No matter how many times one confidently says, "It'll be fine," that doesn't make it true. Herewith, a case in point.
The new rubric means each of us has experience of all of the questions, surely a dubious decision, as it can't possibly help the average mark for the course.
Question 1 was an AP-style SQLj question, with several blanks, and two queries to be filled in. A baffling schema, wherein I am still unsure the point of the OWNER relation, and deceptively fiendish queries to be written made this an unpleasant introduction, for 9 marks.
Question 2 used the same schema, this time with JDBC, and possible allusions to transactions. Three queries and an update to be made, all wrapped in JDBC euchness, for 8 marks.
Question 3 was a baffling diagram, and a description of two clients that use the system based on that favourite schema of ours, from question 1. The question gave an impression of being broadly about transactions, though what we were quite expected to write under a heading of "Availability" is still beyond me. 8 marks for the spiel, here.
Question 4 was more like what we had been expecting, and appeared to be in the form of a standard Section A question from papers gone by. Questions on SQL (albeit no queries to write), three-valued logic, database constraints and Datalog. Each of which were pretty fair, though the question weighed disproportionately heavily - comprising 25 marks and half of the paper. Time was scarce in this paper, and putting the bulk of the paper didn't help particularly.
Overall, in my opinion, a physiologically and mentally draining experience. I'm off for a snooze....