>>The only thing you're adding to the query is a second SORT step, so it
>>shouldn't require any more time/memory than the query's first SORT
>>did.
>
>
> Interesting -- I wonder if it would be possible for the optimizer to
> detect this and avoid the redundant inner sort ... (/me muses to
> himself)
That's somethign I've wondered myself as well. Also - I wonder if the
optimiser could be made smart enough to push down the outer LIMIT and
OFFSET clauses into the subquery.
Chris