documentation

After the Symposium by Seda Gürses

Obfuscation and Resistance by Karen Levy

Exploitating Data Entropy by Melissa Gregg

Slides and Visuals

After the Symposium

Seda Gürses, 30 March, 2014

Every event gains a greater meaning with the little surprises that accompany it all along, and this was even more true for an event on the topic of obfuscation. We were first taken by breaking news that cryptographers made a great breakthrough with code obfuscation that could make software unhackable. Then we noticed the listing of the Symposium on Obfuscation as one of “Five Intriguing Things” of the day under the rubric “Making noise as a political act” by The Atlantic author Alexis Madrigal.

obfuscation-5Then, New York was hit by a snow storm. This allowed all of our participants to experience first hand the puddle battles that often obfuscate the end of a sidewalk and the beginning of a street. Finally, we were able to welcome most of our wonderful participants to a room warmed by their presence as well as a beautiful fireplace, a grand piano (which hosted a Gershwin session), camouflage ornaments and a quirky collection of table lamps. All these little things contributed to the build up of the day around its difficult to pronounce topic.

Having below freezing temparatures surely played a role in pulling our participants closer, too. However, the accompanying snow also meant that especially some of the people who were coming from far away could not join us. Most importantly, we lost one of our speakers to the mayhem caused by cancelled flights and trains: Hanna Rose Shell.

obfuscation-47Hanna had intended to unfold for us some of the intricacies of the development of camouflage throughout its civilian and military history. In an attempt to make up for her absence, we screened Hanna’s short film project “phenomenology of camouflage” during the session on “Clouded Visions”. Finn Brunton and Erica Robles Anderson then shortly contextualized Hanna’s work. They provided short descriptions of some of the interplay between forms of camouflage and photographic reconnaissance, a parallel pattern that occurs also in Claudia Diaz’s overview of obfuscation methods against a “strategic adversary”. Showing Hanna’s work also allowed us to put the camouflage-deco in the room into context, helping us also in avoiding looking like a local military summit.
Nevertheless, the continuity between camouflage, visual aesthetics, military history, civilian tactics, corporate strategies and obfuscation was evident in many of the sessions. Departing from a strictly militaristic reading, Finn Brunton brought to the discussion that obfuscation could be seen as a “weapon of the weak”, a phrase adapted from James Scott’s inspiring work on everyday resistance practiced by peasants in Malaysia. In doing so, Finn framed obfuscation as a mode of resistance available to those at the wrong end of the data asymmetries.
What is valuable about data obfuscation, in comparison to other heavy duty technical privacy measures like encryption, is that it is “often haphazard and piecemeal, creating only a temporary window of liberty or a certain amount of reasonable doubt“. Obfuscation is a relatively easy way to push back against matrices of digital tracking and profiling by introducing some noise, slight twists, or multiple renderings to “truth”. More on the beautiful analogies and metaphors that Finn Brunton painted throughout the day can be found in Melissa Gregg’s blog post on the Symposium.

 

obfuscation-22While most participants agreed with the potentials of obfuscation as a weapon of the weak, it quickly became clear that obfuscation could just as well increase the vulnerability of the weak, or as easily lend itself to the “strong”. Claudia Diaz showed how a website that a user obfuscates towards could distinguish “real” from “fake” queries using statistical methods. She also underlined how the secrecy of profiling practices, and hence the intransparency of the grammar with which databases categorize populations, makes data obfuscation a form of resistance with random results. Her proposal to model better obfuscation tools in response to a “strategic adversary” made clear that very much like technical privacy solutions, obfuscation may also easily turn into a complicated competition between the “machines that protect” and the “machines that detect”.

obfuscation-10Using website “privacy policies”, Joseph Turow demonstrated that, when it comes to online privacy, not only the weak, but also the organizations collecting data had discovered obfuscation as a tactic. Günes Acar on the other hand showed two ways in which obfuscation was invoked in online tracking methods used by the behavioral advertisement and online fraud industry. First, web based device fingerprinting, i.e., the ability to identify a user’s machine from a collection of its attributes, could be utilized by these companies to to counter any attempt by users to obfuscate their activities. At the same time, the tracking companies adopted “code obfuscation”, the art of coding in a manner that is still machine readable but difficult for humans to read, to conceal their tracking activities. While such intrusive tracking practices reflect the tensions over obfuscation as a protective strategy on the web, Günes nevertheless argued that obfuscating user and machine attributes still remains an important strategy to explore against device fingerprinting.

obfuscation-38Laura Kurgan brought us back to a visual world that is created by data and its analysis. She was able to suture the military use of obfuscation in technologies of aerial representation like satellites, their ability to provide powerful images which allow its audience to see certain things, while obfuscating others. As these expensive and powerful technologies of visual observation become available to civilians, activists, artists and journalists, the politics of obfuscation change, and the stories of visual data can be contested.

