Data interoperability is the process that allows the sharing of data between different organisations or researchers.
Central to this is the premise that different types of data have commonalities and different data schemas can actually describe the same thing.
The process of mapping these common data types between different schemas, allows data to be shared easily and reused. For example, suppose dataset A and dataset B contains a series of dates and some corresponding data that was collected on those dates (this may be interview data or data from lab machines).
Someone wanting to link these two datasets and seek out an associational link, must first check that the language used to describe the dates was the same and then usually carry out manual rework of the data to match the datasets together.
More often than not, the labelling and format of the date ranges will be different, so interoperability will seek to specifiy a mapping between the dates in datasets A and B. Typically using a commonly used schema such as the Dublin Core. This will then enable a machine or other researcher, to use the data with the minimum of additional work.