Skip navigation

Monthly Archives: May 2013

The readings from this week were interesting to consider, especially with regard to previous course readings. When reading Kirschenbaum’s “Mechanisms”, I couldn’t help but immediately think back to Bush’s concept of the memex. Kirschenbaum is critical of the evolution of the stored program in claiming that “greater and greater storage capacity will only serve to further dematerialize the media as their finite physical boundaries slip pas the point of any practical concern.” (34) Kirschenbaum is fixated on the loss of materiality in new media, while Bush on the other hand promotes the efficacies that come with advanced storage. It is an interesting dialogue between the two writers as Bush writes from a historical context in which such storage does not exist, while Kirschenbaum is writing in particular about lay computer users. Kirschenbaum uses the term “screen essentialism” to describe new media studies’ insistance on graphics and user interface as more important than the physicality of computing itself. Bush, on the other hand, seems to be a screen essentialist, focusing on the attributes of science and development that allow humans to progress in thought processes and writing based on the efficiency of computerized automation. In addition to Kirschenbaum, it was clear that Sterne’s text on the MP3 developed a similar position with regard to the importance of the format and what it means about communication and culture in general. The fact that MP3s are compressed to adjust to sharing standards and listening standards makes me wonder how much the format influences the mediality or vice versa.

The most engaging reading from this week for me was Galloway’s “Gamic Action”. Video games are so often in my generation understood as yet another form of child’s play, but Galloway really threw that assumption into relief. I appreciated the way he analyzed the terms “play” and “game” itself through examining video games as a medium as opposed to a form of “play”. When we think of standard video games such as Super Mario or Grand Theft Auto, rarely do we critically acknowledge the fact that the machine is as important to the “play” as the gaming operator is. I especially liked Galloway’s discussion of culture and how it is both manifested and developed through video gaming. The example of the arcade versus the home console and computer games really stuck with me. The fact that arcade games are specifically configured, through narratives of “lives”, to profit off of the player in short intervals demonstrates the gaming culture’s inherent connection to a “real world” economy. While you may be suspended from reality in a game, these examples make me wonder how much a game is really removed from the “real world”? How much of gaming involves secondary realities?

So I am missing a blog post, and unsure of which one, so I figure I’ll just rant a bit about some of the concepts that overflow from the space of my paper…

Big Data as precession of simulacra was not something I was able to explore fully, same to with memory as an instantiation of the same logic

Speculation

with memory and with data, the capture of information is necessarily a process of “cooking” but it is cooking for recipes undeveloped based on needs not yet determined

that we do this unconsciously in our brains is a pretty astonishing feat, we think often of the fallibility of our memories but rarely give it much credit

to what extent does the valuing process that our brains perform an instance of simulacra

i think to fairly large degree in that the brain is in fact predicting outcomes and storing accordingly

but this, like financial instrument, is a constantly compounding function, it approximates a wave

or rather it is a wave sometimes

and a particle othertimes

I just read the abstract of a paper attempting to apply quantum probability functions to humans seeming errors in judgement under uncertainty

it was above my pay grade, but i think it begs the question of weather we can understand simulacra as a wave function that doesn’t merely precede the actual, nor merely determine it, but rather embodies the spectrum of possibilities that that the actual will then instantiate

 

I didn’t really like Steal This Film. It was well-made — as a work of art, it’s impressive, and the jumpy, disconnected imagery in some shots is effective despite being a bit cliche for propaganda-themed video. However, the actual arguments were fairly weak. Most of all, the film failed to sufficiently capture my sympathy; much of it was premised on TPB’s fight for existence, which wasn’t very well established as being justified in the first place. It convincingly portrayed youth culture as increasingly OK with piracy and their elders as scared and out of touch, but this doesn’t itself make a good case for the ethics of pirating copyrighted media.

