Linked by Thom Holwerda on Mon 2nd Nov 2009 23:59 UTC
Windows 7 has been out and about for little over a week now, and as it turns out, Microsoft's new baby is doing relatively well. That is, according to the figures by NetApplications: Windows 7 already reached the 3% mark this weekend, and is already closing in on the 4% mark.
Thread beginning with comment 392865
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[4]: Linux has reached the 90% market share
by boldingd on Wed 4th Nov 2009 22:05
in reply to "RE[3]: Linux has reached the 90% market share"
Member since:2009-02-19
To say nothing about the question at hand, size alone is not a sufficient guarantor of the representativeness of a sample. Even very large samples can fail to be representative of the population from which they are drawn; equally, smaller samples can be more representative if carefully constructed. Larger is probably better, but it is not a sufficient condition to guarantee representativeness on its own. So, "it was big!" is not a valid answer to someone who questions whether the sample population was representative; big can still be wrong.
RE[5]: Linux has reached the 90% market share
by strcpy on Thu 5th Nov 2009 05:17
in reply to "RE[4]: Linux has reached the 90% market share"
Member since:2009-05-20
To say nothing about the question at hand, size alone is not a sufficient guarantor of the representativeness of a sample. Even very large samples can fail to be representative of the population from which they are drawn; equally, smaller samples can be more representative if carefully constructed. Larger is probably better, but it is not a sufficient condition to guarantee representativeness on its own. So, "it was big!" is not a valid answer to someone who questions whether the sample population was representative; big can still be wrong.
Fair enough. But in this case I believe the sample may be representative enough. Nothing to support this but IMO.
(But as you probably know, "being big enough" guarantees certain statistical properties via the central limit theorems.)
The point here was that things like "faked identification strings in browsers" hardly matter in the overall picture.
Recent Original Stories
Recent Comments
Headlines
Random Comments
Random Stories
Random OS Link
Member since:
2009-05-20
Really?? So if my web browser identification string does not show the OS, my count is invalid (or do they then simply assume Windows). If my browser identification string is "faked" so that some idiotic website can work and not block me, that count is incorrect as well.
Very accurate - NOT!
You should recap the elementary statistics a little.
With big enough sample, browsers that do not show identification are just white nose.