More times than not publicity, given open source software by major news/information outlets, proves a mixed blessing. Part of the reason is that we've reached an inflection point of awareness as it relates to open source software. The topic continues to be mentioned enough such that I can talk about the basics with my mom and have her follow along quite well. However, there is nothing near a solid understanding of open source that can remotely qualify as pervasive. When combined with the fact that strongly held bold opinions, i.e. flamebait, tends to attract more readers than in-depth, factual analysis, the result is the proliferation of articles like the one from ZDNet Asia which "reports" that there is a Linux identity crisis.
Since I actually commend Don Reisinger's effort to tackle the complex subject of categorizing the inherently distributed and diverse Linux community, I won't spend much time debunking or critiquing his views. Rather, I thought it would be better if I put forward some facts to keep in mind about the subject at hand, for those who are already aware of the following, bear with me:
- From a generic user perspective Linux still suffers from a lack of drivers and consumer software, e.g. games, media players when compared to Windows and Mac OS X: a substantial hurdle yet to be properly addressed and subsequently cleared.
- The average user doesn't go shopping for an operating system like Vista or XP they buy computers. A great many don't even understand what an operating system is...and for the most part they don't need to invest time finding out, as long as their computer works.
- There is a difference between the Linux kernel that Linus Torvalds oversees and a given distribution. The wonderful thing about the kernel being open source is that it is at the heart of advanced-user oriented distros like Gentoo and Slackware where the user can configure everything for him/herself and Ubuntu, complete with its "Linux for human beings" mantra.
- It is entirely possible to maintain the strength and integrity of the core Linux kernel while also improving the usability of the userspace tools and distributions. As a matter of fact, that is exactly what's happening right now. Parties which promote ease of use in the desktop environment aren't advocating that the kernel be dumbed down, and members of that "faction" (myself included) are in no way mutually exclusive. Ubuntu has found a way to encapsulate the intended functionality and potential of the Linux with a relatively easy to use and aesthetically appealing operating system.
- The majority of patches and suggestions sent upstream have more to do with latency/tasking operations present with desktop uses. Adjustments made in light of those issues do nothing to reduce efficiency as a server kernel, nor to sacrifice capabilities. This is partly why I disagree with calls for separate desktop and server kernels.
- Optimization for the mainstream user doesn't point to the irrelevance of other approaches. I look at open source Linux as a quasi society with a variety of citizens, one which must live with the realities of diversity enabled by openness. To strive towards the homogenization of Linux for the sake of producing a Linux distribution that can fare well in the OS market as a real viable alternative for consumers, is baseless.
- For the mainstream user the desktop experience centers on the GUI/windowing experience. Granted, the IO and process schedulers do still need to be tweaked but it isn't that unusual for distros to maintain their own set of patches, anyway.
- Linus Torvalds is, indeed, protective of the Linux kernel. Yet future considerations must be made towards making improvements to cater to how people are actually using computers in order to keep the value proposition of a widely-used kernel that's free, alive. If not, I expect an increasing number of Linux distros will go the way of the OS/2.
- If there is a "liberal" element of the Linux community it is most likely supportive things like extensive improvements to the userspace tools as fuel for propelling further penetration and recognition of open source software, a sustained push on hardware manufacturers to provide compliant drivers or open their specs.
- The Linux kernel community is still strong and resilient. Where the divisions and fissures being reported are because the development process is visible, as opposed to closed off inside the walls of a proprietary company. When Linus makes an announcement on a mailing list it can be linked to, commented on and analyzed by anyone with an internet connection. Debates are a healthy sign that more than one side is being considered and most likely these differences will lead to mutually beneficial solutions.
- AMD/ATI's stated commitment to better meeting the needs of the open source community and Dell's desktop Ubuntu play, marks a point where reluctance to play nice with Linux is slowly becoming non-viable from a business perspective.
- Total annihilation is far from likely, even in the event of a "civil war." In a worst case scenario there's a fork where some distros offer a Con Kolivas patched version of the mainline kernel.
In closing, I can't agree that Linux still isn't "about the money," seeing how the kernel is the recipient of a healthy level of financial contribution, driven not by ideological support but by the fact that growing portions of the commercial sector are stakeholders in its continuance and success. Additionally, whether or not Linux distributions are found at BestBuy in 2-3 years isn't entirely indicative of whether it is, or ever was, meant to be relevant from a mainstream perspective. In the same way that future mainstream momentum isn't a sign of having lost touch with its roots because at the end of the day, Linux is just Linux.