Visualising twitter data helps us inferring many things about the nature of public opinion on issues at hand. The network is the foremost information we get from visual forms, but the most important is the nature of discourse. There are many data analytical techniques that allow us standardised methods of text mining. Such as sentiment analysis and its spatial attributes allow painting spaces with newer information. Visualisation can be geographical or modular depending on the availaibility of the geo-coding of the tweets. This is contrained by the privacy rights of the twitter user, who might choose to provide accurate, inaccurate location of his tweet environment or not to provide at all any information. The problem is also of fake user accounts and the bots, who are easily manipulated with algorithm to disseminate as much number of tweets as possible. These challenges are usually a focus of the IT professionals who would like to sift the data by-passing all these smoke screens. However, as a social scientist we and not bothered by these deceptions if we organise our query carefully based on our empirical understanding of the topic we want to explore and seek only a parametric form of information. These are not very jargonised methods in form of computing algorithm, but the simple coding techniques that have to complied wisely to get the desired nature of queries we want to push through the heap of twitter data. The tweets on any particular topic can amount from few thousands to hundreds of thousands within a day's time. It depends on the nature of the topic, such breaking of a political scandal or a conflict between two great powers. I am using these example because most of my study of tweets is around the themes of geopolitics. As a student of geopolitics I have tried to make a case for cyber-geopolitics that is deployed through social media, such as, the Twitter.