Archive for September 2007

Moebius Transformation Revealed

I would just like to share this interesting video explaining Moebius Transformation. In geometry, a Moebius transformation is a function defined by F(z) = (az+b)/(cz+d), where z, a, b, c, d are complex numbers satisfying the condition that ad-bc ≠ 0. (See Moebius transformation in Wikipedia). In simpler terms, this transformation sends each point on a plane to another point somewhere else on the plane by a combination of simpler transformations including translation, rotation, inversion, and dilation. The resulting transformation can appear very complicated. But when viewed with an added dimension (in this case, a third dimension), the transformation becomes surprisingly simple! 

This reminds me of disco mirror balls. By reflecting light directed at it in many directions, it can produce complex display patterns. A disco transformation? By the way, this video was created by Douglas N. Arnold and Jonathan Rogness of the University of Minnesota and was awarded an honorable mention in the Science and Engineering Visualization Challenge 2007 in the noninteractive multimedia category.  For a complete list of the visualization challenge winners, visit Science magazine’s feature story here!

Poster or Oral?

postersession.jpgPresentations in conferences are typically classified into oral or poster presentations. Obviously, oral presentations are reserved for those topics which are considered “hot” and of higher appeal to the intended audience. If you want to increase your signal-to-noise ratio (here, signal pertains your scientific work, in contrast to noise, which is what everyone else is talking about), oral presentation is the way to go. This is the best way to get maximum exposure or visibility to other researchers. Indeed, this can be conveniently thought of as “parallel visibility.”

Posters, on the other hand, allows you to discuss your work to one person at a time. In contrast to oral presentations, this only allows “series visibility” of your work (and thus, lower signal-to-noise ratio). Nevertheless, this allows more in-depth discussion of your results, and even offers the possibility of networking with other people interested in your work. Some valuable ideas could come up as a result of these in-depth discussions - something that doesn’t usually come by way of an oral presentation, which is of course constrained due to time limitations.

In the biannual meetings of the Japan Society of Applied Physics (Nihon Oyobutsuri Gakkai), there are even sessions where both poster and oral presentations are combined. These are called short oral presentations with poster. Oral presentations are given for 5 minutes (usually in the morning), followed up by a 2-hour poster session (usually in the afternoon). Oral presentations usually just give a highlight of the main results, which allows the audience to choose which more interesting results to go to during the poster session.

If I were to choose, however, I would prefer an oral presentation. I find poster presentations particularly tiring, especially because most poster sessions I attend are usually two hours long. If there are is a considerable number of people who are interested in my work, that usually means that I have to stand and talk for the entire two-hour duration! Also, most of the interesting results related to my work tend to be presented in the same session, so this also means that I miss out on discussing those results with the concerned poster presentors because I have to attend to my own poster. Of course, this can also be conveniently circumvented by sneaking out of one’s poster presentation and doing the rounds of other posters. For a two-hour poster session, I think the best compromise would be to stay at one’s poster for about 1 hour, then use the remaining hour to check out other people’s posters.

What about you, what is your preference? Oral or poster? :)