> which was tested on an MIT building rooftop. The system delivered pure water that exceeded city drinking water standards, at a rate of 5.78 liters per square

> which was tested on an MIT building rooftop. The system delivered pure water that exceeded city drinking water standards, at a rate of 5.78 liters per square meter I would be very interested to see data on this vs. sunlight and climate conditions, in what weeks/month of the year they tested it. I think its effectiveness would be highest at MIT's location from late April to end of September and considerably less in colder/overcast/less sunny weather and winter.