CloudServices/Location/Meetings/2013-08-01

« previous week | index | next week »

Agenda

Stumbler client

• dougt's Fantastic Voyage
  • Questions about data skew when stumbling at excessive speeds: issue #38
  • Other feedback?
• Stumbler Alpha test
  • List of Alpha issues: https://github.com/dougt/MozStumbler/issues?milestone=2&state=open
  • Top issues:
    • Issue #18 -- Allow user to enter a nickname
    • Issue #3 -- Keep local database of scanned APs to avoid reporting duplicates enhancement

Server

• Hanno's pretty leaderboard and map
• Hanno to replace map with a heat map: issue #8?
• Started breaking down incoming JSON blobs into tables, preparation for analysis and more detailed stats
• Alpha blocking issues?

Other

• Status of third-party data providers

Next Actions

• Plan work week for geolocation roadmap (i.e. invite Hanno to California :)
• Tentative Vancouver meeting with mmayo regarding UX
• Submit stumbler to Google Play Store (under a personal account, not Mozilla) so we can get more testers and easier updates.
• Determine how best to synchronize stumbler's three data streams: Wi-Fi scans, cell IDs, and GPS fixes.
• We need to analyze the quality of our data
  • We may want to remove the stumbler's 500 meter accuracy limit or increase it to 1000 meters.
  • Hanno is going to implement a search API so the stumbler can compare the devices' GPS location with the server's prediction.
  • Compare our stumbler data with other sources
  • Hanno says duplicate measurements for the same AP are not a problem.
  • We should increase the variety of devices used because different hardware will have different GPS capabilities, radio drivers and antennas.