 

On the other hand, Liang’s essay was fascinating. I hadn’t considered the interaction between piracy and socioeconomic class. Indeed, I had previously been a pretty strong adherent to the model of piracy as being ethical if (and only if) it was done with no money involved. The for-profit piracy that Liang describes has always struck me as plainly wrong, but he muddies the waters quite a bit by detailing how India’s media economy is largely concerned with and even premised on forms of piracy. Liang’s comparison to illegal housing highlights the ways in which large-scale illegal activity can affect enough people and determine the bounds of enough lifestyles to make it entirely unclear whether increased obedience to the law in question would even be a net social good.

One of the oddest things about browsing Wikileaks, for me, was the banality of much of the experience. I expected it to feel exotic or daring, but the user interface — cut from a common Web-based mold — undermined much of that sense. This may be deliberate.

 

The easiest portal into browsing Wikileaks is the alphabetized list of countries on the left side of the home page. Scrolling through this list triggers, by association, memories of filling out online forms — the ones that don’t bow to American technoimperialism by putting “United States” before “Afghanistan”. The front page itself has an innocuous white and sky-blue color scheme, and the page for each sensitive state document has the instantly recognizable styling of a Wikipedia article. The part that plays best to my aesthetic expectations for cyberespionage is the SHA cryptographic signature accompanying the file.

 

In comparison, http://www.textfiles.com/ — intended to look like an old-school text terminal, and consisting of an innocuous archive of popular old text files that floated around the early Internet — is significantly more impressive. Since very few Internet users actually remember what it was like to use such an interface (and most of them never did use one), it instead conveys a Hollywood-inspired underground cyberpunk sensibility. (Sure, there’ s a category for “Anarchy and Explosives and General Mayhem”, but there’s also one for walkthroughs of text-adventures games. The latter is much larger.)

 

Of course, Assange probably doesn’t want his site to look particularly badass, given the delicate tightrope walk he executes between playing the reckless James Bond and defending the ethics and legality of his actions.

In summation of my engagement with the course, I’d like to examine my response to Professor Chun’s theme of “delightful creepiness.”

It is delightful, which is great. It is creepy, but it shouldn’t be.

I don’t want to say that there is nothing about the Internet, Big Data, AOL search leaks, drone strikes, cyberbullying/rape, or the gamification of culture that should give us cause for concern. The Internet and digital media are changing our culture, our work, our play, and our brains. Wariness is a natural response to change– it helps protect us from the sometimes negative consequences thereof. But we also can’t let it interfere with our ability to change for the better.

Big Data is exactly that: BIG. As mentioned in “Big Data– Creepy?” below, we envision some corporate stalker using the mined data to stalk us, corrupting our minds and emptying our wallets. But the fact of the matter is, each one of us is a single bit, maybe a byte’s worth of data among petabytes. It is our culture that is shaped by the algorithm, and culture has always swept us along in waves. It is now, as it has always been, our responsibility to define ourselves within the framework of that culture. The Internet, more than ever, shapes our individual experience, but we also, more than ever, are empowered to shape that experience ourselves. Websites like Kickstarter, YouTube, and Etsy allow us to be direct patrons of culture and the arts; we take ownership of that which previously belonged to those elites who cultivated their ‘taste’ and imposed it on the masses. I have found myself more and more frequently talking about programming and code in the same terms I would use to discuss magic in a fantasy novel– in this new, digital world, we have the almost-magical power to directly manipulate the fabric of reality. To break the rules. To make the rules.

Yes, there is institutional oppression in this new world. There is violence, rape, and vulnerability: the chance that our secret selves will be exposed, mocked, or– worse– bought and sold. But these things were present in the meat space as well. The digital world is the new frontier, where the real estate is limitless and any man can be a king.

“Is it creepy that corporate companies like Google and Facebook have all of this information about you (your birthday, your favorite movies, your friends’ names, and more), this big data?”

That was the question that was brought up during section.

