SciChat

Inspiration

The scientific community grew to communicate research among its peers but outreach to the general public is still sparse. This is mainly due to the complex nature of modern scientific findings which makes them difficult to break down to a simple level. Alongside this, opportunities to have a dialogue with experts is often limited. We think that everyone benefits from the outcome of scientific studies and therefore a quick and simple way of communicating the results needs to be created. Through a live chat, the general public is able to learn about modern research and ask any questions they have. This also helps scientists to communicate current findings in an understandable fashion.

What it does

SciChat works by connecting members of the public with scientists in the areas they are interested in. By searching keywords, the users are matched with corresponding experts and taken to a live chat where they can discuss the topic of interest in a conversational manner.

How we built it

We developed SciChat as a javascript-based web application without any dependency on a back-end. For our app we needed a database to store our users base as well as a user-authentication system. We used Firebase for that purpose which provides a fully functional API with all backend resources SciChat requires. The chat client module of our app has been implemented using the PeerJS library, which provides a rich API for achieving efficient WebRTC p2p chat communication through a mediator server provided by PeerJS.

Challenges we ran into

We’d like to display recommended papers (relevant to the selected area of interest) to the user if they are waiting to be connected to another member, however we had some challenges using the APIs of scientific journals, and so plan to implement this feature in future.

Accomplishments that we're proud of

Creating a functioning prototype of our idea within the timeframe of 24 hours. Implementing a chat interface that pairs users based on similar interests/research fields via tags.

What we learned

We can do a lot of stuff in 24 hours! The biologists in the group learned to code in html, javascript and the basic structure of creating a website.

What's next for SciChat

To refine the current website, we need to improve upon usability as well as implement numerous extra features we did not have time to integrate. In addition to public-to-academic communication, we hope to add a scientist-to-scientist communication to the platform. Scientists would be able to display a profile including information about their research along with links to publications and professional websites. We would also like to add a reputation system where academics can rate the engagement of the user they chatted with, and vice versa. This will allow for a more sophisticated pairing algorithm in the future. We also hope this will further incentivise the use of the platform. There are several ways in which data generated may be used, such as understanding trending topics and creating user statistics (average rating, number of chats)