#9499 HIGH Not Tri: Track XO use by children more clearly

[Zarro Boogs per Child]

#9499: Track XO use by children more clearly
--------------------------+-------------------------------------------------
 Reporter:  sj            |                 Owner:               
     Type:  task          |                Status:  new          
 Priority:  high          |             Milestone:  Not Triaged  
Component:  not assigned  |               Version:  not specified
 Keywords:                |           Next_action:  code         
 Verified:  0             |   Deployment_affected:               
Blockedby:                |              Blocking:               
--------------------------+-------------------------------------------------
 We could use the addition of a basic usage-tracking tool to the builds, to
 better visualize how XOs are being used, where and for what.  Needed:
 identification of a simple tool to incorporate, testing, discussing how to
 add to the builds.


 There are many ongoing research projects that try to track XO use after
 the fact.  To supplement this work it would be extremely useful to have
 more data on XO usage gathered automatically, subject to appropriate
 privacy controls such as aggregation, anonymization, and the like.

 This data which could be automatically aggregated and logged to the local
 machine, aggregated across a school and anonymized at school servers, and
 reported back to a tracking server.

 We currently gather a tiny set of data - a call-home that reports that an
 XO is online, what its ID is, and what OS version it is running.

 Additional data should be aggregated locally, anonymized along the way
 where possible,  visible to the user, and opt-out. It should be gathered
 only by school or larger group.

 Additional data to gather might include
 * # of hours in use per day
 * # of hours of use during school hours / at home
 * # of activities used, indication of frequency
 * frequency and speed of connectivity to the internet (spot checks)

 And more... a post to the research mailing list would help bring up other
 data that could reliably and automatically be gathered, current best
 practices for anonymization, and the like.

 This is one of those features that would be extraordinarily useful for all
 sorts of reuses - research, regional or school metrics,  self-assessment,
 identifying activity interest/proficiency by region, identifying
 priorities for localization, motivating developers, identifying school use
 or lock-up...

-- 
Ticket URL: <http://dev.laptop.org/ticket/9499>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system