Earlier quoted context omitted.
Nope its actually a totally real and meaningful measure. it measures how much of the market market -> all the people encompasses your share share -> using your stuff Apple has shown time and again that it isn't a good measure of Profitability, but your statement seemed to imply a more general connotation that I can't help with disagree with.
Market share != Usage share If more people are buying product A, but product B has a longer useful life, then it conceivable for product B to have a higher usage share. The problem is that usage share is a lot harder to measure than market share. But ultimately it's the more important number when it comes to things like network effect and developer and consumer mindshare.
80 million OSX (any version) users world wide in 2012 http://www.cultofmac.com/172693/mac-os-x-by-the-numbers-60-m...
110 million windows 8(just windows 8 not windows in general) users in 2013 http://winsupersite.com/windows-8/there-are-now-over-110-mil...
believe it or not their are 500 million windows xp users still http://www.businessinsider.com/microsoft-to-cut-windows-xp-2...
Market share actually does in some respect account for a sum of existing products in use, how that is calculated will depend on the reporting entity.
Usage share turns out to be super easy to capture these days, we do it by looking at the browser string of a request to a web server. Most big sites are putting out this type of data so a relatively clear picture emerges.
For a company like microsoft or apple market share is a more useful measure because it helps them plan/predict their supply chain so that they have the right amount of inventory in the right place at the right time.
But from a developer perspective usage share is more interesting for obvious reasons.
In this case the numbers are still basically the same.