Well, today was the “big day”. I say that because quite a few people impressed upon me that graphics would be the hardest of all of the Computer Science exams this year (most notably Paul). Quite how true that is, I don’t think I could really say.
So I’m now three-quarters of the way through my exam timetable for these end-of-year exams. So far I have sat algorithms and data structures, operating systems and graphics. With just one exam left (coincidentally I have been told this is likely to be the easiest exam)—databases—I am feeling much happier than a few days/weeks ago.
So far I feel reasonably happy/confident about graphics and operating systems. However, I have that worrying feeling I have failed the algorithms and data structures exam quite miserably. I wouldn’t mind too much, but I did fail one exam in January (oddly enough I think it was networks—but that was possibly due to stupid questions). If I have, in fact, failed the algorithms exam that could mean the university will want me back in the last week of August to re-sit that paper, as well as the paper I failed in January. Let’s just say this is hardly an appealing thought considering: 1) I would otherwise be in China; 2) re-arranging my flight would cost me at least £50 (although I would save money once I returned to England (maybe this isn’t true—I’d visit more people in the Hull area, which could be expensive)); and 3) re-sitting exams is never fun.
This year, like last year, I’ve taken a fairly lax attitude towards lectures and examples classes. Too lax? Well… I’m not really sure, actually. On the whole I believe that my knowledge of the subject before going into each exam is probably the same as it would have been had I gone to “more” lectures. If not the same, then only marginally less. I maintain that had I spent, say, a week “revising” (learning) for each subject I could have absolutely aced every paper and question. Unfortunately, this wasn’t quite the case. On the whole my learning structure has been: read through and take notes on the lecture notes during the latter half of the day before the day before the exam (yikes!); make more detailed notes on the day before the exam, try and cover topics that didn’t make a massive amount of sense the day before; most importantly—visit Samia, Ceyda and Karen in the evening (around 6PM) so we can sit together and pool our knowledge to find satisfactory answers to the past paper questions.
Without doubt this is the best revision structure I’ve ever had. I have yet to find out how well I’ll do, it probably won’t be that great, but if I were to extend it a little I can see it going far. So maybe the only thing I did wrong with the algorithms exam was to not spend as much time going through the past papers. The brilliant thing about working together on the answers is just simply that each of us are better/worse at different areas so we can explain it to each other. I hazard to guess that sitting down like this in a small group is massively more productive than being bored half to death in a lecture theatre… but then again, what would I know?
Having just had a look at ARCADE (the Computer Science marks database) I was surprised to find that my highest lab mark is in algorithms at 74.9%. My OS and graphics marks are about 55% (I actually dropped as few as five marks across both of these subjects; unfortunately I have a nasty habit of forgetting to get extensions (or at least leaving before they are handed out)) while my databases mark is a lowly 35.5%—partly due to the fact that I was quite reluctant to sit in front of a keyboard like a chicken and insert data into tables (something not helped by the fact that the Oracle sqlplus command has no command history, which requires the full re-entry of data if a single mistake is made in the previous attempt!). The lab marks only contribute 10% towards the final course mark (I wonder why I bother, really)—maybe, just maybe, the seemingly decent 75% in algorithms might just bump up my exam mark enough to give me a pass, or at least very close. Here’s to hoping!
Anyway, the graphics exam was fairly straightforward: the first section was multiple guess. Of all the multiple guess papers I’ve ever sat this was likely the easiest of them all, with just two or maybe three questions where the answer was not immediately obvious. So, with any luck I’ll have got close to full marks with that section (let’s say so far I have 37% (almost 100% in section A plus my lab mark)—I need just 3% more to “pass”). Section B was equally straightforward—I opted to answer the question on compression, which covered lossy and lossless compression, details of GIF’s lossless compression (including RLE), and JPEG compression (fairly straightforward for a 10-mark question; chances are I messed a few little bits up and got eight out of ten). The only place were I didn’t feel quite so homely was section C, thanks largely to the fact that when I tried to convince Samia and Ceyda that we should go though exactly the same question because I was sure it would come up they went for a shower and some food. I forgot to go through it myself :(
After the exam there was the promise of watching Sin City at theprintworks, as well as getting some much-required grub after just one meal the previous day (and no breakfast!). Unfortunately this was (for me, at least) not to be. More than a little upsetting as I was particularly looking forward to seeing this film having very pointedly not downloaded it and watched it at home. I guess I’ll just have to wait a little longer.
Finally, I’ve imported some more photos. “By my calculations,” I have just my China photos left, and I should get those done by Sunday, at the latest. I’ve re-evaluated which photos to include this time and I picked them by a rather different strategy. I call this strategy “include as many photos as possible, discarding only duplicates and really crappy or boring shots”. Needless to say there are quite a few previously “unseen” (unless you saw my own full-size archive) photos now.
All the images are sorted into hierarchical galleries, along the lines of: Places->China->Beijing->Badaling. This works out nice enough but I’m still not sure if I’m entirely happy with it. Navigating quickly to a given picture can be a bit of a pain but at least I don’t have a single page with a zillion different galleries as I did last time. I might juggle this around a bit later, if I can come up with a better solution.
Otherwise, the only thing left for me to sort out is larger preview images. As a result of using the Kubrick theme on my site, I have to limit the maximum width of previews to a rather less-than-generous size. This works just fine for portrait shots as they still have a lot more height but it leaves landscape shots looking tiny. I guess it doesn’t help that about 90% of my photos are landscape, either. As I see it there are three solutions to this: 1) have small previews; 2) switch to a variable width theme; 3) try and hack Kubrick into a variable width theme. I think I’ll try those in reverse order. Jolly good… time for some CSS hacking for me!
