Scherffig, Lasse (2018) “There is No Interface (Without a User). A Cybernetic Perspective on Interaction”
Scherffig traces a genealogy of interactivity to argue against understandings of interactivity as teleologically moving towards technological advancement and naturalization. He seems to argue against the idea that humans design interfaces, while also arguing against the idea that interfaces design humans. Rather, he seems to understand interactivity as a dynamic feedback system between these two positions of human user and machine. Interfaces, then, are better understood for the processes of interaction that they mediate, rather than the way they serve the user or shape the user exclusively.
Scherffig's genealogy demonstrates different understandings of interactivity as they occurred at different moments of time, relative to the devices of human-machine interaction being developed. Scherffig not only demonstrates the instability surrounding meanings of "interactivity", but also supports his position about how humans "don't actually" design interfaces. The interfaces being "designed" aren't actually registering and responding to the user's desires. Rather, the interface responds to a particular system-specific action taken by the user.
Here is where I find my own critique of his argument: his understanding of "design" should be more nuanced to reflect the richness of his historical evidence. As I understand it, his evidence seems to support the position that humans do indeed design interfaces - but design here doesn't necessarily mean "control". We "design" interfaces in the way that we "design" data - we construct it in a way that is intelligible to us. As Scherffig states later in the article, we don't ever really deal with "raw" interaction or "raw" data - we only work with interfaces and data by making them functional for us.
MacCormick (2012) “What is Computable?”
MacCormick argues that not all things are computable. Specifically, he argues that it is impossible to produce a computer program that can predict all possible errors in other programs. While he proves his argument through a thought exercise (which was certainly a fun read), I can't help but feel that his argument should have been attached to a larger cultural/historical analysis (similar to the genealogies produced by Nunes or Scherffig). Aside from the impossibility of building such an all-knowing, bug-detecting program, we should consider how an "error" might have been understood in different time periods, in different places, and amongst different users. If such a program is expected to exist for a long period of time and is expected to be used by different people in different spaces, it would certainly be impossible to predict what kinds of inputs and errors would occur in all of those different instances (let alone what would be reported as an error). Separately, it's interesting to think about a genre of video games in relation to MacCormick's argument: coding games. These are games where the player progresses through the game either by fixing the game's code or deliberately breaking that code. Not only is the prediction of all errors impossible, but that very impossibility has become a concept of play!
Nunes, Mark. (2011) “Error, Noise, and Potential: The Outside of Purpose”
Nunes argues that error (occurring within infrastructural contexts) provides multiple points of utility, some of which are contradictory. Error reveals ideologies and systematic logics of underlying infrastructures. Error can therefore act as a useful point of intervention for political resistance or creative production. At the same time, dominant ideologies may take these points of error and minimize them (perhaps even naturalize them). Error is coded as controllable. Even more interesting, Nunez suggests that dominant capitalist ideologies can take cases of errors and uphold them as reason for increased control over information production and dissemination.
In his genealogy of error, Nunes directs us to a double meaning of error: deviation from a path that is contained within a larger system of thought, and a kind of deviation (or "wandering") that challenges the intent of the path (or the logic of the larger system) to begin with. While this duality is certainly thought provoking, I think it needed to be accompanied by a nuanced discussion of "functionality". What exactly is this "defined path" against which error is being defined? Does the meaning of this "defined path" change in different contexts? Much of this discussion was a bit too abstract.
I'm not being nit-picky here: it's important to be specific when defining errors and functionalities. In response to the error-based media practices discussed, my main critique is that some of these practices (as well as this book introduction) flattens all errors. That can get dangerous and problematic pretty quickly if we're equating data errors with book errors with body "errors" with sexuality "errors", and so on. (A similar critique has been made of object-oriented philosophies generally.)
Menkman, Rosa (2011). “Glitch Studies Manifesto”
The Glitch Studies Manifesto provides an interesting dogma/methodology for studying visual and aural interruption. However, similar to some of the other readings assigned for this week, I can't help but feel that this manifesto could provide a richer, media-theory practice if it was attached to more specificity. Similar to the Nunes piece, I'm left wondering exactly what "glitch" means. As previously mentioned, it's very important to think about what we're coding as "error" or "glitch", and what we're coding as "functionality" or "flow". The manifesto seems to openly allow for contradictions within understandings of "glitch", but those contradictions need to be located in specific historical/cultural contexts to be meaningful.
Hamish Robertson and Joanne Travaglia (2015) “Big data problems we face today can be traced to thesocial ordering practices of the 19th century.”
Robertson and Travaglia argue that the problems of the big data revolution happening today have their basis in an earlier data revolution from nineteenth century Europe. Interestingly, their analysis isn't entirely based on problems of data collection from the nineteenth century, but rather tends to focus more on disciplinary (as in academic disciplines) formations surrounding information. The focus on disciplinary formations is important, as Robertson and Travaglia seem to suggest that the disciplinary boundaries surrounding data are central for understanding their contemporary problems (problems, which, never seem to be clearly defined). They seem to take issue with the understanding of data as scientific rather than social. Perhaps they're calling for understandings of data as less neat and precise, and more messy and caught up in its social/historical/cultural contingencies?
Cadwalladr, Carole (04 Dec 2016) "Google, democracy and the truth about internet search”
Cadwallader argues that the algorithms behind Google's search engine work to enable the spread of alt-right ideology and hate speech. As Cadwallader seems to be arguing in the position that Google normally operates as a neutral tool for finding information, I feel compelled to ask if the alt-right search results that come up on Google count as a "glitch" or an "error"? In the other readings assigned for this week, there was much discussion of "glitch" and "error" as points of disruption to celebrate and use for creative resistance. But, as these readings did not define "functionality" or "error" very clearly, I feel compelled to press on whether "sudden" alt-right visibility counts as such a "glitch". Further, I'm not sure how to evaluate the sense of alarm in this article. Google, let alone the internet, were never neutral ground to begin with. The beliefs of the alt-right have been visible to some of us for a while - they're directed at some bodies more than others. Perhaps the "glitch" in Google's search engine algorithms is that those hateful opinions have been made visible to people who normally don't have to concern themselves with such intense hate speech?