A lot of people got good to excellent grades in the oral exams in architecture. I don’t know why it always seems, on the first of the three days, that the professors are extremely critical and reluctant to give A’s. On the second, not only I got an A+, but also (naturally) my former and (not so naturally) my present programming partner. On the third (today) one of us got an A+ even though he had quite obviously not studied at all. The professors reportedly credited him with a lot of practical experience with the things covered in the lecture. So while he had not a lot to say, he got the grade.
Personally I did a little calculation and came to the comforting conclusion that, after yesterday, I can no longer get a B as my final grade. With only five more grades outstanding (one of which, the one for the thesis, counts double) and 20 already under my belt, the absolute worst I can do now is an A-. Even an average grade, on average, in all the remaining exams will net me a straight A on the diploma. That puts things in perspective just a little.
Fortunately, with some of the worst pressure off–the things we had had to study for the architecture exam were a lot and extremely varied–I slept well last night and went to the process mining exam this morning–the first written one–in a decidedly more relaxed mood. Unfortunately, however, this exam turned out to be another case of a new professor, or in this case, an old professor with a new field of interest / lecture subject, having no good idea what can be expected of students under exam conditions, and how much of it. 90 minutes and seven assignments, five of which were the usual algorithms: lengthy and error-prone at the best of times. And an exam isn’t really the best of times for things that invariably start with honey-combing several sequences of letters for the number of times the various letters follow one another, and if, then also vice versa, or not. You try to do that right, it takes forever, even if your hands are not shaking. You don’t, you might as well not do it at all, because the outcome after several steps, each depending on the results of the previous, can’t be but seriously off.
And these were the five easy assignments. The other two involved modeling traffic poriority on a crossroads with a P/T net–and modelling notoriously takes forever–and describing, in extended prose, the general idea of process mining for a university administration that has a performance problem. We got 30 minutes more, but in the end most people had still been unable to complete all the assignments. I had, but much later in the day it suddenly occurred to me that I had committed a serious conceptual error in one of the algorithms that will certainly invalidate half of it. So our best hope now is, again, that the professor will grade kindly. In particular, we do hope she will generously award partial points if we drew the right conclusions from earlier wrong results.
Still, one more down. My practicum partner and I spent an hour in the cafeteria just talking of unrelated things, to calm down, and then three hours to review all the lecture slides and our cribs for the distributed systems exam tomorrow. Amazingly, doing this together, systematically, again revealed several points where our understanding was still somewhat incomplete, and at least one where I had missed a key statement on a slide.
So tomorrow will be the final day of the shortest and most intense exam period ever. In fact, I am just now getting into exam mood–that unique constructive-fatalistic state of mind in which you can be amazingly efficient without breaking a sweat, and make tough decisions (like to leave something alone) in the blink of an eye, even though you care greatly about the outcome. Normally it’s the second exam and all subsequent ones that benefit from this stance. Now with the entire exam period rolled into three days, it’s sadly only the last.