 When considering how to make research data interoperable, there's a number of concerns that we need to make sure that we are taking care of, that the data is first loadable into the software that you've got, that when it's been loaded it's usable, you understand the data structures that you're looking at, that when you're looking at the data structures you can actually understand the terms and the numbers that are included in there. They're not just rather strange abbreviations or strings that the units of measure are clear. We are providing here a sort of graduated way of assessing how interoperable your data is with some pointers about how to make it better in each of these concerns, loadable, usable and comprehensible, with the ultimate being that everything's linked out to community-based or standard definitions, schemas, formats, field names, so that the data is understandable and usable and interoperable to anyone within your community. So in this talk we also have a practical example looking at the RL dataset and looking how we can make this more interoperable, so both using our Oslo N5 star tool to guide us as to how to make it interoperable and then implementing it and showing how we can use reference definitions to make it more interoperable. So I'll be far enough to snapshot.