I felt quite honored to be invited to give the Evnin Lecture at my alma mater (Princeton University) in November 2024, on the general subject of art and science integration. Preparing the talk was really useful for this project as I wanted to include the indigo memristor and silk-casting works in progress but had to consider what I would say about where it was all going. On the indigo side, especially, I feel like I made a sort of mental breakthrough in terms of conceiving what kind of final artwork could be practically feasible (within manageable effort and expense), well motivated in both an artistic and “technological” sense, and expressive of the personal and historical themes the larger project is supposed to be about.

Although I wrote in my original project proposal about an ambition to use indigo memristors to make a functional digital/electronic memory, as a kind of exercise in nonfiction Silkpunk (in the spirit of Ken Liu), I’ve had lingering doubts about how a practically feasible construction of that kind could really do something interesting. As in, what is really the significance of a digital/electronic memory device? If our goal is to preserve small quantities of information (it has been hard for me to see how a hand-made indigo memristor memory bank could have more than ~100 bits of capacity), we’re much better off just writing it down with pen and paper (or even better, as Cixin Liu highlights in Death’s End, chisel and stone) in terms of the robustness and longevity of storage. Really, in broad technological context, we only need digital/electronic memories to interface with digital/electronic signals and signal processing. If I’m not going to construct an entire indigo-based digital signal processing system, how Silkpunk would it really be to exhibit some sort of installation that relies on conventional silicon electronics even if it incorporates an indigo memristor memory?

A key mental shift for me was to focus not on the remembering aspect of what an indigo-based memristor is/does, but rather the fact that a single bit memory is by definition a device that can sit quiescently in two diametric states. In computational language, a single bit memory has to be able to sit in either a zero-state or a one-state; in dynamical systems language a single bit memory must be bistable … and with a bit of poetic license a single bit memory is capable of holding contradiction. Looping back to the nascent spirit of my project proposal, this mental shift pointed me in the new direction of thinking about the memristor construction as a way of making indigo the site of holding contradiction–in my project context, contradictory feelings of fascination and dismay that can arise when learning about one’s culture/state of origin.

I mentioned this new perspective in my Evnin Lecture but it took another eight months or so to complete the thought into a concrete idea for an indigo memristor-based artwork. I’m still not sure how feasible the new scheme is but will say more about it in a future post…