Earlier in September, we did a great hackathon day at La Nacion, courteosy of fellow OpenNews fellow Gabriella Rodriguez, (@gaba). Gaba’s the lead on a project called CrowData (which was originated by fellow OpenNews fellow Manuel Aristaran (aka @manuelaristaran)) which facilitates source management at the paper by inviting readers to submit PDFs to be automatically processed. I really like the tool, and decided to help out by creating a Dockerfile for automating server creation.
I felt I contributed a lot to the project because, as I set up my environment to use the CrowData platform, I debugged a lot of issues particular to a virgin linux machine. Since the purpose of the Dockerfile is to generate a successful build of the platform on a virgin machine, my environment was perfect as a sandbox. For instance, CrowData currently requires python, Django, and Postgres; along with their dependencies. Other than python (natch), I don’t have any of those requirements on my machine. So my Dockerfile reflects how a machine might do a headless install when each requirement is initiated.
