 Good afternoon everyone. I'm Sudeesh. So I was a Google Summer of Code student with low-clack. Low-clack is a distributed peer-to-peer anonymous scraper and what we do is there are lots of servers which are distributed peers. So they connect to low, this is called the low-clack net. They connect or to the overall to printer servers across the world and they scrape the tweets one by one and they store them and they harvest them. So this is called the harvester. So what exactly is this? It's a backend search and which keeps data open. So what we know as of now is if you want to use Twitter, if you want to use Twitter, if you want to use Twitter data you need Twitter authentication, the OAuth, but low-clack allows you to do that without a password or without things like that. So it's completely open data. You can do a lot of things with that and there are endless possibilities. In this talk I'll be talking about a few of the things that we have done and what we could accomplish with this data. So first thing is it's anonymous. So you can directly run your own server. You can scrape Twitter data by yourself or you can use the existing low-clack.org. We have an amazing API for you to directly query. You'll get a JSON response and you can use that. So what are the functionalities that we have? There's a very quick search functionality because of the elastic search charts that we are running and it's running on elastic search. You can integrate Kibana, which is a front-end visualizing tool to it and as well as it's running on a Jetty server written in Java. There is very fast response time. So in case you post a tweet right now on Twitter, within the next one second or one and a half second, the harvester automatically takes that. It has peer-to-peer architecture and it takes up very less space. Why should one use low-clack? Because it's open data. We don't need a password in case you're very passionate about privacy. You don't want people to know what all the data you're taking. In case you want to use it for research, be it some kind of sentiment analysis or something to do with behavior, trying to analyze who's going to win the US elections, for example, you can try to do them. So how does it guarantee the anonymity? It does not record your IP address whatsoever, no matter what queries you make. There is complete control over what things you are going to log by yourself and what you want stored in case you're running your own server. And since, and any short links, so in case you have big.ly or any links like that, it automatically un-shortens them for you so that you can just hit it without any directs. So it has distributed peer-to-peer architecture. That means anywhere in the world, people can put that up. And Twitter also has something called, in case you request Twitter too many times, they block you out for some time, like one hour or one and a half hour with low-clack, that does not happen because of the distributed architecture. If one of the servers has the tweet, it will automatically start sharing the tweets within the servers. So what are the possibilities? Some of them are research, some of them are customer support and complain tracking, things like that. One of the applications that we have built using this large amount of data. So we have 170 million tweets that are indexed as of today. And one of the applications that we built was for the government of Tilangana in India. So using the concept of passive governance. So people generally do not want to go and complain to the government officers because they keep bouncing them front and back. But using Twitter, they randomly rant about it. All these rants could be used productively. So using low-clack, we are scraping all of that information, using the sentiment analysis in it and trying to give it to the government, telling that people over here have a problem and these are the kind of problems that they're facing. So it's a very productive tool. It can be used for bettering cities, making smarter and better cities for improving the amount of sentiment analysis research that you have existing as of now. And we also support lots of API. So we have API for Python, a very easy pip install Python low-clack API. API for Node, NPM install low-clack. And similarly for Ruby. So as I said, the government use case is just one of the use cases that we have. There is massive amounts of data visualization that can take place using low-clack. And it's something that's worth exploring. Thank you. So just one thing, in case you want to contribute to low-clack, you can head over to lowclack.org. It's very easy. You can fork and clone the repo, run the AMP build tool on it, and just write bin slash start.sh, which basically starts your server. And it runs on almost all the platforms, Linux, Mac OS, and Windows. Thank you very much.