This project is based on the thought experiment in Dbd, but using different implementation aspects:

• Drop the Context as an intrinsic storage item of a Fact (context can still be added, using regular facts)
• emphasize the streaming/"Event Sourcing" aspect of the fact stream (particularly for the windowed caching option); this also refers to the ideas published on distributed joins for linked data
• Use Scala/Spark/GraphX as a core technology (this is required as the windowed caching/Event Sourcing will trigger a massive amount of data, so a horizontally scalable graph solution is required)