A new visual language for cybersecurity with “SweetWire”
The UC Berkeley Center for Long-Term Cybersecurity convened a panel discussion featuring Greg Niemeyer, a winner of our 2019 Cybersecurity Arts Contest.
On September 16, 2020, the Center for Long-Term Cybersecurity and the Daylight Security Research Lab presented a discussion and live Q&A to showcase SweetWire, one of the winners of our 2019 Cybersecurity Arts Contest. Created by artist Greg Niemeyer, in collaboration with dancer Olivia Caldeira Holston and composer Hallie Smith, SweetWire is a music video project that translates a successful containment of a cybersecurity attack into music and choreographed motion.
The Cybersecurity Arts Contest was launched by the Daylight Security Research Lab with a goal to shift how people understand and identify the harms of technology. Artists from around the world submitted proposals that were reviewed by an interdisciplinary committee and judged for artistic merit, relevance, feasibility, and potential impact, including how they might influence other artists or reach particular audiences.
One of the winners of the contest was Greg Niemeyer, a professor for New Media in Art Practice at UC Berkeley and co-founder and director of the Center for New Media. For the panel discussion, Niemeyer and his co-creators joined Sydney Skybetter, a Public Humanities Fellow, Lecturer, and Director of Undergraduate Studies in the Theatre Arts and Performance Studies Department at Brown University, to discuss their experience in creating SweetWire.
A New Visual Language
The idea for SweetWire began with Niemeyer’s previous project, Network Paradox, a reflection on the history of the internet and coming to terms with the feeling of being overwhelmed by data. Together with co-creators Roger Antonsen and DJ Spooky, Niemeyer looked at the ways in which networks transform. “If a network gets infiltrated, how do the nodes pass on a virus from one node to another? In 2020, the way viruses get passed on through networks is something we deal with every day — the topics of infection, takeover, and viral transformations have only become more relevant,” Niemeyer said.
Moving beyond hackneyed visual portrayals of computation, SweetWire shows that cybersecurity looks more like particle systems than command prompts — more like a network than a man under a hoodie with glowing green letters. These common tropes of information technology graphics often obscure the complexity of what happens beneath the surface. In order to visualize the overwhelming amount of information that flows from node to node around the world, Niemeyer explained that “a new visual language was required.”
The Sound of SweetWire
SweetWire was filmed using point cloud capture systems to convey the complexity and abstraction in the materiality of data. A point cloud is a set of data points in space. Point clouds are produced by 3D and LiDAR scanners and used in technology such as Microsoft Kinect devices, which measure many points on the external surfaces of objects around them. Point cloud capture systems record a scene by measuring where various points cast out of a camera land in a given space.
During the discussion, Skybetter and Niemeyer discussed the “visual polyphonics” of the video and how the viewer sees objects from multiple data layers and vantage points. Niemeyer described how the layering process began by giving Hallie Smith, SweetWire co-creator and composer, an outline of the temporal dimensions to compose the audio. Smith recorded with a modular synthesizer to unlock pitches that are considered “out of tune” by Western standards of music intervals and combined pulsing and droning effects to emphasize tension in the video. The narrative of merging the hacker and the system was key to the music and its texture. Throughout the video, musical phrases are out of sync, build up and intensify, then eventually resonate in harmony, mimicking points at which two visuals come together.
Embodying SweetWire
Smith’s sonic structure was then used as an outline for the dance, choreographed and performed by Olivia Caldeira Holston. Holston described creating a series of contorted dance movements based on the various layers of the music composition. When she arrived to record the dance, Niemeyer challenged Holston to significantly slow down the dance phrases, designed to be quick and jarring contortions, and hold the difficult positions in order for the point cloud system to fully capture and render her “embodied” data.
“The dialogue between data and what we do to our bodies to fit data models is impressive,” said Niemeyer. “The dialogue between Olivia’s dance and how we captured it is a metaphor for the dialogue between our lives and how they’re reflected in the data collected about us.”
The emphasis on capturing Holston’s movements was not merely to represent a body, but to capture her movements as an expression and a statement. Her expressions came through clearly and were amplified dramatically when the data “falls apart” as the representation fails. “Most representations claim totality,” Niemeyer said. “Framed images are presented as something integral and complete….This leads to a lot of misrepresentation.”
SweetWire shows the fragility of the representational method itself. “Half the time it’s hard to make out a signal from the noise, half the time the body feels shaky,” Niemeyer explained. “We like that because it brings into question the integrity of representation. We’re saying that we’re leaving out as much as we show.”
Shooting SweetWire
The video production team, including Jacquelyn Serrano, Oliver Moldow, and Joshua Hyman, described the uniqueness and challenges of filming with a LiDAR scanner for the first time. “The most interesting thing to me was trying to understand the spatial implications of using LiDAR scanners, which are typically used to scan large rooms for architectural purposes,” said Moldow.
The team repurposed a Microsoft Kinect scanner that had a very specific range for capturing images. The filming process was made more difficult because there was no monitor for the production team to see what was actually being recorded. They had to rely on intuition and be meticulous about distance in order for the laser to scatter so that it captured the dance movements and the various corporate building spaces.
The final product does not capture direct images of the dancer or surrounding objects, rather it captures the location of the dancer and surrounding objects in space. Niemeyer and his team then took that data and used another software program to rotate the dancer and objects — a “synthetic, second filming” to change their points of view.
Niemeyer tied this abstraction to cybersecurity by explaining that the meaning of the data only comes once it has been processed and becomes information. How we extract value from data happens not at the level of collection of data, but from data analysis. “You can think of cybersecurity as something that happens when we protect data, but we really need to protect the information, not only from people with illegitimate access, but from people who have legitimate access who sometimes use data to do illegitimate things.”
Cybersecurity Implications
A fundamental question SweetWire aims to answer is whether we are using data in a fair and just way. The issue of cybersecurity is much broader and multi-faceted than it is typically framed. “There is no solid boundary between good guys and bad guys — there is no binary,” Niemeyer said. “It’s much more complicated, and the complexity of this rhetoric is built into the visual and sonic layers of this video.”
Learn more about SweetWire on Greg Niemeyer’s website. Visit the Center for Long-Term Cybersecurity’s website to learn more about the 2019 CLTC Cybersecurity Arts Contest.
