richard is a video index site. It has a very basic admin interface for adding videos by hand one-by-one. This gets very tedious when adding all the videos for a conference.
steve is a command line utility for downloading information for a conference, downloading all the metadata for the videos, and making it easier to transform the data, fix it and make it better. Then steve grew into a library of functions making it easier to manipulate video metadata and push it into richard.
richard and steve go together like peanut butter and jelly. You could use one without the other, but it’s daft. steve uses richard’s API for pulling/pushing video data.
It solves this use case:
MAL sends Will a request to add the EuroPython 2011 videos to pyvideo.org. The EuroPython 2011 videos are on YouTube. Will uses steve to download all the data for the conference on YouTube, then uses steve to apply some transforms on the data, then uses steve to edit each video individually and finally uses steve to push all the data (the new conference, new videos, speakers, tags) to pyvideo.org.
I’ve been working on richard since March 2012. I knew I needed steve and had some thoughts on how it should work, but there were a bunch of things I wanted to do with richard, so I pushed work on steve off.
Then on May 29th, 2012, I finished up the initial bits of steve and thus steve was born.
I worked on it on and off while simultaneously working on adding the EuroPython 2011 conference videos to pyvideo.org and helping Carl with some of the conferences he was working on.
Then I continued working on it until I thought it was at the point where someone else could use it to generate conference metadata. That point happened on January 13th, 2013 when I released version 0.1.