I am a senior in college, in my last semester pursuing a BS in Computer Information Systems/Business Administration. During the course of this semester I am required to write a 15 - 20 page research paper and give a half hour presentation on it. I would like to research the future of computing, that is, whether or not open source will take over Microsoft's monopoly, if so when, and what things will need to take place in order for this to happen. What I have been looking for is some market share percentages, that is how much of the market is owned by Microsoft, by Linux, by Mac OS X, and by the other operating systems. My searches on Google have turned up evidence that IDC publishes reports of these market shares, and I would like to get a hold of the report or some form of it, but IDC sells it for $4,500. For one, is a piece of paper, or digital article worth $4,500? I do not think so. Second, I am a college student and as everyone knows, college students are poor and have no money.
What I am asking is, can anyone point me to any evidence of the operating system market share, credible sources/opinions on how that may grow in the coming years, or even how I might get the IDC report without forking over $4,500? If anyone can help, please do, and I would appreciate any help that you can give me.

Try Google Zeitgiest:

http://www.google.com/press/zeitgeist_nov03.html

It lists the operating systems used to access Google during the preceding year and has other info.

A couple of points to factor in when thinking about the Google "figures".........

What percentage of internet users worldwide hit Google in the period - different languages/search engine choice/cultures....

What percentage of computer users worldwide are internet enabled?

Related : Winmodems are another issue to consider - how many people worldwide are using/playing with Linux in some capacity, but are unable to get on the internet with a Winmodem due to softmodem driver/ISP config problems etc etc (or have temporarily gave up) .........

Yeah, as Skyline pointed out that would not be a very scientifically accurate measurement. Would it not also have to do with the number of times someone used google in a month? Maybe not if Google uses cookies and only counts a computer once.

Skyline or anyone else, do you know of any better sources?