popularity-contest should be enabled by default
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
popularity-contest (Ubuntu) |
Won't Fix
|
Wishlist
|
Unassigned |
Bug Description
Binary package hint: popularity-contest
The entire purpose of the popularity-contest package is to collect statistics. In order for statistics to be useful, they need to come from a significant sample size representative of a population. Currently the sample is neither significant nor representative, as it is a relatively few people who have self-selected themselves to be in that group. The results therefore are likely to be heavily skewed towards more technical users, individuals versus group deployments, and people from wealthier economic backgrounds that have time to fritter away looking at popcon data out of curiosity (in addition to the developers with a legitimate interest).
In short, unless popcon is enabled by default, the statistics from it will continue to be invalid and of very little use, or worse, substantially misleading. Since the stated purpose of those statistics is to guide future Ubuntu development, that is a serious problem.
Now, some people may raise "privacy" concerns with this sort of thing. To address that, we need to make sure that it is clear that the data is submitted anonymously, and that every effort is made to keep it so. (eg. the server shouldn't keep IP address logs at all) As long as there is no way of ever connecting the data to an individual user or machine, there should be no problem with the data existing for the community as a whole.
Changed in popularity-contest (Ubuntu): | |
importance: | Undecided → Wishlist |
Status changed to 'Confirmed' because the bug affects multiple users.