Linked by Thom Holwerda on Tue 24th Feb 2009 13:42 UTC
Apple Whenever we're talking market share and Macs, it'll inevitably get late. There are different means of measuring market share, and different ways to interpret the resulting data, usually leading to heated debates about who is right and who isn't. Ars decided to take a look at the different methods of measurement and see what they mean.
Permalink for comment 350398
To read all comments associated with this story, please click here.
Net applications sucks
by averycfay on Tue 24th Feb 2009 14:16 UTC
Member since:

Even if they are not good for measuring market share itself, you can still use them to investigate where the market is going: sure, Linux may only have 0.5% according to these figures, but if a year ago this was stuck at 0.25, that still indicates a growing trend.

Not really, at least if we're talking about Linux. The numbers change way too much month-to-month to put any kind of weight in them even if we only care about trends.

Check out Linux's share between September and October of 2008. You can't seriously believe that ~20% of Linux users suddenly switched operating systems over a one month period. Much more likely they changed which websites they were tracking (or how much weight they were putting behind certain websites) complete throwing off the numbers. That might more accurately reflect reality if you only care about absolute market share, but it makes tracking trends completely worthless. It's like switching your procedures half-way through a scientific experiment. It completely invalidates the results.

Reply Score: 2