 All right. Hi. My name is Arjun. I am from a team called Hackergram. We're a network of people and communities and organizations that do slightly subversive stuff on and off. But mostly we work with the people who generally don't have access to communication as a matter of course, not as a matter of government mandate. So, I mean, it's interesting. A lot of old familiar faces, Rams here, Kiran's here. So, these are people who I know have been working on access for a long, long time. But I think one of the great things, and I'd love to acknowledge that, is that this current frame that we are in right now has brought a lot of attention on to this issue of internet not being available ubiquitously and easily to people. Most of the places we work in, this is generally the case. We've done work in Karnataka, near Tumkur area, Devranthurga, 2G networks only, that too, on top of a hill, maybe once in a while. We've done work in Chhattisgarh. We've done work in now in Uttarakhand for the last three years. So, this state of internet not being available is something we live with all the time. And so, the kind of solution we came up to for this, and we didn't come up with it, it's an old, old mechanism, is the wireless mesh network. Anybody in here working with wireless meshes? So, wireless mesh is essentially your normal internet is, or your normal internet distribution is a hub-and-spoke model where one point has connectivity and then everybody else connects to that point as the gateway to get out. The alternative to that is a topology where everybody is connected to everybody else or multiple people are connected to multiple other people. Let's not say everybody is connected to everybody else. That almost never happens even in a deployed mesh. But there's more connections, there's more routes to get to a point than just the one route through the one gateway. And that essentially is the idea behind subverting a shutdown or subverting the lack of access or essentially having a self-healing network. So, a bit of history, we started doing work on this back in 2012. We've been working on community journalism with CGNet Swaraj. They had some mobile phone-based journalism platform and they were essentially using IVR systems to call in news reports from remote areas. The challenge there was that it was really costly to do this because you're paying a cell phone provider, airtime money, every single time you were recording a news report or the user was and that wasn't very cost effective. So, we thought of other ways to go about, you know, actually making this possible to broadcast at least at a hyper-local level without having to pay excess money for this. And that's where we started using standard 802.11 Wi-Fi protocol with very, very archaics now out of development. It's called commotion wireless. It was a state department piece of work that let you use dual radio Wi-Fi routers where one radio actually formed the infrastructure layer by talking to other routers and then one radio actually provided access points to everybody else. We've implemented this with commotion in a bunch of places. The challenge that we currently and back then also ended up running into was that the routers are extremely expensive. The current router that can support a firmware flash that will allow you to do Wi-Fi mesh networks, generally between five and eight grand, which is not really feasible for people to buy and host. So, generally what we ended up doing was paying for the hardware and then trying to find if there was a revenue model to kind of pay it back. In this case where internet shutdowns are the question, I feel like people could actually volunteer to set up these meshes by themselves. And so here the key challenge would be to one, find a new piece of software that actually does the job. And I believe Libre Router in Lyme does a pretty good job of it. We've been talking to folks in Latin America, the Rhizomatica guys, they've got an excellent piece of code there. It's called Libre Mesh. That's the one that we are now experimenting with. And the idea is really to have people host routers on their own rooftops and let the other people connect to them and then share their connectivity out. So in an urban context, you could potentially do this from a building top or a rooftop and build a network adjacent to the place that's got a blackout and then shuttle it out from there. The kind of data transfer strategy that we use is called the Sheep Pack Sync Mule Good Shepard model. It's a coin term by a gentleman called TB Dinesh who runs a lot in Bangalore. I'm Shukiran Mohsin. So the idea is that you have units like a mobile phone that basically acts like a pack that can go on a sheep, walk around, collect data and then you do a store in forward. So you've got a bunch of pictures on your phone. The moment you come into a Wi-Fi mesh range, you can start uploading it into a server or a Raspberry Pi that's hooked up to it. It can be as simple as a Raspberry Pi with a hard disk that just takes a cache of all the data you've collected. And then as and when you have bandwidth available, it keeps sharing it out to all the other peers on the network. This would have been a lot better with a little bit of diagram availability but I guess that's largely the architecture of it. I'd rather talk about this with questions coming in. So offline, yeah.