What does “Reference implementation” mean?

Question

I am trying to learn about CachedRowSet in Java, and I am reading some tutorials about it. I have come across reference implementation terminology very often.

For example

Alternatively, you can use the constructor from the CachedRowSet implementation of your JDBC driver. However, implementations of the RowSet interface will differ from the reference implementation. These implementations will have different names and constructors. For example, the Oracle JDBC driver's implementation of the CachedRowSet interface is named oracle.jdbc.rowset.OracleCachedRowSet.

Could anyone please be kind and explain this?

Note, I have found some definitions on the web that are too vague for me to understand.

Answer 1

It means the implementation of the specification that should demonstrate the concepts. Mostly it is implemented by the same guys/company that designed the specification.

You may think of specification as a standard that is needed to allow other possible implementation to be compatible with the rest of the world. And the reference implementation of a proof-of-concept kind piece of software that should show how to do that and encourage others to create their own implementations.

In the JDBC context it means, that there are some interfaces (CachedRowSet) prescribing some methods and there is a reference implementation to these interfaces done in Sun/Oracle.

Answer 2

JSR (Java Specification Request) - theory

RI (Reference Implementation) - concrete, working code proving that JSR can work. Basically, a sample implementation.

You can create your own implementation of any JSR and use RI as an example if you're not sure about something.