India & OpenStreetMap

I'm excited, given I'm off to India for my second time this year. This time, it's not going to be cracking 45 degrees or more at a time of year I typically consider winter.

As I've been thinking about the upcoming travel, I've been working on some of the Humanitarian OpenStreetMap tasks (http://tasks.hotosm.org/job/274) that affected the region shortly after my last visit.

Interestingly, the India I saw must truly be only among the 5% of western facing, tech focused parts of the economy. It's certainly reflected in the amount the country that is mapped.

While Australia's coverage is sparse at times, the idea that a small town or village might be completely outside of anyone's awareness is hard to comprehend. When you look at some of the regions impacted by the flooding earlier in the year in India, you begin to realise that major roads are completely absent. These aren't small service roads or residential roads, but 10, 20 or 50km long stretches which seem to be the only motor vehicle access to some areas.

It would be interesting to know if the Indian military has better maps; and if they share them with NGOs focused on aid relief when natural disasters are pending.

Homeless data collectors

Here in SA, we have a container deposit scheme.
Homeless folk often are able to add a little revenue to their lives by collecting the 10c per can reward.

This benfits all involved, keeping areas clean, waste down and providing a limited sort of charity.

Given the low costs of digital cameras, gps, wifi and ardunio boards, is there a hidden opportunity for businesses?
For example, geocoding and address matching is an unsolved problem - we have data sets but they are not open, or can grow stale quickly.
Grabbing a photo of a location is possible from streetview, but again it is not an open data set.

Could rugged gps enabled cameras, with a 5c per photo reward, be a good solution?
Uploading and tagging against open street map, automatically?

Given a limited amount of review, this seems a feasible alternative to ocr solutions, with a real world benefit to people in need.

For the homeless involved, the fee is less than a can or bottle, time to take photo a bit more from location to location, but easier to carry.

Obviously there would be a discomfort factor for some home owners, but I would wager most folks already harvesting recycables deal with this already.

What other data could the homeless collect, given the right infrastructure?