Thursday, August 28, 2008


Talking Technically

Unfortunately, many people in our field are poor speakers. That does not only apply to graduate students, but also to more senior researchers.
Recently there have been some seminar talks in SoC by supposedly famous professors that were were just amazingly boring and turned out to be a total waste of time.
But today there is finally a seminar talk that is interesting and fun to attend. The topic of the talk is about how to give a good talk. It is given by Dr Terence Sim and the aims at making life easier for students who have to present their research (e.g. PhD thesis defense) and those who have to listen to their talks.

A common misconception about giving a research talk is that the talk should cover EVERYTHING that you have done in your work and that is written in your paper. Instead the talk should tell a story about the research you have done. The paper is of course also telling a story about the research you have done, but it is important to realize that the two stories are not necessarily the same. So instead of "zipping" the content of the paper and squeezing it into a (usually short) talk, the talk should be an interesting story about the work you have done. If somebody wants to know more details about your work, he can always read your paper.

Another important point is to have enough redundancy in your talk:

1. Tell them what you are going to tell them
2. Tell them
3. Tell them what you have told them

The reason is that the audience only has a limited short term memory, which means that the average listener in your audience will only memorize 7 (+/-2) items of your talk at a time. So if you have a lot of details in your talk, the chance is that many in the audience have already forgotten what you said in the beginning when you are in the middle of your talk.

Unfortunately, as with almost everything, it is always easier to agree to good advice to how to give a talk then actually giving a good talk yourself...


Monday, August 25, 2008


You are out of here

My first attempt to publish an academic paper was rejected. The result came in two days ago on Saturday. Its a bit disappointing to see my paper rejected, because of the extra work that I put in towards the final stage of my diploma thesis (which was already under strict time constrains) and secondly because the initial reviewer response were surprisingly positive and we already deemed our submission successful.

Let's see it from a "Squash perspective": Those matches that you loose usually teach you more than those that you win. They don't taste as sweet and nobody likes loosing, but it just gives us the chance to reflect on ourselves, train harder and try again the next time. So back to training/studying. The next conference will come for sure.


Tuesday, August 19, 2008


Why do the Austrians speak German? Did the Germans conquer them?

Sometimes it is striking who little we know about our own history. When it comes to the other people's history, the results are even worse.
Luckily we can spend most of our time without being confronted with these facts, but when we speak to people from another country this can change very quickly.

Take today, when I was chatting with one of my PhD colleges about Germany and other European countries. He admitted that he doesn't know much about European geography and history (I don't know much about Chinese geography and history either), so I quickly opened Google maps to show a map of Germany and its neighbors. I also pointed out Austria and mentioned that they speak German as well. This puzzled my college and he asked why the Austrians would speak German and whether the Germans had conquered Austria before. So I tried to explain that until the 19th century there was no state called "Germany" but that most of the time Germany was a patchwork of many different kingdoms, princedoms and other small states. This caused even more confusion and I tried to give a brief summary of the Holy Roman Empire of German Nations. This was the moment when I felt that my history knowledge was a bit rusty. I could not remember whether Austria was at any time part of the Holy Roman Empire or not. The name Habsburg still came to my mind, but why exactly they speak the same language I couldn't say. When my college asked me some more questions about how the people in Hungary are related to other European people, I had to give up completely.

In the end I took 10-min of my work to browse wikipedia on the subject of the Holy Roman Empire but the question still remained unanswered.


Hungry Ghost festival

The seventh month of the Chinese calendar is the month of the hungry ghosts. During this time, the gates of the underworld open and the ghosts are allowed to walk the earth. To keep the ghosts happy (and probably from doing any harm to the living), people make offerings to the ghosts and there are street performances, like music, plays and Chinese opera to keep the ghosts (and the living) entertained.
So these days I can see some of my neighbors burn incense and light candles outside our block. The burned incense is made of paper and often resembles offerings to the ghosts, e.g. a car or a watch.
Unfortunately I haven't found time to attend any of the street performances, but I would have my difficulties to follow, as the performances are usually in Hokkien.

My knowledge of Chinese believes is still far from perfect, so I apologize for any mistakes in my explanations, please feel free to highlight any errors.


Monday, August 11, 2008



I have started graduate school! Surprisingly graduate school doesn't look much different from my DA research work: I sit in the same lab, same supervisor, working in the same research area,...

However, there should be a slight difference in the work attitude in graduate school. While during your undergraduate studies it might still be sufficient to follow, from now on it should be the case that I am part of those people that are spearheading the research process in computer science. Whow, that sounds really scary :)

And of course, the search for the PhD thesis topic has officially begun now. Finding a topic can be of the hardest challenges in grad school, a process that is hard to control and needs some inspiration.
One of the Prof's told us that he found his topic while taking a shower.
But basically, you have to think about a problem 24/7, eat with it, sleep with it, live with it; then eventually you will have a great idea. I read that Andrew Wiles (the guy who proved Fermat's last theorem) thought about his problem constantly for over eight(!) years: he woke up in the morning and thought about Fermat's theorem, then went to the attic where his study was, thought about the problem the whole day until he went to bed, while still thinking about Fermat's theorem and so on...
(At this point, I of course want to note that Wile's work proved one of the most famous theorems in math's history and that I don't put my PhD anywhere near his work.)


This page is powered by Blogger. Isn't yours?