Sousveillance (Undersight), and Metaveillance (Seeing Sight Itself)
Total Page:16
File Type:pdf, Size:1020Kb
Surveillance (oversight), Sousveillance (undersight), and Metaveillance (seeing sight itself) Steve Mann Humanistic Intelligence Institute, Veillance Foundation 330 Dundas Street West, Toronto, Ontario, Canada, M5T 1G5 http://www.eyetap.org Abstract Surveillance is an established practice that generally involves fixed cameras attached to fixed inanimate ob- jects, or PTZ (Pan Tilt Zoom) cameras at a fixed po- sition. Sur-veillance only provides part of the veillance story, and often only captures a partial truth. Fur- ther advances in miniaturization, together with wireless communication technologies, are giving rise to kine- matic veillance (“kineveillance”): wearable, portable, and mobile cameras, as well as unpersoned aerial ve- hicles (UAVs). These additional veillances give us a more complete picture: multiple viewpoints from mul- Figure 1. “Stealth” streetlight camera by Apollo, and tiple entities bring us closer to the truth. In contrast “Embedded Invisible PTZ IP Camera” by OWLS AG. to the extensive mathematical and conceptual frame- Here surveillance cameras are concealed inside ordinary ob- work developed around surveillance (e.g. background jects like streetlights, which can be placed throughout en- subtraction, frame-differencing, etc.), now that surveil- tire neighbourhoods or cities, for complete surveillance, lance is no longer the only veillance, we need new math- while themselves remaining hidden. ematical and conceptual understandings of imaging and image processing. More importantly we need new tools cealment. In addition to camera domes, concealment for understanding the many veillances and how they is also achieved by hiding cameras inside other objects interact. Therefore this paper introduces metaveillance such as streetlights (Fig. 1), which are placed through- and the veillance wavefunction for metasensing: the out entire cities while remaining hidden in the light sensing of sensors and the sensing of their capacity to fixtures. There is a kind of irony or hypocrisy in the sense. extensive efforts to keep the all-seeing eye concealed from sight. In addition to physical concealment, there 1. Surveillance is not the only Veillance is intellectual concealment (secrecy), as in the example of the microwave motion sensor of Fig. 2. Surveillance is an established field of research and practice[62, 61, 15, 21, 34] in which sensors are affixed 1.1. Truth and the need for a scientific to stationary objects. “Surveillance” is a French word understaning of Veillance that means “oversight” (“sur” means “over”, and “veil- lance” means “sight”), and suggests an omniscient au- The opposite of hypocrisy is integrity. In Greek thority’s “God’s eye view” from above [19, 40]. mythology, Aletheia is the goddess of truth and the A common surveillance camera housing is the dome, Greek word “alethia” means “disclosure”, “not con- in which the camera is concealed behind a transpar- cealed”, “truth”, or “sincerity”. Alethia’s opposites ent but darkly smoked plastic hemisphere. Inside the are Apate (deception), pseudologoi (lies), and Dolos hemisphere there is also usually an additional opaque (trickery)[57, 56]. black shroud to further conceal the camera. A com- Science is a human endeavour in which we attempt mon design objective in surveillance systems is con- to go wherever the truth may lead us, in pursuit of new 1 1 Metaveillance Metaveillance Univeillance Equiveillance Society Society Veillance Surveillance (oversight) Countersurveillance Sousveillance (undersight) Coveillance Counter- McVeillance veillance Figure 2. Microwave motion sensor concealed behind plas- Stativeillance society Autoveillance Society tic grille. When we look inside to see how it works, we Kineveillance find that the numbers have been ground off all the chips. Dataveillance This is another example of the hypocrisy of “smart things” Figure 3. We can no longer look only at surveillance (over- watching us while revealing very little about themselves. sight). We must also consider other veillances like sousveil- lance (undersight) and coveillance (lateral watching as hap- discoveries[31]. The term OpenScience was coined by pens when people watch each other). We live in a “Par- ticipatory Panopticon” [14] where we’re watched and also S. Mann in 1998 to emphasize this need for truth and sur watch back! The extent to which these veillances are per- open disclosure [48], and if veillance cannot deliver mitted or prohibited, relative to one-another, defines the on that promise, we need to look at other veillances! kind of society we create, in the intercardinals of the “Veil- lance Compass” [41], (Right) shown simplified [17]. This 1.2. The many Veillances broader intellectual landscape requires new tools to help us Recent advances in miniaturization and wireless understand it. One such new tool is Metaveillance, which is the sensing of sensing: sensing sensors and sensing their communication have made mobile sensing practical, capacity to sense. Metaveillance gives us new ways to un- giving rise to a transition