Archiving is an interesting and, to me, tricky intellectual enterprise. Generally speaking, archives seek to preserve a record of our culture, as much as possible, as best as possible. The digital explosion of the last fifty years has create the need for digital archives, to preserve a record of digital culture.
There are many, many deep problems with this enterprise—digital or not—starting with the question of who would use this data, and what (tiny fragment) should be preserved, and just what preservation means at all.
In this light, I was interested to look at The Living Archives project, which centers in performing arts. The project is motivated by the goal of, “Revitalizing public archives into living social resources implies shedding the conception that they are the dormant and disembodied narratives of a dominant culture.” They approach this with, among other things, a track called “Performing Memory.
Archives are so often “dormant and disembodied” because they are designed by, well, archivists, who are concerned about curating collections. (Archives are expensive, so they do tend to be implemented by wealthy elites. Controlling the official record is almost the definition of being a “dominant culture”, no?) So what will performing artists do differently?
First of all, the project dives right into tough issues far outside the technical issues of curation, such as the tension between open data and personal privacy. It’s interesting to see how performing arts may bring these issues to light, and help people make intelligent decisions.
One recent “performing memory” project that caught my eye was Performing Encryption—an intriguing title!
This is an interesting interactive dance workshop, that invites participants “to generate a digital encryption key using your own gestures and movements.” Cool! A whole body interface, implementing “personalized” encryption key generation!
I think the overall idea is to make a very concrete and visceral connection between you and personal data collected about you, and the role of encryption in controlling access to personal data.The point is made nicely because your dance is captured in video, which is encrypted with the key you generate via your dance. At the end, you get a copy of the video which no one can view unless you share the key with them.
Technologically, this project is pretty straightforward. They use a key generation program that is designed to use random mouse movement as the random input to create a secret key. This software is modified to use traces of you body movement, collected with Microsoft’s Kinnect. It uses the same principle, but you use your whole body. I imagine that everyone who did this workshop has a much better intuition for how key generation works, however you seed it.
The underlying cryptography is standard and widely used, and frankly, doesn’t care how the private keys were generated. While the project materials make comments about how the key is personalized because it derives from your own unique “dance”, that is rather misleading. The key is derived from numbers sampled from measurements of the motion, and even if the sampled numbers are meaningfully related to the motion, they are basically used as random numbers in the key generation. The resulting key is derived from your data, but is not semantically representative of it—that’s actually the critical feature of a private key, it must not be possible to know the original input.
It should occur to the you and the workshop participants that you might be able to use this methodology to create a lock that opens only if you dance the correct “secret”. I.e., you should be able to generate the same key, if you can dance your dance the same way. This would be cool, and I’m sure there are or soon will be simple versions of this available. (This is probably not a very practical security measure—how do you keep people from watching you dance your “secret” code?)
I’m pretty sure that won’t work with this particular technology, because it would be difficult to match the movement closely enough to replicate the key. In fact, it may start with a different initial seed (e.g., from a timestamp), which would make it impossible to replicate the key even in you danced exactly the same. (This is the usual design for encryption, in order to prevent hackers from capturing and replaying a recorded stream of data to recreate the password and break in.)
Anyway, the workshop is kind of cool, and I suspect that many people will both enjoy this approach and feel more personal attachment to the resulting key, and to data encrypted with it.
“Trace the body in the algorithm, reaffirm physical presence in Big Data, and see if your digital encryption key feels more like yours if it was generated as a duet between your movement improvisation and computational processes.”
I imagine that the workshop discussions help relate this exercise to all the other personal data that we create, and perhaps some understanding of how we might choose to use encryption to control personal data.
I don’t know if this workshop advanced the cause of digital archiving, or not. But it is kind of a cool participatory dance project, and a real, honest to goodness, whole body interface that actually does something practical.
I have not really given sufficient attention here to the overall “Performing Memory” thread of this project, which has some deep and intriguing ideas about the relationship of bodies and action to digital life. I will return to this project in a future post.
Robert McGrath's Blog 