-
Notifications
You must be signed in to change notification settings - Fork 12
Home
This project is a part of the Data Science Working Group at Code for San Francisco. It was initiated in order to provide all Code for America brigades with a tool for performing user research with Twitter data. Other DSWG projects can be found at the main GitHub repo.
There is a genius in our ability to identify situations where people's lives can be improved by building new or combining existing technologies. Given the choice between making a small or large impact on the quality of life in our community, we will all choose the larger impact. Often the difference lies in how well we understand the problem. If we set out to engineer solutions based on untested assumptions or incomplete domain knowledge, we are gambling with our chances of making the biggest impact regardless of how much effort we make. The realm of UX Designers and Product Managers exists to assist software engineers with the user and customer data needed to validate problems and inform decisions among competing solution ideas. On many "Code for America" teams, volunteers with backgrounds in user research and customer development are missing, or the team might be so small that competing priorities makes sufficient due diligence impossible. This tool is designed to let all members of a project team-- from engineers to product managers to designers-- acquire exponentially more user research data than through conventional methods of surveys and interviews alone so they still stand a great chance to make a huge impact.
Code for America brigade volunteers come with a variety of backgrounds and experience levels, but generally have a very limited exposure to the full spectrum of technologies that might be applied to any given problem. Most would love to know more about machine learning (ML), APIs, and pipeline workflows as these technologies are powerful, in-demand, and currently dominate tech headlines. The intersection of interest and usefulness presents a great opportunity to save a number of teams from having to replicate a tool that all can use, so a key objective of this project is to build a tool that generalizes well to the widest range of project research needs. Our assumption is that most brigade projects could benefit from the same tool with only minor if any modifications. Our goal is to create an easy to understand and usable library with clear documentation so that any team member can quickly pick it up and start doing user research.
Envisioned as a pipeline of integrated components, we expect all research tasks will be able to use the exact same component for pulling data from Twitter (using their own API creds). Projects would choose additional components to add to the pipeline as needed to accomplish whatever task they're trying to accomplish.
- If you haven't joined the SF Brigade Slack, you can do that here.
- Our slack channel is
#nltweets - Feel free to contact @Josh Freivogel, @Vincent La, or @daniel.zou on slack with any questions or if you are interested in contributing!