-
Suggests a topic you might be interested in.
-
Loading our request.
-
Getting a story from the community.
Inspiration
When we watch the news, we see political commentators battling about why some candidates are better qualified than others, or why some celebrities are better than the rest. The community's voice may only be heard from polls or the occasional quote. But we wanted to hear what the community was actually saying about topics of interest to us. We literally wanted to hear it.
What it does
We take in the user's preference of topic and form a story sourced from Twitter users.
How we built it
We used the Twitter tweepy module, markovify module and responsiveVoice API to gather, process, and output a summarized opinion on the user's topic. We used extensive javascript and python, alongside jquery and php for the backend. For the frontend we used HTML and CSS on an offsite https encrypted domain.
Challenges we ran into
We all lacked experience in web development. With that we had trouble incorporating responsiveVoice, and the tweepy module. Javascript being a synchronous (single-threaded) scripting language prevented us from doing some asynchronous tasks, such as being able to file i/o. We found a workaround by making an XHTML request instead. Scaling it also became challenging because if we had multiple users it could cause conflicts in output. We also had challenges in cleaning Twitter data for output.
Accomplishments that we're proud of
Making the design of the website very nice, simple and clean. Being able to get all the languages we were using to work together and click.
What we learned
Lots of web development. Twitter API. General experience with Web Dev. Flow of data in Javascript. Putting our app on the cloud and distributing it.
What's next for StoryTime
We want to integrate other backends such as reddit (already started), news article comments, quora comments, etc.
Log in or sign up for Devpost to join the conversation.