“With India’s coronavirus crisis becoming increasingly more desperate and beds, medicine, and oxygen supplies scarce, people in cities across the country are relying on Twitter and the kindness of strangers for help during a time of national upheaval.”
The above extract is from an article by the Washington Post about how the COVID crisis in India is growing out of proportion resulting in a shortage of resources and communication gaps between the suppliers and the people in need. All of this is resulting in people tweeting asking about the availability of oxygen cylinders and other medical care resources in their area.
The problem is that many of these tweets don't get 100+ retweets or likes and are hence not being noticed by people who have resources or leads.
Many of these tweets are being ignored for hours and hours without getting even a single reply. This is demoralizing for the person who posted the tweet and is directly or indirectly hurting the person who the resources are being asked for.
While there are many people who are helping these people for free, that is still not enough compared to the tweets people are making every minute. There is a need for another incentive for more and more people to help them and using methods to make it competitive so they come back for more.
There were many problems that we ran into. While we were still ideating the project, one of our biggest concerns was the verification system. You can read the Notion document that we’ve attached in the links to understand the full product more clearly. The verification system is very essential as we need it to protect the system against scammers and spammers. For this, we now have 2 tests and 3 ways to prevent spam. You can read more in the Notion document. Thinking about the business plan was also quite tough for us, as we have little experience in commerce but we believe we were successful.
Ideating the individual database models was also quite challenging. As we were using a NoSQL database i.e. MongoDB, we had to make sure that the models are optimized for faster database reads and the system was foolproof. You can check out the models in the repository’s directory
Using the Twitter API was also quite a task as the docs are not very clear but we were able to pass that fairly easily.