CS not a science, not about computers

Although the name of the field, Computer Science, does not imply it, computer science is not really a science and is not really about computers. From Abelson and Sussman (1985 and 1996), The Structure and Interpretation of Computer Programs (the intro CS course text at MIT):

Underlying our approach to this subject is our conviction that “computer science” is not a science and that its significance has little to do with computers. The computer revolution is a revolution in the way we think and in the way we express what we think. The essence of this change is the emergence of what might best be called procedural epistemology—the study of the structure of knowledge from an imperative point of view, as opposed to the more declarative point of view taken by classical mathematical subjects. Mathematics provides a framework for dealing precisely with notions of “what is.” Computation provides a framework for dealing precisely with notions of “how to.”

This is spot on. Abelson mentions in his lecture that surveying techniques originating in ancient Egypt led to geometry (earth measure), but not until geometry transcended the survey and encompassed formalisms did it become applicable to so much more. The same can be said of computer science: the work we do today in formalizing notions of "how" will one day transcend the computer and apply to so much more -- and it's just starting to happen with our realization that computation shares much in common with biology.

Oh, and computer programming is something that most computer scientists know how to do, but is not what they really do. It's like how musical composers know how to play an instrument, but it's not really what they do and there's much more to music than just knowing how to play. It's a skill. Knowing how to program (or play a piano) and being able it do it well helps a lot (and many times pays the bills), but it's only the surface.

In my next post: the redefinition of computer science in hopes of a revitalization.

i want a polaroid camera

Polaroid cameras almost completely faded into non-existence with the advent of the digital camera, but surprisingly enough, they are making a comeback in some of the off-beat artistic photography. Actually for some people, Polaroid never left. I had a Polaroid taken of me at a restaurant in Taiwan—a place where they liked to hang customer photos on the wall. I was amazed at seeing a Polaroid camera still being used and thrilled at seeing the photo develop right then and there. The last Polaroid I appeared in featured a cardboard cut out of Ronald Reagan (current President at the time) in front of the National Air and Space Museum.

I found a spectacular stop motion animation on youtube made from 987 Polaroid frames, shot with a Canon camera without any computer manipulation.



I like how Process Enacted (above) is a meta-movie: a movie about making the movie itself. This reminds me, I have yet to see Adaptation, which is supposedly a movie about making the movie Adaptation. It brings me back to some of my undergraduate computer science classes where we talked about reflection: when a program examines and modifies itself while it is running. (Computer science is really more about philosophy than it is about science—more on that in my next post.)

If anyone wants to unload a 1970s-era Polaroid SX-70, 1980s-era SLR 780, or an 1990s-era SLR 690 camera, let me know! Oh and despite the song telling you to "shake it like a Polaroid picture," don't shake your Polaroid pictures :)

music animation machine

One of the people who attended the Experimental Two-Day Course by Edward Tufte last month was Stephen Malinowski, the creator of the Music Animation Machine.



Malinowski's work was shown in Tufte's course as an example of a multivariate information display. A key to interpreting the dimensions in the display:

• Musical themes are represented by color.
• The pitch of the notes are represented by placement on the vertical axis.
• The duration of the notes are represented by the horizontal length of the lines.
• The combination of different notes being played at the same time is shown by vertical overlap.
• And all of the above is being plotted over time (via scrolling).

That's five dimensions of data being mapped onto two dimensions, horizontal and vertical.

The point of this display is to help listeners who cannot read sheet music to visualize the music as if they did know how to read sheet music. By focusing on certain elements in the visualization, one can listen critically for details and hear things and anticipate things that would otherwise be difficult without any sort of visualization.

It's sort of interesting how one more clearly hears the notes that one is seeing in the display. Plus since we humans are so good at pattern recognition (relative to any computer), the visualization makes it easier for us to spot and hear little repeated themes in the music. This is the world that sheet music readers live in -- they can follow the score and see the music while they are listening to it. The great thing about Malinowski's animation is that it's intuitive enough that non-sheet readers, and terribly slow sheet readers who never learned how to read the bass clef (like me) can use it in place of sheet music. It's not meant to be a replacement of sheet music -- just an alternate presentation of it.

So what's the point? Is this just something novel and fun? Well yes, the Music Animation Machine is fun to watch and listen to, but there's a greater lesson to be told here. Visualization helps us draw out (in the figurative and literal sense) our understanding what is happening, not only in a piece of music, but in anything we are interested in studying in the world. The world is much more interesting than a line graph because it's so multivariate. However, the challenge that remains is how to effectively present multiple dimensions of information to help us understand the system at hand and tell the story of what's really happening.