Quarter Life Crisis

The world according to Sven-S. Porst

« iTunes stuffMainA trip to the south »

Antrittsvorlesungen

1547 words on

This is about the highly interesting topics that are university and maths. So run away now if you think you can’t handle that.

The first thing that comes to mind with regard of that topic is money. More precisely the lack thereof. That’s quite a notorious feature, particularly in departments like mathematics where there’s no industry support and shiny graphs or diseases for publishing are rare. This means that many things happening in the department can be a bit depressing or at least nerve-wracking.

Thus it was nice to have a positive event for a change. Becoming a professor is quite a feat. Those are life-time posts after all. And thus, following old traditions, every professor is asked to give a public lecture when he or she starts the job. This is at a fairly general level as the idea is that the public or people from other departments can come as well and learn what the new professor is working on. It’s also one of the rare opportunities for the department to show off what it does and for the new professor to give everyone the impression that it’s actually important.

And then there’s the second thing that seems quite characteristic for universities and math departments: everything happens rather slowly. And if it doesn’t need to be done immediately, it may be postponed over and over again. Which is exactly what can and does happen to those public lecture, or Antrittsvorlesungen. In fact for the maths department we had ended up with three new professors who hadn’t given their talks yet. One who had started last year in the numerics institute, one who started in 2003 in our (i.e. pure maths) institute and one who had started in 2002 in the statistics institute.

And apparently our dean came up with the good idea to make all three of these talks happen at the same time in a nice old assembly hall in one of the old buildings. Thus, giving the whole thing a festive character and making sure that people from all the different institutes actually come and see each other once in a while. And while the attendance could’ve been higher and had an almost complete lack of young students, it went rather well.

After a welcome by our dean (who isn’t much of a public speaker and often forgets to greet people and so on, but who did really well today) and by some vice chancellor (who seemed mostly to be impressed by himself and whose two minute talk and leaving early during a talk nobody would have missed) we got to hear some piano music which was also played in between the talks (Schumann, Schubert and Debussy, IIRC). Our junior professor played. Rather well I thought.

And then, the first lecture was given, by ‘our’ professor. His post is probably the most prestigious in our department, with people tracing it back to Gauß when they’re in a good mood. And having him around has been quite good for the department as besides being a good mathematician he’s quite good at many other things, from negotiating for things like more library money to getting people over to give talks. He’s also been at most famous universities from Moscow to Japan – and that’s going the long way via Paris and Princeton, so he know quite a few people and things.

His talk focused on a simple topic – integers. Well, actually it was called ‘Rational Points’ and about algebraic geometry but there are relations between that and number theory as people who followed the whole ‘Fermat’ thing may know. So while he may not be working with just integers most of the time, they were always lingering around and everything can be followed back to the most simple mathematics. If you’re generous at least.

His idea for the talk was quite good. With some more refinement it could easily be really good. It still had some passages that left me with the impression that non-mathematicians would lack the background to see the point. Some more delicate maneuvering could have helped there. And perhaps he should’ve actually practiced it with an audience beforehand. Then they might have noticed that he should have talked a bit more loudly and that pointing at the screen of your computer doesn’t really help the audience when all they see is the projected image from a video beamer (do people actually use this expression in English or is it another English word people invented over here?). That was quite funny…

His slides were refreshingly simple, mostly black text on white background. The images he used were a bit too small and low-contrast and the slides were done in PDF, apparently the mathematicians’ favourite computer slideshow format.

Next up was the statistics guy. He’s doing statistics for medical research which I know nothing about. I learned some statistics as an undergrad but never got the hang of it. But as those talks were for the general public this shouldn’t be a problem. It wasn’t. But I didn’t learn too much from the talk either. Somehow what stuck is that with a lot of theory they tested a new drug in comparison to an existing one and compared the survival rates of the patients, ending up with a graph for each drug, with one being ‘above’ the other one all the time.

Now, forgetting everything I ever knew about confidence intervals and so on, I was tempted to say that, well, the drug where more people survive is better, and to ask where all the fancy statistics are needed. I’m pretty sure there’s a reason for that but it wasn’t apparent in the talk. While I was underimpressed by that (particularly as the guy is said to be quite a show-off-y young guy), the end was downright anticlimactic when he mentioned that the whole study actually turned out to be useless as they should’ve compared with a third drug. Huh? I thought. Couldn’t he just have kept that detail to himself? It doesn’t affect his theories anyway but it spoils the story.

And the slides weren’t good either. Many of them had different styles, things weren’t aligned properly or just crammed in. The usual PowerPoint junk, I guess. However, he did have a laser pointer to point out stuff on the slides. Which also meant that he used that too much and didn’t look at the audience a lot because he was busy pointing out the words he was just saying on the slides.

The third lecture was about optimising railway networks. Apparently all railways in Europe have similar usage rates with the exception of Switzerland where there are many more users. And one of the main differences seems to be that there are more stations in Switzerland. But of course having more stations means that trains have to stop more frequently and thus will be slower. That is, the system has to be optimised between two ‘ideal’ states, one where the coverage with stations is best but journeys take ages and one where there are no stations, thus giving the fastes journeys for the people who live on the train.

The way to solve this is to create a clever model. Then you’ll find out that it’s NP complete, i.e. too hard. So, do some clever tricks to make it simpler. But then it’s still NP complete. Hmmm, then you simplify, making it really simple and keep an eye on the errors that you get this way. If you’ve done a good job, those errors will remain small but you can solve the problem. As she is from the numerics department, there was also a step with huge matrices containing lots of zeros and ones, which they do computations with.

The next topic were railway connections which can be tackled with similar approaches once a good model has been found. She said that doing those computations for Deutsche Bahn they found out that door-to-door journeys could be something like eight minutes shorter on average with the network of stations they computed. But she wasn’t sure this was actually going to change anything… I guess you can’t just move a station a few hundred metres down the road.

She was definitely the best presenter on the computer, which may or may not come with being the youngest of the three (I think) and working in a computer oriented area anyway. She simply used the mouse to point out things on screen. And she had made the effort of taking some photos of her kids’ toy railway depicting some of the situations she described which was a very nice touch. But there was one big problem apparent from the very beginning… the little ‘1/84’ in the corner of her first slide. Even with funny photos and examples there’s no way to get through those in 45 minutes. So we did see some ‘fast forwarding’ in between. Something that the speaker could have easily anticipated, I suppose.

After that there were some drinks, snacks and hanging out with the mathematicians. Not too exciting, really. One thing I learned was that entering Göttingen and Party into a certain search engine will give one of my web pages in the first couple of results. Colour me amused.

January 27, 2005, 23:05

Tagged as uni.

Add your comment

« iTunes stuffMainA trip to the south »

Comments on

Photos

Categories

Me

This page

Out & About

pinboard Links

♪♬♪

Received data seems to be invalid. The wanted file does probably not exist or the guys at last.fm changed something.

People

Ego-Linking