From distrowatch
The (presumed) rapid increase of Linux Mint's popularity continued to receive attention in many online publications throughout the week. One of the authors commenting on the issue was Bruce Byfield of Datamation who believes that, based on user traffic and key design changes, it is not entirely inconceivable that Mint is indeed becoming an Ubuntu killer: "At first glance, the idea is absurd. Given that MGSE modifies the GNOME 3.2 release, you might convincingly speculate that Linux Mint has provided the solution for the many who are unhappy with GNOME's current directions. But challenge Ubuntu? Canonical, Ubuntu's commercial arm, claims twenty million users, and is promoting the distribution heavily. By contrast, Linux Mint is a much smaller, non-commercial organization that appears to be less organized, and to have fewer resources to draw upon. In fact, it relies on donations and ingenuity for funding. Yet is the idea even technically possible? Certainly Linux Mint's team and its supporters think so, considering that for several years they have been calling Linux Mint the fourth most widely used operating system, which sounds like a deliberate challenge to Ubuntu's claim to be the third. One way or the other, a closer look seems in order."
I think that possibly the two pregnant sentences are:
a) ubuntu claims 20 million users
b) mMint claims fourth largest user base.
So to speculate one would have to actually SEE the data for the number of actual downloads of Mint.
Lit us be kind and say that it is approximates an exponential curve in the beginning.
If it is the fourth largest we can only speculate that say, for the sake of argument:
ubuntu 20 million
distro 1, 15 million
distro 2 ten million
distro 3 5 million which is Mint.
That means that to knock ubu off the top seat Mint would have to be having downloads that would at least approach 15 milion because to be the "largest" it would have to have taken away at least ten million from ubuntu proper and then be above the fifteen million of distro 1.
To do that the arc now would have to be, for the sake of argument,
downloads of let us say a quarter million per week.
That would supposedly accelerate as more and more people switch from ubu proper to mint.
so given that we might say, a quarter million in week one, then .50 million in week two, .75 million week three, 1 million week four,
and that is just a linear extrapplation so given a linear extrapolation that would take 15 months to pass the distro 1
if it were exponential it would be on the order of say.... 8 months to pass the fifteen million mark.
let us say that the distrowatch ranking reflects this.
That means that the numbers at distrowatch would probably go through the same arc.
Thus one could take the ranking at distrowatch and do the same kind of math for it, and if thenumbers increased in the same linear fashion or ecponential fashion then one could speculate, with validity, that possibly Mint is the giant killer.
Given that the distrowatch numbers are ANYWHERE near being remotely related to the real numbers.
http://distrowatch.com/weekly.php?issue=20111205#news
It really is hard on one's brain being a geek.
woodsmoke
The (presumed) rapid increase of Linux Mint's popularity continued to receive attention in many online publications throughout the week. One of the authors commenting on the issue was Bruce Byfield of Datamation who believes that, based on user traffic and key design changes, it is not entirely inconceivable that Mint is indeed becoming an Ubuntu killer: "At first glance, the idea is absurd. Given that MGSE modifies the GNOME 3.2 release, you might convincingly speculate that Linux Mint has provided the solution for the many who are unhappy with GNOME's current directions. But challenge Ubuntu? Canonical, Ubuntu's commercial arm, claims twenty million users, and is promoting the distribution heavily. By contrast, Linux Mint is a much smaller, non-commercial organization that appears to be less organized, and to have fewer resources to draw upon. In fact, it relies on donations and ingenuity for funding. Yet is the idea even technically possible? Certainly Linux Mint's team and its supporters think so, considering that for several years they have been calling Linux Mint the fourth most widely used operating system, which sounds like a deliberate challenge to Ubuntu's claim to be the third. One way or the other, a closer look seems in order."
a) ubuntu claims 20 million users
b) mMint claims fourth largest user base.
So to speculate one would have to actually SEE the data for the number of actual downloads of Mint.
Lit us be kind and say that it is approximates an exponential curve in the beginning.
If it is the fourth largest we can only speculate that say, for the sake of argument:
ubuntu 20 million
distro 1, 15 million
distro 2 ten million
distro 3 5 million which is Mint.
That means that to knock ubu off the top seat Mint would have to be having downloads that would at least approach 15 milion because to be the "largest" it would have to have taken away at least ten million from ubuntu proper and then be above the fifteen million of distro 1.
To do that the arc now would have to be, for the sake of argument,
downloads of let us say a quarter million per week.
That would supposedly accelerate as more and more people switch from ubu proper to mint.
so given that we might say, a quarter million in week one, then .50 million in week two, .75 million week three, 1 million week four,
and that is just a linear extrapplation so given a linear extrapolation that would take 15 months to pass the distro 1
if it were exponential it would be on the order of say.... 8 months to pass the fifteen million mark.
let us say that the distrowatch ranking reflects this.
That means that the numbers at distrowatch would probably go through the same arc.
Thus one could take the ranking at distrowatch and do the same kind of math for it, and if thenumbers increased in the same linear fashion or ecponential fashion then one could speculate, with validity, that possibly Mint is the giant killer.
Given that the distrowatch numbers are ANYWHERE near being remotely related to the real numbers.
http://distrowatch.com/weekly.php?issue=20111205#news
It really is hard on one's brain being a geek.
woodsmoke