The code is based on an experiment originally carried out by Liam Green-Hughes who adapted one of the off-the-shelf Google TV templates to load data using a SPARQL query. For more background on that see Liam's blog post:
Starting from the same template, this application builds a similar channel but using the Prelinger Archives data stored in Kasabi, rather than from the Open University.
The code performs a single SPARQL query to extract the basic metadata about each of the films described in the dataset, grouping them into categories for each decade.
The actual media files are stored on the Internet Archive servers and are streamed directly to the browser from there.
The code could easily be adapted to use additional, similarly constructed datasets. The code to crawl the Prelinger Archives collection on on the Internet Archives can be found at:
Leigh Dodds (email@example.com)
The application is a very simple Sinatra application. It's dependent on the Kasabi.rb ruby gem.
You need to configure your Kasabi API key in the KASABI_API_KEY environment variable before it will run. You should then be able to just:
You should also subscribe to the Prelinger Archives SPARQL endpoint on Kasabi, to ensure your API key is authorised to use the API.
This is currently known to run on:
- Google Chrome
- Chromium (need to install chromium-codecs-ffmpeg-extra package for video playback)