tomtom on the Neo1973

Paul McMillan Paul at SemiCubic.com
Mon Apr 2 21:58:07 CEST 2007


Something else worth mentioning, regarding road usage, is that on a much
larger scale, we could get fairly accurate road data automatically if every
phone recorded where it went (unless the functionality was turned off) and
just uploaded that data automatically to a server when it was near an access
point. Combine large amounts of path data with velocity information, and
you'd have a pretty decent set of tracks that were likely to be roads. You'd
probably have a pretty accurate representation of areas where many NEO users
went regularly.

It's probably too large of a project for this community in the foreseeable
future, but it would be useful with enough users.

If you wanted to take it a step further, and make it more useful in the
interim, you could combine speed, time, and location information to generate
predictive traffic models[1]. I think this may be an extremely useful
application of the phone, aside from road mapping altogether. Obviously this
data would have the potential to be anonymized for anyone to be interested
in participating. I've done some prototype work in this area already,
analyzing gps tracks from standalone devices and averaging and overlaying
the tracks over google maps (simplest way to usefully display the data) to
see if the concept is useful.

-Paul

[1] Also, these tracks could be used to predict where you were likely to be
going, and provide an estimate of how long it would take. This functionality
would require no maps at all, or could make use of incomplete free map sets.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.openmoko.org/pipermail/community/attachments/20070402/fb3a8fe7/attachment.htm 


More information about the community mailing list