News/Research
25 Aug, 2022

Summer Research Reports: Kevin CK Lo

We're pleased to support our students in their summer research. Read about Kevin CK Lo's work building a modular geo-spatial web template below!

UC Berkeley's Summer Research Award helped provide time and resources toward building a modular geo-spatial web template that sends any device's location, orientation and touch data to a MQTT server (provided the sensors exist), that can then be accessed by other credentialed users for artistic projects.

This is a template that can be easily modified to suit anything from live performance, to classroom demonstrations of networking with a locative media focus. It is functionally as "copy and paste" as I could make it. Several problems were worked out, one problem remains in relation to use of this on campus wifi – this will all be elaborated below. In any case, one major result of building this infrastructure has made conjuring projects far more creatively free. Since starting work on this, I have thought of multiple uses for BROADSPATIAL (a working title, not to be used in production) that didn't figure at the beginning of summer. Upcoming, I will be programming a performance at CNMAT that uses this audience interaction as a major component. I'm also co-teaching Music 158A this semester (Sound and Music Computing with CNMAT Technologies), where I will now use BROADSPATIAL as a teaching tool, for one of our "capture the flag" learning exercises. An installation piece I am thinking of, provisionally titled "no one looks at each other anymore" (a poetry line by Chris Nealon, which inspired this idea), would use BROADSPATIAL to, in essence, allow people across the world to look at each other, no matter where they are. Major problems solved: how to keep the code light and simple; how to abstract the data cleanly so data traffic is light; the architecture of having a central server that processes incoming data, that can aggregate and resend as needed as the best solution; how to ID each user in a way that is light and performant. One unfortunate problem to be solved is that the MQTT server that I use (and continue paying for via the Research Award) blocks data on UC Berkeley wifi. A provisional solution is that mobile devices can turn off wifi and use mobile data instead, but ideally, there is less fiddliness in setup (especially in the context of a performance situation), and I will have conversations with Prof Ed Campion to see if access to shitfr.io can be allowed for data traffic. When it comes to any other non-corporate use, there are no issues.

I'm looking forward to using this in performance, and appreciate that this has developed into standalone platform that, aside from the shiftr.io server access, does not need another intermediary platform to create its art. In contrast to this, I think of newart.city, which my co-lecturer frequently uses, but I find ultimately clunky, and most importantly, less designable. I'm looking forward to receiving feedback from participants who can interact in a live performance setting, in this "spooky" way. Much appreciation to the award for its support and to the BCNM team who make it possible!