In the embedded space, there's been a lot of talk about whether Android is fragmented, and if that fragmentation will ultimately hurt Android, because developers won't know what version to code for and users won't know which one to use.
The former argument may be resolved soon, with Google detaching a lot of the current core functions in future versions of Android so that no matter which version a phone/device vendor picks, consumers will be able to download up-to-date versions of those once-core applications (like Gmail).
The latter argument was never true in real-world application: users don't give a plug nickel about what version of Android they've got... they just know it's Android (and they may not even know much more than that, depending on how technical a sales pitch the phone store salesperson delivered).
This goes beyond the cute little green robot icon... that's branding, which I've discussed before. But branding is part of a broader sense of identity--something else that Android has in spades.
Identity, like beauty, is in the eye of the beholder, and can vary from different perspectives. I, for instance, look at my mother and see... my mother. But I know that there's thousands of people out in the world who see her as their teacher, or their child's teacher. To them, that's her core identity. Of course, neither one is wrong, but unless I complete my plans for world domination, I have a suspicion my mother will be remembered by more people for what she taught than for being my mom. And that's cool.
The same applies to products like Android. I look at Android and I see "Linux-based embedded platform that is currently kicking the iPhone's butt," but likely many consumers are seeing the same platform and thinking "this is cool like the iPhone, and I can use it on my network and not switch to that lame AT&T."
I would suspect that it's this identity for Android that Google, Verizon, HTC, Motorola, and all the other players in Android space like to see. Verizon has sure pushed this impression in its ad campaign against AT&T--a campaign AT&T is fiercely fighting on the airwaves, even as public outcry about the problems of the AT&T network continues beyond any negative ad campaign from Verizon. This has become, then, AT&T's current sense of identity.
Identities can change, deliberately or otherwise. Once Google was perceived as the savior from the shackles of Microsoft. Now, that identity is increasingly tinged with suspicion as people start to wonder just what Google plans to do with that data, anyway. Facebook was once the hip place to connect, and now it's increasingly becoming identified as a swelling monolith of our personal information.
Identities are deeper than branding, because it's the reflection of how a company or product is really perceived--rightly or wrongly--no matter how that company or product is packaged. "Wrongly," because sometimes those perceptions are unfair or completely wrong.
The question of identity has been at the forefront of my thoughts lately as I examine the leading desktop distributions of Linux for various reasons. Because of the maturity and similarity of Fedora, openSUSE, and Ubuntu, it's becoming a challenge to get a sense of identity for each of these projects.
Of the three, Ubuntu is probably the easiest to identify: most popular desktop Linux distribution. That's a laudable goal, but right now that strong sense of identity could work against Canonical, which is also trying to position Ubuntu as a strong server platform and a cloud client. Look for a push to build some sort of meta identity for Ubuntu soon, I would expect.
The question of identity becomes murkier when exploring openSUSE and Fedora. These are each great distros, with few differences between them, other than preferred desktop environments (KDE and GNOME, respectively) and packaging systems. openSUSE has YaST and the openSUSE Build Service behind it, and Fedora does well with yum and the venerable RPM package format--as well as PackageKit, coming with Fedora 13.
And while I am sure the individual members of each of these projects will be able to point out the technical benefits of their respective distros, such arguments do little towards revealing the identity of these distributions. What is Fedora known for? What is openSUSE known as?
These are questions that go beyond the so-called top three. You can apply the question of identity to any Linux distribution: "what makes insert distribution here so special?" And no, the answer isn't "because I like it."
Some distros can answer this question more strongly than others, but there is a definite blurring of individual distributions' identities--more so every day as delta between the technologies used by distributions continues to narrow.
This is the kind of thing that will bring out the Linux detractors with arguments like, "if all of these distros don't have a strong sense of identity for users, then what does that mean for Linux as a whole?" Like it or not, that's a fair question. Is Linux harmed by so much apparent redundancy?
I am still weighing this, though my initial gut reaction is "no." There's simply too much collective wealth to be found in the sharing of open source code to ignore the value of diversity.
The solution to the identity question is to strengthen the sense of identity these distributions already have within their own communities. Fedora people have a sense of purpose and identity about their distribution--so do the openSUSE folks. They just need to enhance that internal identity and project that to users.
Distributions need to reveal their true identities to give users better choices.