On 19 May 2009, at 20:03, Tobias Ivarsson wrote: > Since you are planning to publish your code under AGPL on github, do > you mind me asking what the system you are building is doing?
I'm afraid I can't talk about the application yet (we're going to get a commercial license for Neo4j). I just want to release a general- purpose application template, to help others (and myself on future projects) hit the ground running. > I would like it if I could model each entity as one node, but if it > makes > things more manageable to model it as more than one node, then I > will do > that. Sounds like we're thinking along similar lines. > This sounds like a good way to model things. In my system it is more > common > to to change / add / remove the relationships than the entities. So > perhaps > I need to introduce a relationship indirection to be able to do the > same > thing for relationships in my system. ie let each relationship in > the model > be represented by two relationships with an intermediate node with > your > suggestion for history management. That sounds like a neat solution, good idea. So how would you implement an update to a relationship? Create a new intermediate node, delete the relationships from the old intermediate node to the two entity nodes, create new relationships from the new intermediate node to the two entity nodes, and create a relationship from the new to the old intermediate node? I guess any properties which you would have previously attached to the relationship (in a non-versioned data model) could now be attached to the intermedate node instead. I'm just wondering how the type of relationship should be represented. Should the two relationships to the intermediate node be of the type intended by the domain model? Or should they be a special relationship type (e.g. "versioned relationship") and the domain model relation type encoded as a property? Either way the graph traversers will have to be modified to take into account the intermediate node. I'm wondering if it's possible to write a wrapper around a traverser which operates on a graph of nodes with versions, but makes it look as though it was a simple unversioned graph (i.e. abstracts away the intermediate nodes and finds the latest version of each entity). It would be good to separate the versioning from the application logic. > My gut feeling is that you will make things complicated, with an > explosion > in the number of relationships for the target nodes, if you copy the > relationships. Yes, I had that fear too. Your solution sounds a lot better. > To the rest of you: all ideas are welcome, any input is good to get, > please chime in! Indeed. By the way, are there links to any publicly accessible apps built on Neo4j somewhere? Or are the production uses all hidden away inside big banks? ;-) Best wishes Martin -- Martin Kleppmann Managing Director, Ept Computing Ltd. Web: http://www.eptcomputing.com Blog: http://www.yes-no-cancel.co.uk Twitter: http://twitter.com/martinkl _______________________________________________ Neo mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
