Supporters like to claim that one of the advantages of FOSS is that it encourages diversity. Unlike Windows, FOSS is supposed to welcome new ideas and to be less vulnerable to viruses because most categories of software include several applications.
The reality is somewhat different. Examine a user poll, and you find a consistent pattern in which one application or technology has 50-65 percent of the votes, and the next one, 15-30 percent.
For example, among distributions, Debian, Linux Mint, and Ubuntu, all of which use the .DEB package format, won 58 percent of the votes in the 2012 Linux Journal's Reader Choice Awards, compared to 16 percent for Fedora, openSUSE, and CentOS, which use the.RPM format.
There were only two exceptions to this general pattern. The first was Best Desktop Environment category, where the diversification of the last year was reflected in KDE receiving 26 percent, GNOME 3 22 percent, GNOME 2 15 percent, and Xfce 12 percent. The second was Best Web Browser, in which Mozilla Firefox received 50 percent and Chromium 40 percent.
Overall, the numbers fall short of a monopoly, but in most categories, the tendency is there. The best that can be said is that, without the profit motive, being less popular does not mean that an app will disappear. But if competition is healthy, as everyone likes to say, there is some cause for concern. When you look closely, FOSS is not nearly as diverse as it is assumed to be.
By 2004, FOSS had reached the point where people could do all of their consumer tasks, such as email and web browsing, and most of their productivity computing using FOSS. If you ignore the hopes for a free Bios, only wireless and 3-D drivers were needed to realize the dream of a completely free and open source computer system.
Nine years later, many of the free wireless drivers and some of the free graphic drivers are available—but far from all. Yet the Free Software Foundation only periodically mentions what needs to be done, and the Linux Foundation almost never does, even though it sponsors the OpenPrinting database, which lists which printers have Linux drivers. Given the combined resources of Linux's corporate users, the final steps could probably be taken in a matter of months, yet no one makes this a priority.
Granted, some companies may be concerned about so-called intellectual property in the hardware they manufacture. Perhaps, too, no one wants to reverse engineer for fear of upsetting their business partners. Yet the impression remains that the current state of affairs exists because it is good enough, and too few care to reach the goals that thousands have made their lives' work.
A few people might be aware of some of these taboo subjects already. Probably, however, there is something in this list to peeve everyone.
However, my intent is not to start nine separate flame wars. I'd have no time for them even if I wanted them.
Instead, these represent my best effort to identify the places where what is widely believed in the community needs to be questioned. I could be wrong—after all, I am discussing what I have grown used to thinking, too—but at worst, the list is a start.
If there are any other taboo subjects that you think that the FOSS community needs to consider, leave a comment. I'd be interested in seeing what I might have missed.