I've never really had a problem with the database connector packages in R, I just worked with them not knowing any better. Clearly they caused pain for someone, and after having read about this package I can see how it does make things easier. I'll be trialling it on my next project.
I'd be very interested in some performance measures. I have found that for larger datasets, it is quicker to hold as much as possible in memory, then use data.table for queries[1], rather than a complex SQL query for each return.
