FirefoxSummit/2006/ProposedSessions/StalkingElusiveFirefoxUsers: Difference between revisions
Jump to navigation
Jump to search
Line 33: | Line 33: | ||
* [[User:Asa|Asa Dotzler]] | * [[User:Asa|Asa Dotzler]] | ||
* David Rolnitzky |
Revision as of 20:28, 10 November 2006
Session Title
Stalking Elusive Firefox Users: How big is our herd?
Session Leader
K Lars Lohn
Summary
Figuring out how many active Firefox users there are is not as simple as just summing the numbers of unique ids we see in the server logs. Have our estimates been inflated by naive counting? How can we get an accurate count when our users, in protecting their own privacy, are armed to deceive us?
Agenda
- Review: our current technique
- the AUS cookie
- the formal definition of Active Users
- summing over a range of days
- the consequences of "doing the right thing"
- Research: mining 178G of server logs
- tracking individuals
- who are these obsessive cookie clearers?
- unique ids, average gap size, average run length
- where do we go from here?
- a formal definition for a new metric
- is this accurate?
Interested Attendees
Please add your name here if you're likely to attend this session, this will help prioritize sessions and minimize conflicts
- Asa Dotzler
- David Rolnitzky