The Function of the Interface

This is the third research note from Anand Kumar Jha, one of the short-term social media research fellows at The Sarai Programme.


Ross Lovegrove - Bottle
A Bottle designed by Ross Lovegrove.
Credits: Evan La Ruffa.


Affordance as the interface as the function

The first set of tools made by the cave dwellers appears very hazy while typing the closing braces on the Terminal [1].

Our usage environments consist of products like spoon and knife. The concaveness in the spoon suggests the dipping end and the flatness suggests the holding end. A knife speaks similarly to our cognition with the cutting edge being thinner than the handle. We often need one sense such as touch or two senses such as touch and vision to resolve the complexity of interaction with the tools around us. Once we are able to use a spoon, we know what to do with a ladle or a scoop. Mind recognizes patterns and interpolates utility. Most of the objects, conceived pre-industrial revolution present a remarkable simplicity in their usage. While one admires the creative complexity in a device like pulley, it is not as intuitive as a spoon. The material dictating its handling and thus affecting the form of the product to create these specific interaction points called the affordance, became a prominent explanation for tools based on non abstract use cases. Post industrial revolution the world of the pulley grew and spur gears, chains, conveyor belts and pistons happened. One could safely park them 500-600 years away from the birth of the gesture controlled gaming console. The machines like gears, shovels or pistons had handles, cranks or other extensions suggesting where and how much force should be applied. One could see what happened when the force was applied and thus understand the personality of the machine.

Machines did become complex but still bothered explaining themselves and thus could be made by anyone who had the material resources.

The “covered”, “out of sight” “Invisible” machinery and the idea of “Under the hood” gained its first ground with ford launching its T Mobile and recruiting clay artists and sculptors to make a new skin every year to cover the non evolving machine beneath it [2]. Very soon, what an automobile was and what it looked like started becoming two very different things. This is when the machines fundamentally became “two headed”. There came into being, one head that would face us and one that would face away. The machines became a meaningful systems of multiple simple machines well synchronized with each other and assembled together to serve a task mechanically complicated but ritually simple [3].

However it would be dangerous for us to assume that this was the only direction of growth for the machines. Manuel Castells articulates this very beautifully in his book “The Rise of the Network Society” [4]. He talks about the experience of time through the practice of life. Historically building the argument, he refers to the ability of the feudal class to manage, lineate and thus in multiple ways “own” time. Contrast this against the peasantry’s dependency on agrarian cycles or a craftsman’s dependency on natural processes of smelting and casting objects. Under capitalism, managing time continued to be the privilege of a certain class. One can see the direction of industrialization in two axis, automation (which is making objects with the help of machine, faster than a human would) and scale (making more objects in the same timeframe). Both of these axis are chasing time, constantly doing more and more in the same timeframe and thus commodifying it to an extent where the organizations started paying workers based on their billing cycles, the one who sat at the top of the pyramid of scale and automation (deciding for millions of workbenches earned more and the one closer to the machine and managing one work bench. To quote him:

Under capitalism, time became money, as the rate of turnover of capital became a paramount form of profit-making. The faster you could secure your return, and the faster you could reinvest it, the greater the profits to be made. Finance became constructed around the sale of monetized time. Credit was based on time. Speed became essential in financial transactions. The more capitalism went global, the more differences in time zones made possible the proliferation of interdependent financial markets to ensure the movement of capital around the clock. And so, a new form of time emerged in the financial markets, characterized by the compression of time to fractions of a second in financial transactions by using powerful computers and advanced telecommunication networks. Furthermore, the future was colonized, packaged, and sold as bets on future valuation, and as options between various future scenarios. Time as sequence was replaced by different trajectories of imagined time that were assigned market values. There was a relentless trend towards the annihilation of time as an orderly sequence, either by compression to the limit or by the blurring of the sequence between different shapes of future events. The clock time of the industrial age is being gradually replaced by what I conceptualized as timeless time.

