yes, that's certainly something worth adding :). Here, the fact that the "latest" data is duplicated in the versions and "current" tables would help a lot... Though checking all data in a database could take a long time. But providing an interface for this should be doable. One problem is how not to miss a revision (if there are two or more changes before the batch process fires).
Additionally, the interface could support receiving notifications from other services, which could contain information which rows changed (for example the IDs of that rows, or even the whole row data).
If you'd like, create a JIRA for that :)