I've long thought the key reason that Linux beat the BSDs was the license. Linus Torvalds seems to think so also.
4 plus ones
Shared publicly•View activity
- But did he know that at the time he was choosing a license? I can't help get the feeling there's a little bit of confirmation bias [https://en.wikipedia.org/wiki/Confirmation_bias] going on here. There are plenty of failed projects that have also used GPL(v2).
I've always felt Linux's success was due to the same processes that cause other, random, projects in tech to succeed: people, timing, and sheer luck ;-)
I've often heard this "Linux isn't fragmented" statement and scratched my head in wonder. Back in the day you wouldn't dare run a kernel.org kernel on anything non-x86. Those kernels were outside mainline for decades before anyone cared enough to merge them (and in some cases they still are the defacto kernels). Many vendor and distro kernels are so heavily patched, they can barely be labeled "Linux" anymore!39w
- I believe Linus chose the license because it codified the so-called "golden rule" which is widely seen as a valuable ethic. So I think he did choose it because he thought it would "work".
"defining factor" is certainly hyperbole - there are multiple factors, such as those you list, which were all necessary. However I think GPLv2 can be seen as an important factor in the transition from "successful hobbyist project" to "successful commercial project" because it provided the legal framework for corporations to work together in the same way that hobbyists already did.
So it possibly is justified to claim that the GPLv2 was a defining factor in the "commercial" success of Linux. Then there could be a separate discussion on whether the commercial success was necessary for the project to be "successful" in some absolute sense.39w