Laura’s presentation not only opened a path into the politics of data and its interpretation, but also invited us to consider data aesthetics. Nick Montfort followed suit and was able to ground the discussion in the aesthetics of obfuscation, as he demonstrated “Just Another Perl Hacker” (JAPH). In what could be described as a performance in debugging code for the sake of aesthetic appreciation, Nick argued that “code obfuscation” as practiced by JAPH and International Obfuscated C Contest (CCC) communities differ from those discussed by Günes Acar. These programmers have no functional or utilitarian scruple, or even one of resistance. Rather, assuming these things are separable, the participants of obfuscated code competitions have a deep desire for the aesthetics of obfuscation. Inspired also by Nick Montfort’s intervention, Karen Levy’s short piece below on the Symposium takes the argument a step further and suggests that aesthetics and play can, in fact, be important considerations in reading resistance to surveillance.

obfuscation-44Susan Stryker was the final speaker of the day and invited participants to revisit some of the assumptions in the way we discussed obfuscation. As part of her talk, Susan showed the first part of her documentary film “Christine in the Cutting Room” which depicted Christine Jorgensen, “the bomb [that was] dropped on the gender system that blew up the body’s meaning, […] the destroyer of binaries for a world split in two”. Christine’s story as a transsexual spectacle embedded at the heart of the cold war underlined that a normative understanding of obfuscating one’s binary gender identity, e.g., the ability to misrepresent one’s gender as (fe)male, as a liberatory act needs to be turned upside down. Instead, Susan argued, if we regard gender as code and transgender as obfuscation, this could allow us to think of obfuscation both as a way of challenging the idea of stable and representable identities, but also as a source of great vulnerability.

obfuscation-36In addition to the plenary sessions, we had two workshop sessions during which three obfuscation tools and games, Anonymouth, Vortex and Ad Nauseam were discussed. The tool makers, Rachel Greenstadt, Rachel Law and Daniel Howe were accompanied by Elana Zeide, danah boyd, Mic Bowman, Ted Byfield, McKenzie Wark, Vincent Toubiana, Ken Anderson, Jamie “Skye” Bianco, Ed Felten, Joe Bonneau, Lakshmi Subramanian and Alex Campolo, who were designated as discussion ringers. During the sessions, the tools were critically assessed. The discussion included assumptions the tool makers made with respect to users, the difference in obfuscating for machines vs. for humans, discussions on ethics of funding sources, how to sustain an independent tool with subversive objectives, the legality and the security of the proposed tools, as well as their future potentials.

obfuscation-26

We ended the day with a discussion which integrated all the different threads from the whole day. Some challenging and exciting questions were raised. For example: Obfuscation is often described using sensory metaphors, e.g., noise, legibility; how can we assess whether some of these metaphors restrict us in our thinking about obfuscation? What values are ascribed to consistency and consistent identities when we talk about obfuscation? What can we learn from the process of (code) de-obfuscation about the code itself and how it functions? What if obfuscation becomes a normative activity? How far do the database holders care about the quality of data, and how likely is it that they may be able to sustain obfuscated data just as they sustain errors in the database? What are the economics of obfuscation? And, if we don’t know how we are legible through databases, where does obfuscation of profiles get us?

The list of questions and the points made were much longer, suggesting that obfuscation as a strategy to protect or protest has great potential and deserves further critical scrutiny. We are absolutely indebted to our speakers for this fulfilling result. They were engaged prior to and during the symposium and seeded creative, welcoming and reflective discussions that allowed us to collectively explore the horizons of obfuscation. We certainly hope that the valuable participants of the symposium enjoyed the Symposium on Obfuscation as much as we enjoyed their presence and engagement. Even more importantly, we hope that all of those present in the room continue to think of obfuscation in its many manifestations in our lives and continue to search for its potentials and limitations.

A final word has to be said for those who were less visible and yet crucial in making this event happen. The Symposium on Obfuscation was only possible due to the generosity of our sponsors: MCC, ILI, Intel ISTC-Social, and NYU-Poly; our workshop co-organizer Carl DiSalvo; our moderators: Malte Ziewitz, Heather Patterson, Joris Van Hoboken; our program committee: Helen Nissenbaum, Erica Robles-Anderson, Finn Brunton, and myself;  and most importantly the organizing team made up of Alex Campolo, Jamie Schuler, Nicole Arzt, Emily Goldsher-Diamond, and our photographer Alp Ilyas Klanten. We are greatful to all of them for all their contributions to making the Symposium on Obfuscation a wonderful gathering.

Obfuscation and Resistance

Karen Levy, 18 February 2014

A topic is worthy of study if, when you begin to study it, you end up with many more questions than answers. This is without doubt true of obfuscation, as last week’s Symposium confirmed. Saturday’s gathering of lawyers, sociologists, artists, technologists, gender scholars, anthropologists, and computer scientists (among many others) brought to the fore a multitude of new perspectives on obfuscation: what it looks like, who practices it, and what purposes it serves.

