"Community fiber" is working towards a practical implementation of ambient connectivity that Bob Frankston's been talking up for years through IEEE and elsewhere. Some of his essays lay some good groundwork for governance and expansion. http://frankston.com/Public/
With several miles of elevated highway, a moderate volume of volunteered & OEM supplied vehicles and open minded infrastructure partners a pretty broad range of hacks/trials could be accommodated. Actively driving vehicles off the road and/or damaging them would be frowned on. However, sudden avoidance of "soft" obstacles under different scenarios definitely works.
Not sure how to work in the scoring aspect of a flying v, but would definitely be fun to model fuel savings of cars doing the same.
My work on smart homes led to what we're doing w/Carvoyant. How about connecting your car? I'm a little biased towards "lock the door, arm the alarm" but it doesn't have to be all functional. We've got lights at office that flash when someone speeds (recently unplugged by an unknown offender) but that could also be flashing patterns based on who's coming home next.
Personally, I'd love to see Raspberry Pi & Ninja Blocks in that mix and would be happy to foster a few more connected car/smart home mashups.
One nice thing about a connected car and a good rules engine in the home, is that our "things" can take independent action like dim the lights, adjust temp and lock up on their own, all the time. More timely adjustments can maximize the savings and reduce user intervention.
Even without the car, I'd really consider any design options/devices that can reduce or eliminate your intervention. The homes not so "smart" if you're always issuing commands.
It's nice to see a good balance between useful App and useful API. I used this feature previously but now I can easily point friends and the rest of the team to it. Data portability alone is worth paying for and the API flexibility lets the available features grow faster than Fullcontact's efforts on their own. Kudos on good work for the front end of identity.
Whoa, already down from 500 problems per 1000 vehicles (last March) to 400 in 1000 now, and pushing hard to 360 in 1000? I'm definitely curious the structure of their metrics What would that performance translate to if iOS or Android had similar "problem" levels?