One of the biggest pain points in any IBM product are the non-sensical error messages that are generated by the products when something goes wrong. The exception itself usually doesn’t provide much of a clue as to the root cause of the issue. To proceed you either need to be an expert in reading stack traces, you open a PMR with IBM (a lengthy process) or you pop the exception message into google and hope that you hit something (like say the awesome danzrobok.com 😉 ).
What I’d like to see is IBM take control of this information about explaining in greater detail how an exception can occur and likely resolutions. I’d also take this one step farther and take this information from the infocenter (where it starts out as a shell) and populate it in a wiki format. The wiki would allow people who experienced the exception to describe how it occured, attach project interchange files and create discussions based around specific errors.
When resolutions are discovered, they can be updated directly into this central location for information about this exception, aiding future users. This solution would decrease the number of support calls that IBM recieves and it empowers the users of the product to resolve their own issues. Win – Win.