"Linux" desktop/laptop marketshare is indeed miniscule (estimates vary between 1% and 2%) and in this context rethinking of strategy is in order. The fundamental problems are: 1) "Linux" is not preinstalled on nearly all new personal computers; installation/configuration is perceived as risky for data and complicated by ordinary users. 2) "Linux" has inexistent marketing support to entice and reassure ordinary users. The first problem is related to the second - OEMs obtain no marketing advantage by preinstalling GNU/Linux; there are no effective selling points for the software against Windows, the chief competitor. What is "Linux" anyway? For most nontech users, it is the desktop or "window manager". Millions of users purchased the first EeePC netbook; these users could tell us that they tried "Linux", without being able to name the Xandros desktop - indeed, they might only be able to affirm that "it wasn't Windows". Supermarket chains offer specials on "Linux" netbooks, but one would be hard-pressed to know if GNOME or KDE (or Xfce, or LXDE, etc.) is offered - the main selling point is low price. And for the uninitiated who hear about Ubuntu or Fedora or openSUSE or Mandriva (or Debian or Gentoo or ...), the desktop is totally out of the picture - not one of these distros even mentions which desktop users will work with by default. More often than not, desktops are referred to as mere "upstream" projects, reduced to incorporated code, a "shell" or "GUI". This tradition of minimizing the key role of the desktop hurts GNU/Linux marketshare because for the ordinary user, the desktop *is* the system. Casual users of free Linux OSes (e.g. people who purchase entry-price netbooks with, say, Linpus) attribute "Linux" to the first desktop they encounter. This confusion is understandable - in the Windows and MacOS ecosystems, there has always been just one desktop associated with the underlying system, with very few exceptions (Apple's dual-desktop OS 9-OS X during the transition in 2001-2002, Xerox TabWorks preinstalled on Compaqs way back in the Windows 3 days). This lack of identity keeps GNU/Linux systems in the margin to stay. Distros and desktops have not worked together enough to promote a "Linux" brand. In this context of lack of support by distros, the best approach for desktops is to minimize the underlying system(s) and market directly to users. Browser projects such as Chrome and Firefox take this approach (Mozilla marketing work is excellent). Google is doing this with Android (with of course massive OEM deals, attacking problem #1 directly). Jolicloud is perhaps a better case study, focusing on what ordinary users want to do with apps and data across multiple devices. This is by and large my approach with Sugar (I say by and large because Fedora figures more prominently in our marketing than other distros for historical reasons). So should distros give up on branding? Absolutely not. Distros could very well take the approach of "ingredient" branding while supporting a "Linux" brand. In the short term this would have no effect because of the marginal marketshare and brand awareness of each distro. However, that could change if OEMs come on board and for that to happen, I believe it is essential that 1) a "Linux" brand be developed and 2) desktops be marketed as differentiators to Windows. So the best way forward is for distros and desktops to band together, to create a brand to rival Windows (or even OS X) and to allow desktops their place in the sun. However, for that to happen, there must be a fundamental change in the way FOSS projects are managed - marketers must have a seat at the table. This is blasphemous in the FOSS context where there are so many engineers suspicious of marketing, including many who started their careers with marketing-driven proprietary software. Many FOSS projects consider marketing's only role is contributor recruitment, leaving the challenge of mass marketing - increasing marketshare - to someone else. Year after year, there is no one else, and growth is incremental with no possibility of a breakout. We have no marketing leverage to offer OEMs. What's the next step? A marketing summit, a working group? Through the Linux Foundation? DesktopSummit? elsewhere? I am not sure. But joint work in this direction could have very positive results. Sean Sugar Labs Marketing Coordinator Some analysis from the past three years: SJVN Five ways the Linux desktop shoots itself in the foot http://blogs.computerworld.com/14911/five_ways_the_linux_desktop_shoots_itself_in_the_foot Alexander Wolfe: Linux AWOL From Desktop Upgrades http://www.informationweek.com/blog/main/archives/2009/11/linux_awol_from.html 7 Reasons Why Linux Won't Succeed On The Desktop http://www.informationweek.com/news/201807072 Bruce Byfield: Branding: Even For Linux and FOSS, It's Everything http://www.linuxplanet.com/linuxplanet/reports/6776/1/ Serdar Yegulalp: More Reasons Why Linux Misses The Desktop http://www.informationweek.com/blog/main/archives/2009/11/more_reasons_wh.html On Wed, Apr 20, 2011 at 5:12 AM, Rahul Sundaram <metherid@xxxxxxxxx> wrote: > Hi > > http://dissociatedpress.net/2011/04/18/fedora-and-gnome-branding-drama-missing-the-big-picture/ > > "I’m not saying this to belittle Fedora — but to put this whole thing > into perspective. Arguing over branding here is /just plain silly/. > Depending on which side you’re on, it may validate your “team” (GNOME > vs. Fedora, I guess) but it’s /utterly irrelevant/ when it comes to > having an impact on Fedora adoption. At best it’s a distraction from > actually increasing the market share of Linux — and at worst it’s a > validation of the argument that community driven projects fail due to > infighting." > > Rahul > -- > marketing mailing list > marketing@xxxxxxxxxxxxxxxxxxxxxxx > https://admin.fedoraproject.org/mailman/listinfo/marketing -- marketing mailing list marketing@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/marketing
