The global multi-year exercise kicked off on today when the www.osscensus.org went live.
The census takers figure the results should surprise just about everybody and that when companies realize how dependent they already are on open source the floodgates will open wider.
How authoritative such numbers turn out to be depends on such uncertainties as the size of the sample, how the numbers are messaged and what they’re said to prove.
It will be up to IT departments to run an open source tool called OSS Discovery and pick which systems to scan for open source code and send the result back to the Open Source Census database anonymously. Once they’ve contributed they get access to reports that summarize their own open source use and provide comparisons to similar companies.
Depending on the size of a participating company, OpenLogic is suggesting a scan of anywhere from 100 to 1,000 machines.
The web site will list the number of times each project and versions has been installed according to the results provided.
Sponsors, who are supposed to help make the catch-as-catch-can sample meaningful, include IDC, CollabNet, Unisys, Olliance Group, Open Solutions Alliance, Open Source Business Alliance and O’Reilly Media.
Advisors include Tony Wasserman, director of the Software Management Program at Carnegie Mellon West, and Jim Jagielski, chairman of the Apache Foundation.
IDC says it’s supposed help ensure that the data is interpreted consistently and will combine it with other IDC research for a more complete picture of open source adoption.