I'd like to use this blog post to share some of my thoughts about Fedora's Spins -- both the idea behind having spins, and some of the marketing-related challenges that exist as a result.
What is the purpose of having different spins of Fedora?
There are over 10,000 packages in the Fedora repositories, and the question of how you choose some subset of those packages to present as a "finished product" is central to the spins idea. In a community as diverse as Fedora, shouldn't we make it as easy as possible for people to highlight specific pieces of Fedora's repository set, and share that with an audience? Spins allow for this, without requiring someone to install a potentially large set of additional packages that aren't part of a default installation, which makes the process of getting your machine into its final state more complicated and error-prone. Spins allows us to have multiple default installations, and lets users select what appeals to them from a menu of choices.
Doesn't all that choice simply lead to confusion?
It can, and this is the first marketing-related pitfall that we need to be wary of. I submit to you that spins fit into one of two categories -- strategic or cosmetic -- and that as an organization, Fedora wants to encourage the creation (an engineering activity) and the spreading (a marketing activity) of the strategic spins while redirecting the efforts of those who would create purely cosmetic spins to another area of contribution.
What do you mean by strategic and cosmetic?
The most popular Fedora 12 spins as of this moment are KDE, LXDE, XFCE, Games, Brazilian Open Office, Fedora Electronics Lab, and Education.
The three desktop spins are strategic because they highlight some of the primary choices and the leading desktop options in the open source ecosystem. It's strategic to have them because it demonstrates that Fedora cares about a wide variety of desktop options, and that we are a community that is capable of not only packaging, but also contributing upstream to all of these projects.
The Fedora Electronics Lab and Education spins exist to serve a niche, and I see the strategy in those as well. If we send Chitlesh to a class of electrical engineering students, it will make a fantastic impression upon them if they can perform a default installation of Fedora that provides them with all of the tools that they could possibly want for their lab work, homework, and studies. Swap electrical engineers for designers, and I think you've got a compelling strategic reason to produce a spin for that group of people, because those audiences will want a bunch of packages that the "average productivity user" probably doesn't care about. We can be the curators of Fedora's package repositories and provide people with options that they might never otherwise discover. Update: Rahul's blost post regarding the Fedora security spin fits exactly into what I'm talking about in this paragraph.
What about the Games spin or the Brazilian Open Office spin? I'd be curious to see how much those spins differ from the default desktop option before making any comments about them. Obviously they are attracting users, but that's different from trying to think in a vacuum about the strategic value of a particular spin. What does the games spin do other than just add a bunch of games packages? Does it have anything in it that helps with heavy graphical performance? Are there things we can do in a spin like that to really maximize the hardware capabilities of a machine that we don't do in (for instance) the default desktop spin? Those are questions that can enhance the strategic value of a spin. Does the Brazilian Open Office spin solve a particular legal problem? The spin feels a bit like a one-off special case. If that's true, is there a better solution that having a separate spin entirely, which can help consolidate our messaging?
What is your bottom-line?
The attention of an audience is finite. Successful marketing is about having as simple a message as possible, and amplifying that message as loudly and as far as you can. Choice is good, but too many choices means that none of the options are well-explained. Spins are a useful tool that have strategic value. The marginal cost of creating new spins is fairly low, but the strategic opportunity cost is the factor that should be looked at most carefully when we decide which spins to highlight and spend the Fedora Project's resources (both release engineering and marketing) creating and promoting.
One of my next activities will be to look more closely at this page, and to suggest any updates or diffs that come to mind.