>>What's the main difference of those 2 SQLs? Am I right to say that the first one is faster than the second one?
>>
>>SELECT ELEM_ID FROM ELEM WHERE EXISTS (SELECT ELEM_ID FROM ANSWER WHERE RATER_ID=254 AND ANSWER.ELEM_ID=ELEM.ELEM_ID)
>>
>>SELECT ELEM.ELEM_ID FROM ELEM,ANSWER WHERE ANSWER.RATER_ID=254 AND ANSWER.ELEM_ID=ELEM.ELEM_ID AND ELEM.CHILDREN
>
>where is the elem.children in the first sql?

Your correct is should have been

SELECT ELEM_ID FROM ELEM WHERE CHILDREN AND EXISTS (SELECT ELEM_ID FROM ANSWER WHERE RATER_ID=254 AND ANSWER.ELEM_ID=ELEM.ELEM_ID)

>in theory (according to the books) the first one should be faster...

So, if the tables grow, the first one will always return the SQL immediately because it benefits of full Rushmore and the second one not?