"You are asking PG to return multiple unrelated data sets
Who talked about 'unrelated'?
Just as one can return min, max, average and standard deviation of column C in table T in one call, one can, for example, return that, the top ten records with highest C value and the records with the top ten most common values for C.
Yes, you can do two queries, but doing them in one go can be faster, some times much faster.
The risk, of course, is that one gets "one stored procedure per screen", but as long as one is aware of that, I don't see anything inherently wrong with that.
Who talked about 'unrelated'?
Just as one can return min, max, average and standard deviation of column C in table T in one call, one can, for example, return that, the top ten records with highest C value and the records with the top ten most common values for C.
Yes, you can do two queries, but doing them in one go can be faster, some times much faster.
The risk, of course, is that one gets "one stored procedure per screen", but as long as one is aware of that, I don't see anything inherently wrong with that.