Published Beaver photographer

It was still light after Non-linear Dynamics finished at 5!

Non-linear Dynamics was really interesting today. Besides all the usual theory, which is pretty absorbing in itself*, we got to have a go at making predictions: he'd brought in a board with many rows of nails hammered into it, down which he was going to drop a golf ball, which could then end up in various slots at the bottom. Apparently we were the first ST418 group to actually get to see the board- it's pretty heavy it seems but he'd had to have it with him to show to some conference of European weather forecasters in Reading that morning so it was no extra trouble to show it to us. I can only assume in previous years he's dropped the golf ball at home and revealed what the outcome was after everyone's made their predictions. Anyway, we were given three sheets of paper with information for three different starting points for dropping the golf ball about the percentage frequency with which several golf balls had landed in the various slots in an experiment carried out the previous evening, and odds, ie how many times your stake you would get if the ball landed in each compartment. The odds were designed to be fair given the probabilities of the ball landing in each slot, but actually there was a typo on one sheet with three slots being given 15% probability and corresponding odds when it should have been 10%. So he gave us the option of changing the odds for those slots to the odds for 10% probability or leaving them at those for 15%.

We had to pick which slot we thought the ball was most likely to land in, and had a hypothetical pound to bet- in units of a penny, so it was possible to bet on several or indeed all of the slots, dividing up the amount how we wanted. I divided mine in each case between all the slots with more than 10% probability, and put on each more pennies than their probability of coming up (this was possible because I was putting nothing on the ones with less than 10% probability). I was pretty pleased with this strategy because it gave me a 70 or 80 percent chance of making a profit- since the ball was only going to be dropped once it was thus pretty likely that I would come out of it better than I went in (if the ball had been dropped 100000 times then with fair odds however I had bet I believe I would have broken more or less even. Though maybe not. I didn't really have time to think about it properly). And indeed, I did make a profit on each go- the most I got was £1.25 (ie 25p profit) and the least £1.10. I thought that was quite good actually- if I'd bet £1000 that would be £100 to £250 profit which is really not bad. Then we did the same thing, this time with a rubber ball: this was supposed to represent a change from a situation that followed the known model exactly to 'Reality'. The lecturer was purposefully unclear about whether we should change whatever strategies we had been using; I decided to do a similar thing but bet on a few more low probability slots, therefore overbetting on the higher probability ones by less. In fact, I won again every time, but of course as I had bet less I won less (still not less than £1.10 though). For the record though the ball landed in high probability slots each time. Even though I won less, I was still pretty happy to have made a consistent profit. And I had always thought I was no good at things like gambling and investing and basically anything that involves risk taking, since I was always tempted to play it too easy (in this case there was a moment when I considered betting in pennies the same number as the percentage probability, till I realised that this gave me no chance whatsoever of making a profit and was thus rather pointless), and that the reason why I was tempted to play it too easy was that I would lose badly if I was brave enough to take a risk. So I was really pleased that I actually made a profit each time, and was even beginning to understand the attraction of the whole City and financial thing, which was something of a first, and wondering whether I could find any real life situation to employ this strategy in so I could make some real money out of it... Then at the end of the lecture they revealed a winner of the golf ball version, a winner of the rubber ball version, and an overall winner, none of which was me, incidentally, and though they didn't say how much people won I got the impression that just making the kind of profit I did actually counted as doing quite badly. I was talking to the teaching assistant at the end and he seemed to be saying the way to do well was to bet a lot on the less likely slots. But that doesn't entirely make sense to me as the ball actually landed in the most likely slots- except I suppose it did land in the 10% ones a couple of times. I suppose I could have bet just enough to cover the really likely slots and overbet as much as possible on the 10%s- since overbetting there would have had a greater effect if the ball went there than overbetting on the more likely ones if the ball went there (due to the greater odds). Anyway, I still don't entirely get it but apparently we're going to get another chance near the end of the course after we've learnt more.

They didn't really say if we did better than the European weather forecasters, sadly.

