Hi, As discussed at our last meetup, we should really document our public APIs. Many had requested this be completed before Hive 1.0.0.
As such I have created an uber jira to track this https://issues.apache.org/jira/browse/HIVE-9362 and created two sample sub-tasks for how I imagine this will play out. The most important task is documenting what we consider to be the public API. Additionally even if the API is not public, our user community, e.g. SparkSQL, Pig, etc, should document the API's they use. This way we'll at least know who we are breaking when we make a change. Cheers, Brock
