Hi Marko, Thank you very much for the tips, indeed, we will follow this route as well to guarantee type safety without causing a negative impact on performance.
Romiko -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Marko Rodriguez Sent: Monday, 20 June 2011 11:54 PM To: Neo4j user discussions Cc: Jamal Abreu ([email protected]); Tatham Oddie Subject: Re: [Neo4j] Category Nodes and Type Safety vs Performance and Scaling Hi, > Notice, here we now have a relationship where from the name we can see that > the types coming back will always be an Agency. This means relationship names > have information about it's target Node type, making the amount of > relationships proportional to the number of node types. I tend to do the same thing when I don't want to deal with filtering when I get to the end of a traversal. For example: marko---followsPerson-->peter marko---followsFeed-->horicky In the above, I don't have: marko---follows-->peter marko---follows-->horicky because, then when I traverse, I don't have to do this: marko.out('follows')[[type:'Feed']] Instead, I can be safe knowing that marko.out('followsFeed') will always get me to feeds and thus, no filter needed. Also, the latter plays well with Frames (http://frames.tinkerpop.com) which is what I use for graph-object mapping. Hope that helps, Marko. http://markorodriguez.com _______________________________________________ Neo4j mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user _______________________________________________ Neo4j mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user
