit bigger and more professional than I originally even envisioned."
Lesson for businesses
Show unfinished work to real users early. Many companies hide a product until they consider it complete. Asking "what do you want?" at an early stage exposes mistakes sooner and turns users into partners.
Second, modest claims build trust. Torvalds didn't promise a revolution; he said it was a hobby. That honesty meant people arrived without false expectations, and what they found was better than what they had been told.
Third, in collective projects a leader's job is less about doing everything and more about choosing well: accepting good contributions, politely declining weak ones, and explaining the reasons for both.
Next: Torvalds' kernel was not a complete operating system on its own. The missing pieces had been built years earlier by another group with a very different motive. And one small licensing decision in early 1992 changed everything.
منابع / Sources
- "What would you like to see most in minix?", Linus Torvalds, comp.os.minix (Usenet; archived copy at CMU), 1991-08-25 — https://www.cs.cmu.edu/~awb/linux.history.html
- "Historic Linux kernel versions (0.01 onward)", kernel.org, 1991– — https://www.kernel.org/pub/linux/kernel/Historic/
- "Git approaches 1.0 (quotes the 1991 and 2005 Torvalds lines)", LWN.net, 2005-07-27 — https://lwn.net/Articles/145194/
