@Jesse: You have to compute the numbers yourself. Many solutions were rejected because they hard-coded the numbers in their solutions. --Brad

Pragnesh... I agree with Jacob on this one. --Brad

Pragnesh... I suppose it is open to interpretation as to how one should center the Month/Year. In the example you gave, the Month/Year was 10 characters. Subtracting that from 29 leaves 19 characters to be divided by two. We decided to split it as 9+10

That should be changed. GO is not an actual T-SQL command... it's just a batch-end indicator/separator. It should never have a semi-colon after it. Jacob or Ramireddy, can you change that and get rid of the semicolon?

1) True, it is a single query. I misunderstood your original question, though. We want you to generate the fibo numbers from scratch as opposed to writing 92 UNIONed queries with the fibo numbers hard-coded. That's really the whole point of the exercise

@Misieq: Yes, the Fibonacci numbers have to be generated by you... in the same single query where you produce the final result. Your single query should generate the numbers, analyze them, and spit out the result. I'm not sure I fully understand your se

@Sergejack: If your solution happens to come out in the correct order, then you will be okay, but there are no guarantees. There really is no such thing as a "natural order" or an "order of item creation" in SQL Server... at least nothing that you ca

For this particular challenge, there can't be a "tricky" test or "load" test... simply because there is no data supplied... the query generates its own data from scratch. After checking the accuracy of the output, we'll just rank based on (CPU,Read,Durati

@Jeff: Perhaps "set-based" is not the correct term... that terminology may be from older times in trying to disqualify CURSOR-based solutions. We can change the guidelines to state the requirements more explicitly. Regarding your other comment: Ass

@Magoo: Please add a version suffix.