 Hello. My name is Noel Wilson. I work for Catapult Design. We're a design firm that offers design services to socially driven clients. Those clients can either be big development organisations, NGOs, for-profit enterprises, or even small entrepreneurs. And in 2012 we worked with a small entrepreneur called Wellow Water in India, specifically Rajasthan, to develop and redesign a water barrel that they were using to transport water around in rural villages. So we got to visit a lot of rural villages and do a lot of heavy contextual research of water use, of water sources, and also of different opinions of who should do the different water tasks, and really just uncovered as much as we could about water culture in Rajasthan. We spent a month doing that, but we spent the other two months of the engagement prototyping, and I wanted to explain some of the different levels of prototyping that we got into within that two months and the different values of them. So first and foremost we used their original product that we already identified the challenges of within the brief, and we were able to tweak that and see how we were able to achieve the different solutions that we were coming up with back then. And we kept sort of riffing on those solutions, and we hit a point where we had to start making a different device, and we were able to make some devices to test handle height, some devices to test capacity and weight, some devices to test size of the opening. And we first did them pretty low fidelity, and we did them with local fabricators, and we were able to work with welders, with benders, with tailors, with cobblers, and with a multitude of different crust people within the town, mostly in Udaipur. And that was very rough prototyping that it taught us a lot, and we were still able to make functional prototypes that people were able to fetch water with. And when we did that we'd learnt most of all that the capacity was going to be in the original prototype, and we cut it by a good third. So now this was under 50 kilograms, actually around 45 litres of water that we were able to transport. We also were able to work out that all of the water containers that were traditionally used had a pretty specific form and were able to be, one carried on the head, but two poured very easily. And so this was very low hanging fruit for us to incorporate that form into the already shrunken device that we had created. And we kept on with this process of prototyping with the people that were going to use these products or buy these products, and we were able to learn that storage was also a priority for them. So for this device to be able to sit well in the home and be used well in the home was another priority, and we were able to riff on that again and continue testing. And the low fidelity prototypes hit a wall and we had to start getting into more sophisticated form and were able to 3D print. We went through multiple rounds of 3D printing with a local mobile printer and were able to test different forms, and we did probably at least 20 different CAD models as well to test different forms with the users again and with distributors and with manufacturers. And when we were testing with manufacturers they were able to communicate the smarts and tricks of the process to get the most out of the process. We started with roto-molding and ended up producing a prototype for then-detected pilot which is boy-molding and which is able to be made a little quicker than the row-molding. And they first started making tens of these, then made some changes, made hundreds of them and piloted them and then made more changes. And now three years later, they're in the thousands but they're still prototyping them and the process has required a lot of patience for them. But they're definitely invested in the culture of prototyping. And that's really the model of the story is that prototyping will ultimately lead you down the path that you need to go but you don't know what that path is originally. So you have to explore many different paths and being able to explore and iterate is unconstrained by your budget.