SDShare is a simple protocol that allows clients to stay in sync with data changes on a server. It can be used to integrate RDF data from multiple sources, where some can be legacy sources like SQL databases. It's published as an open specification for anyone to use and implement, and there are a couple of open source and commercial tools out there.
The spec has a large usage area. It could, for example, be used to meet 5 out of 11 use cases identified by the W3C Linked Data Platform group: www.w3.org/2012/ldp/wiki/Use_Cases_And_Requirements
We've been implementing this in commercial projects for a couple of years now, and would like to do a talk about the protocol itself and our experiences with using it.