Just wanted to draw attention to a JIRA ticket that is getting some focus: https://issues.apache.org/jira/browse/TINKERPOP-1642
I'm hoping to see some pretty big improvements to the performance of the core API in 3.2.5 (i.e. speed up gremlin-core, not get benchmarks running faster as of a result of changes to TinkerGraph for example). So far the changes have been quite fruitful, but there's still work to be done. If you have any thoughts please feel free to bring them up on the ticket or here on this thread. Thanks, Stephen
