Weekly Head Voices #12: Ceci n’est pas une bibliothèque.

Welcome to the latest edition of the Weekly Head Voices, in which I briefly touch upon the, to my mind, mention-worthy events that took place within my field of observation during week 5 of the year 2010, and with which I too finally have an excuse to (ab)use the famous words of Magritte for my dubious ends. :)

WE INTERRUPT THIS PROGRAMME FOR THE FOLLOWING IMPORTANT ANNOUNCEMENT:

I’m officially not supposed to talk about this until the next edition of the Weekly Head Voices, but it’s too big and too cool to keep quiet about until then.  Google has just released Buzz, their location-based status / media updating system, and it’s fantastically cool.  I’ve just posted my first Buzz via Google Maps 4.0 on my E71 (I think I’m the third buzz in Delft EVAR).  Don’t know what Buzz is? Check this YouTube clip:

YOU WILL NOW BE RETURNED TO YOUR NORMAL PROGRAMMING.

The reason for the title of this post is my Saturday visit to DOK, Delft’s unique library concept. It’s not really a library, but more of a fantastic place of gathering that coincidentally contains thousands of books, CDs, DVDs and, err, a coffee shop! They even have a number of sonic chairs that one can make use of to listen to music via the mounted Macs. This Saturday, live music by one of the artists who exhibited at the DOK.  I made you a short snippet:

As even this short edition should end on a philosophical note, I’d like to conclude with an interesting discussion I had on whether the type of research we do in (medical) visualisation can be considered to be science. Very strictly speaking, the scientific method consists of observation, hypothesis forming and finally experimentation to prove or disprove the hypothesis. A large body of visualisation work is concerned with making stuff that solves hard problems, i.e. formulative research as opposed to the more traditional evaluative research. Although the question of whether making stuff that solves hard problems constitutes science is a complex discussion that deserves a whole year of blog posts, I am going to conclude with one possible and simple take on situation:

By taking this constructive approach we are, besides actually solving problems (a neat by-product, no?), discovering how to create effective visual representations of complex phenomena hidden in even more complex data. By doing this, we are in fact observing the supremely complex system consisting of the whole pipeline from data acquisition to insight, all the while experimenting with parameters (in the widest possible sense of the world) and thus confirming or disproving hypotheses concerning the nature of the pipeline and its various components. Together, these hypotheses make up the model that governs the effective extraction of insight from data via the human visual system.

5 thoughts on “Weekly Head Voices #12: Ceci n’est pas une bibliothèque.”

  1. I think one shouldn’t worry too much on whether a “real science in progress” label could be applied to our visualization work. First, much of our more practical results can be considered as enabling tools for science itself in the sense that they enable “real” scientists to investigate data. Second, as soon as we investigate or measure (human) performance on our tools were doing “the real thing” ourselves.
    Just like in other disciplines, and not only those in Computer Science, we can also lean much more to design-oriented approaches (from industrial design, architecture). With this in the mix it makes visualization much cooler than “just” classical science, don’t you think?
    On the other hand, what if we would continue this reasoning, would we end up here: “Professor dr. Kunstenaar” http://weblogs.nrc.nl/cultuurblog/2010/02/02/professor-dr-kunstenaar/

    1. I agree with your points, although I *do* think we should worry, at least constructively worry. This type of constructive self-criticism is good for the field when we keep on analysing and improving.

      With regard to Professor dr. Kunstenaar: I think that a certain type of art does play a small (but important role) in visualisation, but I wish to remain on *this* side of the spectrum, thank you very much.

  2. That must have been a really interesting person you had this discussion with. No really, I mean, think about it. Visualisation & Science. Pure. Genious. :)

    I assure you that there is going to be a proposition on this in my thesis and albeit slightly premature, I hereby challenge my committee to counter it ;).

  3. I join the disscussion for a few words. :)
    Maybe holding the goal of solving hard problems, and on the way of exploring, new discoveries would come out, without prophet, in “real” science. :P

  4. I largely agree with Gerwin’s view. We create the tools for the “real” scientists. But I think visualization is special in a different way also. The tools that the “real” scientists use, or as Charl calls it: “the pipeline from data acquisition to insight”, for me consist of two parts. The first part is imaging, where physical properties are mapped to raw data, which is physics, a “real” science. The second part is the image processing/visualization part, where the data is processed into visuals in such a way that the “real” scientist can interpret the data. Now, this interpretation step is important (and potentially very dangerous, but that is not my point right now). It is as if the “real” scientist is wearing glasses that distort reality in a way that we, as visualization scientists, think is right. In a sense, we create the reality as the “real” scientist perceives it. With respect to the “real” scientist, we create the truth, we are God. Visualization is not about science, it is about creating a belief system, creating a religion. (Something you should be careful with, but that’s a whole different topic… :-P)

Leave a Reply

Your email address will not be published. Required fields are marked *