The nograph project is my attempt to make working with various NoSQL and GraphDBs easier. Plenty of other projects and tutorials exist to help you with your favorite implementation. I wanted something different. It needed to be easier than JDBC but not as heavy as hibernate. Seeing how GraphDBs are usually reduced to nodes and relationships, and these in turn can be reduced to documents, the overlap between these two approaches becomes obvious. In fact GraphDBs are now commonly listed as a type of NoSQL. One common breakdown has four elements.
My favorite indexing technology is Lucene. When I started seeing others write about how Lucene was one of the earliest NoSQL data stores, I knew I wanted to build a GraphDB on top of Lucene. It turns out that others have been doing that for a while. (List other Lucene graphs including those using SOLR and RDF).
Even though others are doing this, I still wanted something easier (at least to me) to use for storing nodes and relationships. The nograph API allows for easy insertion of and searching for nodes and relationships. Traversals are still a bit of a work in progress. However, the NoSQL aspects make it even easier to use Lucene as a document store. Simply set properties on the generic node implementation and the API will create a document and add it to the lucene index. No schema needed, no field definitions or mappings. Set some properties and save.