![]() ![]() Now, all that remains is to add up the weighted sum and divide by that number…Īnd the answer is 1.18% for GNU/Linux’ global share of page-views. ![]() That’s almost exactly what StatCounter puts out for their “unweighted” number of 1.17%. This data is for all operating systems, not just desktop.Ī surprising result from my calculations is that GNU/Linux flattened around 2010/2011. That could be the result of M$ killing the netbook and along with it a bunch of new GNU/Linux PCs consumers could use or StatCounter rejigging their stats or my weights for ~2013 not being valid back then. Still GNU/Linux is not dead and will rise again. There’s a lot of room to grow and with M$ being unable to count the opportunities continue.ĭrLoser wrote, ” Even random numbers would be better than no weighting…I suspect that we can both agree that you mis-spoke on that point.” The baseline has doubled in 5 years despite a lot of competition from Android/Linux. overcounting A and undercounting B, a random re-sampling can in part remove the bias. If you randomly fire while shaking the barrel a bit, you may hit the bull. If you don’t shake you will never hit the bull. I presume you have heard the term “circular definition,†Robert? If not, look it up.” A poll of ~1K randomly-selected people may well tell the story better than selecting 100K people from Winnipeg.ĭrLoser also wrote, ” If you add up random 1s and 0s you get a normal distribution. ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |