Spark 1.6 introduced the next version of DataFrames namely the Dataset. As such, it's worth investigating support for it even though it is marked as @Experimental and the documentation clearly indicates it will be changed in the future.
Hence why this is a low priority task.