Reply to post: Re: Same old, same old

US data suggests Windows 10 adoption in business is slowing

Richard Plinston

Re: Same old, same old

> unlikely to represent a random sampling of the population

Many of the usage stats are gathered using javascript linked to from self-selected site. There is nothing random about which sites use particular stat services, there is nothing random about which users visit which sites, there is nothing random about which users block javascript and/or monitoring sites.

It may be useful to look at which users of a particular OS upgrade [sic] to a new version because those users are likely to continue to visit the same sites and use the same add-ons. It is unlikely to give any meaningful information about different OS usage because those users will visit different sites and may block in different ways.

For example: my machines will not be in _any_ of those stats.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon