TDDE 131 Week 1: Measurement & Uncertainty, Scores

As a means of archiving our course, we’re going to attempt to write up a short review of what we covered, along with links to whatever media/exercises/ideas that came up.  For those of you not fortunate enough to be sitting in, please feel free to check out the links below and try the exercises at home!

Introduction & Context
Measurement & Uncertainty
Laughter Scores

Pre-class reading:

Chapter 7 of The Drunkard’s Walk: How Randomness Rules Our Lives by Leonard Mlodinow


6:00-7:00pm: Introductions, ice-breakers, course information

we opened with a “who we are and what is underappreciated”; many TV shows came up (X-files, Torchwood, Arrested Development) and … hobbies (as a whole category of underappreciated things)

name-games: Michael is now to be referred to every third time as “Jason”

Why this course? The idea grew out of Tara, Michael and Adam’s desire to blend the tools and techniques of Astronomy with the tools and techniques of Art Design; to break down the language, perceptional, and functional barriers between our practices; and frankly to have some fun with what we’re doing.  We are a society of mutual admiration and subtle envy over what the others are doing. We are not trying to just make art out of science; Tara’s life goal is to actually have art advance scientific discovery (after which she will devote herself to hobbies).

Grading: to fit with the theme of the course, we’ve proposed something a little experimental and downright scary to the physicists: the metrics and breakdown for assessment will be decided as a class around weeks 3/4.  yes, the proletariat student body will have control of the means of grade production.  don’t freak out.

Illustrative examples: Michael demonstrated some beautiful examples of data-driven and process-driven art:

Mouse-clicking app IOGraph developed by Anatoly Zenkov produces maps of where your mouse moves on your screen and how you click, a “live” work that is both data-driven and aesthetically interesting.

Hatsune-Miku is a “collaboratively constructed cyber-personality” (a “wiki-celebrity”) which began a singing voice synthesizer and has grown into a global cult phenomena; Tara is currently making a documentary about the Hatsune-Miku.

 Dan Deacon’s remix of an acapella version of Carly Rae Jepsen’s “Call Me Maybe” was created by inserting a layer of the song on top of itself 147 times in each measure; the result is something the diverges from the original song into a distinct audio experience.



Mark Hansen & Ben Rubin’s Listening Post culls fragments of text from 1000s of chatrooms and presents them in a flattened, homogenized format that changes the connotation of the personal statements made (see this YouTube video of the piece)


Laptops talking to each other (see this link on animated chatbots doing the same) – an example of data-driven performance by rules-based agents.


Copyright Symphony by R. Lee Montgomery takes a series of images that have been in infringement copyright cases and transposes their pixels directly into sound.  The piece is a good example of transsensory transformation (converted the visible into the audible); play several of these at once.

The Shape of a Song by Mark Wattenberg visualizes song structure by mapping the patterns across a song; a Jave tool on the website allows you to create similar patterns for any midi file.


Visualization of eye tracking of a picture, based on specific prompts, from a 1967 book by saccadic pioneer Alfred Yarbus



Lev_TIMEMAGLev Manovich and Jeremy Douglass’s created a montage of all Time magazine covers from 1923-2009; this massive visualize draws out patterns that can be linked to choices in saturation to the sitting President’s party affiliation (see more examples here)

7:00-8:00pm: Measurement and Uncertainty

As we’re working with astronomical data this term, it is essential to get an idea of the fundamental source of data, measurement and its associated uncertainty.  Echoing Mlodinow, we emphasized the difference between measurement of a thing and the thing itself, and the inherent role the observer plays in the measurement process.

We then performed an exercise in measurement: groups of 3 had 20 minutes to measure the size of a stone, a whiteboard, and the building we were in using only their bodies as references.  Each was left to their own device as how the measurement should be made. Some interesting observations:

(1) When asked which was the hardest, there was no consensus.  The stone was hard for one group because it was so different in scale to the building; the white board was hard because it the bottom was elevated with respect to the floor; the building was hard because of its sheer size compared to the human body

(2) One group devised different scales of measurement: one corresponding to the stone, one corresponding to an arm’s length, one corresponding to the distance a stone could be tossed.  Ultimately all three measurements were converted to the largest size

(3) Three groups devised measuring scales for the building that were fundamentally based on time and motion.  One group counted snaps as they walked at a regular pace; two other groups timed a person running. The decision to go with a time-velocity approach mirrors the use of light-years as a natural unit in astronomy.

(4) Each group ultimately came up with completely different and essentially incompatible distance measurements. Our inability to compare our measurements illustrated the importance of unit standardization, a step (in the late 18th century) that was critical to being able to make measurements beyond our body scale to astronomical objects.

We then discussed at length uncertainty and its artistic counterpart, variation.  Several parallels were drawn between measuring an thing and judging a thing (e.g., what is good changes in time and culture, Jazz as an obvious example); and performance as measurement (e.g., Bach being played by different musicians will sound different even though its the same piece).  Uncertainty is quantifiable in science and desired to be small; variation is not easily quantifiable in art, and degree of variance is more nuanced; having some variation may be good, having too much may not (an extreme tonal design may cause a piece to not be “music”).

We also talked about the nature of uncertainty; is a measurement number “the irrefutable real deal”?  Maybe the object you measure is absolute, but the process of measurement will always induce uncertainty to some degree, it is unavoidable.  However, even in science it may be hard to know the uncertainty if, for example, if we can make only one measurement (e.g., detection of the Higgs boson by the LHC, an experiment that will happen perhaps once in our lifetime).  Similarly, variation in art is easy to spot but hard to predict ahead of time.

8:00-8:50pm: Laughter score

We examined the Laughter Scores of Edward Jessen, who performed a scientifically-modeled study of laughter.  He identified categories, made measurements and recorded them in the notation of music:

Michael led the class in trying to replicate some of these laughter scores, and emphasized how much more interesting and engaging these were as scores to perform than simply a study to be written about.  Science can act in much the same way.

Assignment for next week:

With your group of 3, measure something, create a score, and turn it into a 2-3 minute performance, with the limitation that you can only use your bodies to “perform” the score – tapping, talking, looking, singing, crawling, what have you (no  nudity, cruelty to animals, people or plants, or open flames please!)

%d bloggers like this: