[anigbrowl]

See, that looks cheap to me. DK claims 1000 liters/day for a cost of 0.2 cents/liter; let's be pessimistic and assume 500. (Maintenance would be my biggest worry, but let's keep it simple.)

Now if we take grinding poverty as a per capita income of $1/day and consider a village of 100 people, that means everyone in the village could have 5 liters of pure water a day at a cost of 1 day per week - a 14% 'tax' for 1 year.

That's a big economic hit for a very poor village, but it would not surprise me to find that such a village might lose more than 14% of its productivity to a combination of acquiring cleanish water and lost productivity due to water-borne diseases.

$5200 is not big bucks to us in the developed world. Surely it would not cost too much to buy 5 of them and do a year long comparison of 10 villages where half of them have the system and the other half serve as a control. Even if you add in setup costs the total bill for such an experiment would only be $75k.