from static veillances to kine- derstand all the veillances and the extents of their bound- matic veillance (“kineveillance”), i.e. sensing from a aries, as well as that which falls beyond their boundaries. moving/kinematic frame-of-reference. More generally, surveillance (oversight) is no longer the only veillance. We also have sousveillance ten involves surveillance. Some surveillance is harmful (undersight)[30, 60, 7, 39, 4, 25, 63, 51, 6, 22, 59, 13, (“malveillance”). Some is beneficial (“bienveillance”), 5, 55, 3, 65, 54, 58, 35], in the form of wearable com- like when a machine senses our presence to automate a puting and self-sensing [38, 49] (now known as “Quan- task (flushing a toilet, turning on a light, or adjusting tified Self”), and the wearable face-recognizer [64, 37], the position of a video game avatar). and computer vision systems to help people see and HI (Humanistic Intelligence) is a new form of intelli- remember. gence that harnesses beneficial veillance. HI is defined Surveillance happens when we’re being watched by Kurzweil, Minsky, and Mann as follows: and sousveillance happens when we do the watching. “Humanistic Intelligence [HI] is intelligence that And with social networking, we now have co-veillance arises because of a human being in the feedback (side-to-side watching, e.g. when people watch each loop of a computational process, where the hu- other) [50]. Thus we must think of Veillance in a man and computer are inextricably intertwined. broader sense beyond merely Surveillance. (See Fig. .) When a wearable computer embodies HI and be- comes so technologically advanced that its intel- ligence matches our own biological brain, some- 1.3. Surveillance and AI (Artificial Intelligence) thing much more powerful emerges from this syn- ergy that gives rise to superhuman intelligence Surveillance is a broad concept. It goes beyond vi- within the single ‘cyborg’ being.” [52] sual (camera-based) sensing, to include audio surveil- lance, data surveillance, and many other forms of HI involves an intertwining of human and machine in surveillance. When we say “we’re being watched” we a way that the human can sense the machine and vice- often mean it in a broader sense than just the visual. versa, as illustrated in Fig. 4. When police are listening in on our phone conversa- HI is based on modern control-theory and cyber- tions, that is still surveillance. So, more generally, netics, and as such, requires both controllability (be- “surveillance” refers to the condition of being sensed. ing watched) and observability (watching), in order We often don’t know who or what is doing the sens- to complete the feedback loop of any kind of HI. In ing. When a machine learning algorithm is sensing us, this way, surveillance (being watched) and sousveil- that is also surveillance. AI (Artificial Intelligence) of- lance (watching) are both required in proper balance 2 Senses Effectors Senses Effectors 1 Human 2 1 Human 2 Random 3 4 Delay 4 Observability (Sousveillance) Controllability (Surveillance) Machine Machine Controllability (Surveillance) 5 Sensors Actuators 6 5 Sensors Actuators 6 Humanistic Intelligence Machines of Malice (HI) ("Digital Lightswitch") S. Mann, 1998 S. Mann, 2013 Figure 4. The Six Signal Flow Paths of HI: A human Figure 6. Systems that fail to facilitate sousveillance are (denoted symbolically by the circle) has senses and effects machines of malice. Example: an art installation by au- (informatic inputs and outputs). A machine (denoted by thor S. Mann, consists of a pushbutton (push-on/push-off) the square) has sensors and effectors as its informatic in- light switch where a random 3 to 5 second delay is inserted puts and outputs. But most importantly, HI involves in- along with a 10 percent packet loss. The parameters were tertwining of human and machine by the signal flow paths adjusted to maximize frustration in order to show a neg- that they share in common. Therefore, these two special ative example of what happens when we fail to properly paths of information flow are separated out, giving implement a balanced veillance. a total of six signal flow paths. works 90 percent of the time, and, combined with the delay, users would often press it once, see no immediate Veillance is the core of Humanistic Intelligence effect, and then press it again (e.g. turning it back off before it had time to come on). See Fig. 6. Senses Effectors ) Human ) 2. Sequential Wave Imprinting Machine This section describes some unpublished aspects of veillance veillance a wearable computing and augmented reality invention Controllability (Sur Observability (Sous by author S. Mann for making visible various other- Machine Sensors Actuators wise invisible physical phenomena, and displaying the phenomena in near-perfect alignment with the reality Humanistic Intelligence to which they pertain. As an embodiment of HI (Hu- (HI) manistic Intelligence), the alignment between displayed Figure 5. HI requires both Veillances: machines must content and physical reality occurs in the feedback loop be able to sense us, and we must be able to sense them! of a computational or electric process.