The Cautionary Tale of J Random Newbie. Found whilst reading Joel Spolsky’s commentary on programming culture, which discusses ESR’s new book The Art Of Unix Programming. Spolsky’s article is great. Sounds like the book is pretty interesting too.

Spolsky’s core point is the differences in culture between Unix and Windows. He discusses how this applies to Linux on the desktop. It is his view that Linux on the desktop will fail (or at least has done so far) because the Unix culture values stuff that’s useful for programmers more than stuff that’s useful to users. He uses OS X - a version of Unix made for users, not programmers- as an interesting example . OS X dares to do things like superimpose a highly simplified filing GUI on top of /, something which works well, despite much paranoia from the community before they got to play with it (including myself). It’s because of things like this that OS X is now the most broadly distributed version of Unix. As an aside, OS X’s deft blending of user and programmer tools is why I love it and would never switch over to either of the polarised OSes unless forced to do so.

I still think it would be nice to have a desktop Linux, or something like it, to compete with the other two big desktops. I like to believe that the value there is in the openness. But after reading Spolsky’s article, I wonder if this is just practically valuable to programmers and hackers and not really all that useful to the end-users that make up 99% of computer users (J Random Newbie discusses one reason why). The hope is that the increased openness helps lead towards better software for everybody. I think it’s hard to say if this is actually true. If we went back in time and replaced fledgling Windows 1.0 with an open source OS that then grew to the size of Windows today, what would be better about that OS, the apps that run on it or the world at large? Could an open source OS even have grown that much without the might of the single unified Microsoft behind it?