 

obfuscation-8

For me, the Symposium brought to mind new ways of thinking about resistance more broadly, a topic I’ve been struggling to tackle lately in my own work. In my research, I’m exploring the tools and practices truck drivers employ to violate federal laws that limit their driving time – laws that are increasingly enforced via electronic monitors installed in their trucks. The more truckers drive, the more money they make, so they’re motivated to break the law, even when it leads them past the point of exhaustion. I’ve been thinking lately about how to make sense of the wide variety of mechanical, organizational, and social means drivers use to resist control over their behavior. The tactics truckers use seem to have very different flavors – smashing an electronic monitor with a hammer, for instance, seems qualitatively different than quitting a company, or than driving under multiple driver profiles – but I’ve been coming up short in my thinking about what the real differences are here and what they mean.

The Obfuscation Symposium was a lightbulb for me. One of the key issues that animated our discussions was: what is obfuscation for? What are its goals? When can it be deemed a success? Different participants addressed these questions in different registers. For Joe Turow, obfuscation is a tactic used by the powerful to reinscribe their power, a boundary object that separates those “in the know” (here, corporations and their lawyers) from outsiders. (Bourdieu would have a similar reading.) Or obfuscation can be an act of protest that confronts economic processes by upending algorithmic assumptions about patterns of individual behavior, as tools like Ad Nauseam illustrate. Obfuscation can be an act of play, a game of plain-sight hide-and-seek, like the code poems Nick Montfort spoke about. Or maybe, as Susan Stryker suggested, obfuscation can be understood as having to do with individual identity and self-expression, the “codes” we all bring to our daily interactions with one another.

obfuscation-23

Among these possibilities, there are major axes of variation: whether obfuscation is intended to be noticed or under-the-radar, to bring about institutional change or to be an end in itself, to reconstitute the self or to be a collective endeavor. And I think these variations are applicable to resistance more generally; like any tool, it means different things for different people at different times. So this has been helpful for me in thinking through my truckers, and explains some of the variation I’m seeing. For a trucker smashing his monitor with a hammer, that form of resistance is an act of self-expression that reasserts his occupational identity; for a trucker hacking around a permission restriction to access the solitaire game on his unit, resistance is an act of play; and for a trucker surreptitiously editing data streams to eke out some more work time, resistance is a covert means toward accomplishing ends of economic necessity. Blanketing all of these actions under the term “resistance,” without digging deeper, obscures the very different things that are happening here. By calling these variations to light, the Symposium was of great value to me.

Slides and Visuals

Günes Acar

Claudia Diaz

Rachel Greenstadt

Joseph Turow  

The JAPH (Just Another Perl Hacker) that Nick Montfort presented at the Symposium:

$ _=
 qq
.CG
T--A
A---T
A----T
 C----G
  T----A
   A---T
    G--C
     CG
     CG
    C--G
   G---C
  G----C
 C----G
A----T
C---G
G--C
 AT
 CG
A--T
A---T
G----C
 A----T
  G----C
   C---G
    A--T
     GC
     CG
    G--C
   A---T
  G----C
 G----C
G----C
A---T
T--A
 AT
 CG
T--A
C---G
A----T
 A----T
  G----C
   A---T
    A--T
     CG
     TA
    T--A
   G---C
  G----C
 C----G
A----T
C---G
G--C
 CG
 TA
T--A
C---G
T----A
 C----G
  A----T
   C---G
    G--C
     GC
     TA
    G--C
   C---G
  G----C
 G----C
G----C
A---T
T--A
 CG
 AT
G--C
A---T
A----T
 C----G
  C----G
   A---T
    A--T
     CG
     CG
    G--C
   G---C
  G----C
 A----T
T----A
C---G
G--C
 CG
 TA
A--T
A---T
G----C
 A----T
  A----T
   C---G
    A--T
     GC
     TA
    G--C
   C---G
  T----A
 G----C
G----C
C---G
T--A
 AT
 CG
G--C
G---C
T----A
 C----G
  G----C
   C---G
    C--G
     CG
     AT
    T--A
   C---G
  G----C
 G----C
A----T
T---A
G--C
 CG
 TA
G--C
A---T
G----C
 A----T
  G----C
  .;@_{A
    => C
     =>
     G=>
    T=>}
   =0..3
  ;s;. *
 (\w).*
(\w).*
\n;$_
{ $-
 ++
 /9
%2?$
2:$ 1
};gex;
 s;(.)(
  .)(.)(.
   );chr
    64*$
     1+
     16
    *$2+
   4  *$
  3   +$
 4 ;gex
; eval

Leave a Reply

Your email address will not be published. Required fields are marked *

*