Most people agree that climate change is happening. New reports, including the most recent UN one, tell depressing outlooks about our planet and our own lives if we don't act. Yet, we continue to almost ignore one of the biggest problems of our generation.
What it does
Enter Kommunity, a new place for environmental activism that starts with local communities and builds upon big data. With statistics and trends that relate to people and their local areas, environmental issues are brought up front and center, and all they need to do is start to take action. Luckily for them, Kommunity helps them do that too, with calls to action for events and activism based on problems in their area.
How I built it
I first focused on developing the client and filling in the API endpoints on the backend side when needed.
Challenges I ran into
Finding data and getting it into the client was one of the main challenges, as I had never used ArcGIS before. The next part was actually using that data to create trends and calls to action dynamically. Unfortunately I wasn't able to learn Tensorflow and Google Cloud ML in time after I finish most of the rest of the app, but I was able to show basic data and trends manually for now.
Accomplishments that I'm proud of
One of the main things I am proud of is the user experience. I meant for it to be a very natural, friction-free onboarding process where the user just performs actions and their profile evolves from there. I wanted continuity in the design through transitions and CSS helped a lot there.
Another thing I am happy about is how well the data from the maps came together.
What I learned
I learned a ton about trends I didn't even realize about environmental issues in my area, surprisingly, so I actually learned as a user of my own service! But with making the app, I learned a ton about geographical focused APIs and big data
What's next for Kommunity
You may notice that despite the web app's current focus on the environment, the slogan is "Bringing people together to do great things." And that generic-ness is entirely intentional. I believe that Kommunity can be a powerful platform for any great activity. Hopefully in the future Kommunity will be able to expand to deliver data driven activism to even more topics.
But working at that scale will also require a lot of platform development in terms of machine learning. I really wanted to add machine learning to this project, but unfortunately was not able to integrate into the application to show trends between environment and health data due to time constraints and application scope. So before Kommunity expands to support more topics, I want to add big data crunching through machine learning to see what interesting trends can be found from the power of Google ML.