Linux Growth Hidden by Poor Statistics
Information Week reports on a USB survey of 60 CIOs that found a slow down in the adoption of Linux, but there are problems with the survey methods. First, 60 is too small of a number even if the CIOs were from different industries.
The UBS research is based on interviews with more than 60 CIOs in a range of businesses including financial services, energy and utilities, technology, and health care.
Then there is a problem with a bias in the survey group:
More than 90% of the responding CIOs in the survey who are not current Linux users said they would not deploy the open-source operating system in their servers in this calendar year.
as pointed out by Savio Rodrigues points out in InfoWorld:
Setting aside the Linux adoption question, we should consider the Linux usage & revenue question. The analyst doesn't appear to address the growing use of Linux in shops that already use Linux; the whole, 'try it, like it, use more' scenario. Also not accounted is whether the 10% who are going to adopt Linux will add one copy each or add 100 copies each.
Matthew Aslett summed it up nicely in the Open Source Weblog:
So those that have not adopted Linux are expected to continue to not do so, but there is room for considerable uptick amongst those that have."Statistics are like a drunk with a lampost: used more for support than illumination."
Sir Winston Churchill
Email This!
Digg it!
Sphere it!
Del.icio.us
Reddit!
Newsvine
