windows 8 market share for October, 2014

It appears Net Applications has made some changes in the way it collects data and has eliminated some data it was collecting from China that was almost 100% XP. This gave Windows 8 a boost while XP took a dive in the numbers from prior months.

Windows XP's October market share tumble was overstated, NetApplications admits- The Inquirer

Edit; I have to wonder as well if those trying out Windows 10 are reported as Windows 8.x users

 

Again, why am I not surprised...remove data that counters your claim of more W8 but then get caught and say ops we left some out...yea right sounds more like oh let's remove these bad numbers so W8 looks better....doesn't seem to fly very well I see.

 

The issue is that the data was posted without notice of the change. Because of this a lot of sites were proclaiming a huge come back for Windows 8.x by comparing the numbers of the prior month. Accidental or not the left themselves open to criticism.

 

True but the issue here is you can no longer use any of the prior data now to compare current data, not unless they go back from the saved data and remove those other groups from there as well. Apparently it was not a statistical anomaly but actual intended and directed data manipulation.

By not disclaiming they have also given fuel to the conspiracy people. Thankfully the marketing people at M$ did not bite for if it did and started a marketing campaign using these numbers to try and bolster sales and/or proclaim Windows 10 a marketing winner, well an ensuing disaster would be putting it mildly.

 

What makes them believe that the current numbers are "more accurate?" This really brings into question the accuracy of all of their data. They are just picking and choosing sites to sample from....

 

Agreed, almost any and all data can and should be taken with a grain of salt as they say. I have learned long ago you can make numbers say just about anything you want. This is why you have to place careful scrutiny to all of it. Even then findings can be way off as while someone sees the glass half full others out there see the glass half empty and draining out quickly.

 

Really is that the best mustard reply I get? Maybe you been in the cave to long but if you remember M$ themselves called testers and preview testers DUMB....and I didn't hear you make a peep about that and how they were treated? Malicious Intent so whom do you think did it to themselves...not the testers.

 

Mustard? MS calling people dumb?

 

Must been in the cave alot here...it's already in many news stories how bad M$ treated testers and IT telling them what they are doing was bad. Then to name call those testers dumb....now if you can't get your head around that...then nothing will.

 

Enlighten me.

 

TBH, while never specifically called "Dumb" by M$ actions speak many volumes louder than words. We need not rehash those here as it is way off topic.

Back on topic, now that this can no longer be used to compare one iteration to another for performance, is this intentional? We no longer can say, at least with this metric, how well Windows 10 is doing over time in relation to XP/Vista/7/8.x.