Some people will say “Yes, it is creepy.” But why? I speculate that one of the reasons  is because we imagine that there is a person out there (in Google or Facebook) who is personally collecting the information to learn about you and provide a personalized experience — basically we have an irrational fear of having a stalker. But the difference between a stalker, an individual, who is collecting information and a corporate is that the personal data aggregated by the corporate is not analyzed by a person, but rather a computer program that uses an algorithm to personalize your experience based on the given data. To a computer program, they are just statistics that is not matched to a face or a personality.

It is only when a person, a stranger, decides to use the information stored in Google or Facebook that the creep factor comes in. It is the person who tries to collect the stored big data for personal use who is creepy, not the corporates.

Boston Marathon Suspect & Sunil Tripathi

The left picture is a photo of one of the suspects for the Boston Marathon incident captured by the surveillance camera and the on the right was the photo of a missing Brown University student, Sunil Tripathi. As you can see, the picture on the left had very poor resolution and one can only guess who the person was. With news about Sunil Tripathi circulating, a Redditor eager to help with identifying the suspect jumped the gun and created a false connection between the two. The Reddit post pointing out the false identification blew up online and thus began the witch hunt for Sunil Tripathi. What is even worse, is that newspapers antsy for any kind of update with regards to the issue took the unconfirmed information as truth and spread the false word.

Later, it was revealed that Sunil Tripathi was completely unrelated and Reddit issued a public apology. But the poor resolution of the image in combination of high tension created an unpleasant situation for Sunil’s family during that short time period of Sunil’s witch hunt. So do we blame the poor resolution of the image for the “prosecution” of Sunil Tripathi? The answer is an obvious no, but this gives some evidence as to how image resolution can affect the people. We as a society depend on the notion of “Seeing is believing” and when what you see is unclear, people can draw to the wrong conclusion without proof in order to do away with the uncertainty. Thus, we should strive to do away with images of poor resolution, not only for aesthetic, but also for clarity.

In Manovich’s Whither Internet Control, the following passage was the most thought provoking:

Google already bases the ads that it shows us on our searches and the text of our e-mails; Facebook aspires to makes its ads much more fine-grained, taking into account what kind of content that we have previously “liked” on other sites and what our friends are “liking” and buying online. Imagine censorship systems that are as detailed and as fine-tuned to their “users” (targets) as the behavioral advertising that we now see every day. The only difference between the two is that one system learns everything about us in order to show us more relevant advertisements, while the other one learns everything about us in order to ban us from accessing relevant pages.

It was shocking to think that even now, we are only a small step to losing our Internet freedom due to the amount of information we publicize. But I think that was the general point of Whither Internet Control: to enlighten the readers of how Internet doesn’t just mean democracy, it could also easily lead to a tool for government censorship and control. But I don’t think this means we have to reject the Internet for the fear of the possibility of the government re-purposing the Internet for authoritarian uses. Many things can be used for “good” and “evil” and it just depends on how the people decide to use it.

Lev Manovich’s “Navigable Space” provides interesting insight to the way we interact with new media on a virtually spatial level. The most interesting part of his argument comes through his analysis of space as a cultural concept. Discussions of art history and comparative historical spatial perspectives in aesthetics demonstrate that the way we conceive of and use virtual space probably has cultural connotations and implications. His discussion of space through the lens of the narrative also works to provide a cultural insight to virtual space. Manovich’s discussion of the American narrative at navigable through physical space and nature versus the European narrative as navigable through physical and psychological interaction with others demonstrates reasoning behind why virtual space in video games, for example, are so representative of the physically “navigable” quality of the American narrative. I especially liked how Manovich discusses the separation between things and space as “haptic” and does a good job of explaining why computer space is haptic and aggregate. It makes me think of the notion of websites as “sites”; separate locations. However I can also see how space on the web can be considered as infinite and therefore cannot be isolated from the things it contains or provides (e.g. websites). Regardless, it is the cultural analysis of navigation that allows Manovich’s text on computer space to truly make its mark in new media studies.