Nobody wants a "me too" Technology
As much as I like Linux (some flavors thereof, not *all* Linux distros), it still irritates that living *** out of me how they ceaselessly continue to copy ideas from OSX and Windows. This is why many non-Linux folks refuse to "switch" to it for a desktop alternative. For servers, the debates are entirely different. For desktops however, there is nothing Linux is doing, or has done, to revolutionize the PC desktop world in general. All of the major innovations seem to have come from Microsoft and Apple over the past 10 years. Next to them, it would be software vendors like Adobe, Autodesk, and scores of lesser known vendors and developers targeting one of those two platforms. Every Linux nut I know (and I know quite a few I have to admit) gets asked the same question immediately after they hit me with their arguments of superiority: "What major innovation has Linux introduced that has been copied by Windows or OSX?" They can't name one single item.
Beryl? Compiz? GNome? KDE? I'm sorry, nothing really "new" there.
A million Windows app clones? Pa-leez.
A few Apple OSX app clones? I'm falling asleep.
ZFS? You've got to be kidding me.
Modular installations and replaceable shell UI's? Nope.
Free? Ok, that's a given.
This is what the Linux and open source community needs to address. They need to stop focusing on jumping up and down in the back of the room, hand waving hysterically in the air, and blurting "ooh! ooh! Me too! I can do that to! Watch me..." It's stupid. It's embarrassing. Equally as embarrassing as Microsoft trying to compete in markets they don't belong in (search, browser widgets, portable media players, etc.). When Linux starts to actually "lead the way" with innovative ideas that make others chase them, that's when the public will take them seriously on the desktop end of things. When software vendors start targeting their products at Linux equally with Windows and OSX releases, that's when it will matter. Words like "free", "stable" and "secure" matter mostly to server environments, not desktop environments. Desktops need to be "cool", "simple" and "exciting" as well as "compatible" in order to make a dent on the shelves of Best Buy or OEM PC shipments. "Me too" doesn't cut it.
I like Linux. It has a place in data centers and server farms. It has a place in the embedded world beyond any doubts. It has other places as well. But on the desktops of Joe and Suzy Homemaker it ain't quite there yet. In fact, it's a longer way off than anyone wants to admit. Every year is claimed as the "Year of Linux on the Desktop" and every year slips by without making any tangible gains. The numbers seem to clearly point to an increasing lead by the Windows and OSX camps. People really tune in when Apple announces a new OSX release like Leopard. They sort of tune in when Microsoft announces a new Windows version. Nobody pays much attention to a new Linux release, mainly because there really isn't one. It's always an incremental upgrade and the funny names make it tough to take seriously in the mainstream press. "Longhorn", "Leopard", "Vienna", "Tiger", and so forth sound more serious than "Feisty Fawn" or "Hoary Hedgehog".
What the Linux "community", if there really is such a thing, rather than a name applied to a dis-integrated scattering of developers around the world that each focus on their own agenda, needs to do is actually unite behind a common face or maybe a single vendor. They need concrete organization and representation. IBM can only do so much for them. Oracle and Sun are really only involved for their own revenue models, not the good of the cause in general. They also need to focus on real innovation. They need to shake the Earth with some features nobody has ever seen before, and that nobody can ignore either. Something so cool that Microsoft and Apple have to chase, instead of being the ones chased. We need impressive results, not more "me too" stuff.