Yesterday was indeed hectic, though I did manage to make it to the Sutherland meeting. The SSLC went quicker than I had expected. Once again we had quite a few problems to note, including the Regression teacher's computer classes, where he pauses for literally (and I only use that word when I really mean literally) five seconds between telling us to have a go at a particular question and getting the teaching assistant to do it on his computer which projects onto the board. And he often goes off at a tangent and starts answering something a bit different. The problem is that, aside from the whole issue of not getting to try things for ourselves to see if we can do them before getting the answers, it's really hard to go as fast as someone typing commands into a computer when you have to look at what they're typing on the big screen in order to know what to type yourself (which slows you down compared to them) and when often after they hit return the output pushes their command off the screen. And it's really really hard to keep up with all that and try to actually understand what's going on as well. Sadly, though the comments had been passed onto the lecturer before the class he seemed to have decided not to take them on board. I hadn't noticed before LJ (I think it was) commented on it at the SSLC, but he always finishes at 6.15, so it's not like he has to go that fast to get through all the material, and he finished at that time again yesterday. As she says, it's not much good to us because when we finally get the chance to try it for ourselves he's done it all already. The really worrying thing is that I think he actually thought he was adjusting to suit us, since he was giving us about 10 seconds to do everything this time and making comments along the lines of 'Well, I'll give you a chance to try that yourselves'.

Things may improve however. As class representative, it probably should have been me who did this and I do feel a little guilty that it wasn't, but if it makes him change that's the important thing: N put her hand up and asked him if he could put the R history files (ie the saved list of all commands typed in during the session), cleaned up, in the Public Folders. This would make a big difference as, though we still wouldn't get a chance to try things for ourselves, we would at least not be lost if we didn't get everything down. I could really appreciate the point about asking for cleaned up versions though- we had printed version in a handout (not quite as useful as it sounds because he was going off at quite a tangent and most of what he did wasn't on there) and after I'd typed in the same command about three times alternating with a couple of different ones (following what was written down) I realised that these files included all the going wrong and mistakes that are a normal part of a session. Anyway, he said he would put the R history files in the folder, so here's hoping. I heard her talking to him later (after he'd stopped doing things on the board at 6.15) and they seemed to be working out a class strategy more tailored to our needs and wishes (I believe I caught a reference to giving us time to work things out by ourselves), and he really seemed to be serious about it- he was talking about a complete overhaul. I really feel that that should have been me talking to him, and that I was deserting my duty rather there. Especially because earlier when she asked for the R history files he was saying things like 'It's all stuff you've done before' and making her sound like a bit of an idiot and I should really have piped up and said I wanted them too and indeed if the SSLC was anything to go by the whole class did. I kind of feel I let her and everyone down. But at least it seems like there might be a change for the better.

The Beaver finally published one of my photos. It was small and muddy looking, and definitely not the one I would have said was the best I sent for that topic, but it was one of my photos. I was pretty happy about that- but then I later realised (when I was showing it to Ginger; we went to the NFT again last night) that apart from that photo by me all the other credited ones were by the photography editor himself, and I started to wonder if maybe he only printed mine so it wasn't an issue with only his photos. Still, there were a couple of uncredited ones I could swear were mine, and as they were uncredited it suggests they weren't there just to show that his weren't the only pictures included. They were uncredited because they didn't feature as part of news stories exactly- one of them was inside the front cover which seems to have been re-designed and formed a new thing called Reader of the Week (one of those ones where you come and get a prize if you are the person circled in the photo)- other new things inside the front cover included Picture of the Week which was credited to the photography editor- I'm not quite sure how that works: surely he didn't pick his own photo in which case who was judging. Anyway. The other was in among many photos of the Library, all uncredited (but I suspect mostly by the photography editor) which illustrated a diverse range of pieces from the serious to the surreal about said library. Both pictures were actually ones I sent in as candidates for the calendar the Beaver will be producing with the theme of LSE life, and had been cropped quite a bit. I was happy to have them published, and certainly if it was a choice between published uncredited and not publised at all I'd go with published, but it was slightly annoying that people reading it would attribute to me only the rather small boring and muddy one.

In other exciting news, the lecturer conducting the SSLC mentioned in passing that we should really be starting to revise last term's courses at least for the summer exams... This is scarily early for me (at UCL I would always plan to start revising in the Easter holidays, and the exams there were a month earlier than here- and I always started about two weeks later than I originally meant to), but I think I actually will start revising from next week: we don't have that much work this term (at the moment), but we will have projects at the end of term to do over Easter, and it's quite worrying how much work I need to do on a fair few courses! So please hold me to being able to write next week that I've got cracking on Principles and Methods...

*though the lecturer doesn't so far seem to have taken on board our comments in yesterday's Staff Student Liaison Committee and actually put all of the slides he uses in the lecture into the handouts, rather than about one in four (on a good day)- it would just about be ok if they were actually up long enough to get them down, but they mostly go so quickly; I think he has at least received the comments from the lecturer conducting the SSLC though as he made an aside at one point to the effect that the reason he didn't give us time to copy everything down is that he doesn't need us to have it all down. I still say he should just give us the slides on the handout sheets...