Potential impact of JSR-310

With JSR-310 recently accepted as a new JSR, I think it is safe to predict that this will form a significant component of what will eventually become Java 7. While it may not be the most exciting new standard -- at least in terms of bleeding edge technologies, it's impact will be felt heavily across many established a APIs. Personally, I would have loved have this effort run concurrently during my tenure as JDBC Specification Lead, particularly in the early stages of JDBC 4.0. The frequent mis-understandings in design, intent and general functionality of the JDBC java.sql.Date, Time and Timestamp always stood out high as some of the interfaces were long over due attention, especially considering the strained relationship between java.util.Date and java.sql.Date. Not to mention some of the inherent performance inefficiencies, that are rooted in the java.util.Date class, the general malaise was tolerated for too long. My current wish list for this JSR runs as follows...

• Disentangle the relationship between different setters on java.sql.Data,Timestamp and Time and their parent java.util classes while seriously considering a full refactoring of these interfaces.

• Date calculations - addition/subtraction/deltas
• Take a long hard look at the .NET Framework Date handling - they some nice things that the Java community would enjoy.
• Improved exception handling - I'd like to see some improved functionality for the new API to catch invalid dates

Finger-pointing aside, I must give credit to the co-specification leads, Stephen Colebourne and Michael Nascimento who I believe have an unenviable task. In the new spirit of Java openness they have also added a wiki where you voice your opinions directly. For some well informed opinions, I recommend checking out the Javaposse.com for some well thought out opinions on this topic. While undoubtedly they will be swamped with comments, we should all set out with the knowledge that it will be next to impossible to keep everyone happy. Nonetheless, I am willing to bet that this JSR will quickly reach the shrill levels of the JDO/EJB/Persistence debate at least in the next couple of months...