The State of Linux Distributions: an analysis of Distrowatch rankings
Submitted by srlinuxx on Sun, 05/14/2006 - 10:41.
I have always been interested in Linux distributions and how they evolve, not just technically, but also in terms of their popularity and their -wanted or unwanted- position among the Linux users and on the broader market.
In this regard, the Distrowatch web site is a fairly effective tool at tracking linux distributions no matter how small or short-lived they are.
What I would like to do here is to give the interested readers a rather personal outlook on how the Linux distributions have evolved in terms of popularity, which are hot ones, why, and what I noticed behind the daily, weekly and monthly rankings of Distrowatch. I have been able to uncover some interesting patterns, and ones that are not often talked about.
What I found by looking at these rankings for several years now is rather interesting. To give a brief overview, there are distributions who do not move or not so much (say from one place up or down in one year) and they typically range from the 5th to the 10th positions. Now you may wonder what is going on from the 5th to the first place.
How DistroWatch ranking is manipulated ?
When tiny Linux distros appeared, we have studied DistroWatch ranking system. We have discovered the secret. Advertizement on DistroWatch helped Mepis and Ubuntu. But the hits can be influenced by the frequency of new releases. 8 weeks periodically gets improvement of standings. Puppy and FeatherLinux did well. Damn Small Linux works very well at DistroWatch. Austrumi improved ever so slowly because of timing spacing, but Byzantine(99th placed) got dropped because they did not announce their 08/04 release.
The secret of success on distroWatch maybe manipulated, after all. None of the distros in front are really bug free. And the bug free distros did not release their versions often enough to get noticed on ranking improvements. You have to advertize to get better ranking is the business aspect of popularity contest. New version releases are cheap advertizing. Planned infrequent releases will lose popularity.
PcLos can continue to have new releases against the current trend of less releases. All PcLos had to do is map the official Linux kernel versions and their driver inclusions. Make a list and provide the secret addresses of sockets and backports of hardware drivers. Then provide connectivity to stand alone application software to be able to use all the official drivers in kernels. Watch PcLos gaining popularity with each new release every 8 weeks.
Footnote:
DistroWatch does not allow one person to vote more than once in each of the poll of 100 candidates. But, it does not identify how many picks each person votes. Its eatirely posible that the total number of persons are only a few thousands, not enough samples to be better than 4% error. On the other hand, if PcLos has 2000+ registered users, each vote from them can put PcLos into first place. Feathers users did what they could before Feathers went dorment; but DistroWatch users are still curious enough to hit feathers' data.
Linux distro version number effects popularity ?
When Linux kernels change to ramdisk operating system, and distributed data is stored in local ramdisk instead of remote hdd repositories, kernel 2.6.x was born out of necessity. Any server software application had to take advantage of 2.6.x, but not desktop applications. So, generally Linux distro version 9.x is perfectly workable as desktops.
This however may mean that 2.6.x udev drivers still had to be installed on desktops, making version 10.x attractive to users. If your desktop version is 9.x, you have to upgrade to 10.x with all the drivers availble in kernel 2.6.x, namely all /usr/drivers(names). The kernel 2.4.x /dev/driver(names) can not be in conflict. And all drivers have to have a killer application program for each driver installable.
Gnome can drag your distro down the drain, because of a few programs are written with Mono codes of emulation. Once your program is emulation such as wine, you have to access them from terminal commanline input to bypass other emulation programs. Or else, emulation gui has to be stand alone. This avoiding Mono(except when programs of Mono are stand alone), will protect KDE which is an operating system of its own, base on Qt codes(gui based). As such it has its own drivers. These KDE drivers have to be upgaded with Linux kernel drivers, so that they don't crash on each other.
Whoever can streamline to version 11.x will be using Linux 2.8.x. adding 2.5.75 thread creation kernels and use prex(posix emulation) for console to run all servers offline or online(virtualization).
Good Luck on your ranking competition.
Footnote:
On version 10.x, Control center can adjust configuration files. But more sophisticated controls should be in application programs that can control the device in use. Printers need all the fonts and colors and formats which wordprocessor can control. TV card need channel change but antenna tuning on signal strength meter. Burner need multisession that can be plug into file managers. Skype needs answering machine features. Media players has scheduling features. The list goes on and on. We are just beginning to program sophistication into Linux applications.