The comparative commodification of time with various kind of commodities, agrarian, industrial, financial and predictive (futures trading etc) clearly tells us of the scale and automation of the capital distribution(who makes money), production of value (the tangible to intangible) and people who engage in production of value(for example: farmer, welder, trader and big data specialist). We see the interface moving from a handle in a sickle and a hammer to turner in a lathe and to multiple components in a furnace to a trading panel selling agro-based shares and debentures to a dashboard aggregating the data about the rainfall, the availability of the agricultural labor, the migration trends, the global fluctuation in commodities markets and multitude of such complex factors and then telling us how much would a Kilogram of wheat fetch in next year winters.

The interface is not just hiding the complexity of the system; it is also hiding the politics of the system. This now helps us come to the question, “what is the function of the interface”.

Design and the execution of the interface agenda

The Interface, unlike The Cinema, does not have an emotional affordance [5]. As Galloway points out, one experiences an emotional peak(laughing or crying or similar behaviors) while watching a film; however one experiences no such emotion while working on an interface.

Make no mistake, both of these are coded systems. They are architected in a way to manipulate human behavior and curate a precise response from the audience; Interface more than the Cinema. Interface design derives its methods extensively from ethnographic research to produce user representations known as personas. Personas define the requirements that the product must meet and the acceptance criteria of the solutions proposed [6]. Product usage scenarios often called task-flows or work-flows reveal the situations under which the product would be used and the specific things user would do in a sequence to accomplish a task on the Interface. An artifact called the storyboard captures the detailed workflow sequences. This process is also used extensively in Cinema and Comic Book/ Graphic Novel Industry to document frame by frame narratives often from Camera angles, coded from a first, second, or a third person perspective [7]. Part of the design of the interface (user research) is to agree to the mental model of the user , contradicting is the objective of visual design to pitch a perfect rhetoric to them. At times, it is very difficult to find out if the user found the product useful or was attracted to it, both of which could result into a purchase.

There are however, two interesting differences between the Interface and other mediums like cinema and comic. Interface is neither produced, nor consumed linearly. It may use techniques from storytelling but it behaves very similar to Architecture. It has a structure and structure is more often hierarchical. A user navigates this structure based on their motivation towards visiting the site. Occasionally user is in with no motivation beyond browsing and skips the “ Primary Navigation” and goes the “Search bar” way, feeling privileged about the well curated choices meeting him/her at those exact moments. This brings me to the Second point, which is the power equation between the media and its consumer. Cinema has a passive feedback loop. It gets watched, it does not watch. While the projector runs, there is no feedback loop from the viewer. The only way a viewer can register his/her displeasure is by not repeating the show and by discouraging others from watching it.

The Interface behaves like an intelligent slave [8]

The Interface is crafted not in accordance with the model of machine, it is crafted with the user’s mental model [9]. It is done so that the user doesn’t have to bother with understanding how the machine functions to make it work. An interesting example that is given is that the user does not need to know the Page–Rank Algorithm (named after Larry Page, the founder of Google) to do a Google search. The closer the interface to the user’s mental model the more invisible the machine becomes, almost similar to the servants in the feudal palaces who appear and disappear at will of their masters. The interface is intuitive and convenient and helps you accomplish what you set out for.

Let’s get back to the Architecture analogy for a bit. While the user remains empowered to exercise the choice by navigating through an application at his/her will, the fact that this “will” was the study material behind the construction of such a structure, renders the idea of “free will” doubtful. Precision oriented methods such as EEG devices [10] and Eye tracking [11] involved in testing the usability and the cognitive mapping of web and mobile interfaces, tell volumes about the sophistication of such information structures, commonly referred to as Information Architecture [12]. Interface Studies researcher Johanna Drucker [13] and several User Experience practitioners such as Alan Cooper [14] and Jesse James Garette [15] cite Gestalt’s principles [16], Hick’s law [17] and Fitts Law [18] as theoretical cornerstones of Usability, providing the much needed academic validation to the method. The next obvious question is, so what does the “Slave Master” Interface do apart from being user friendly.

The days of information sites are limited. Web now serves websites as software application, media platform, a retail store selling millions of products, accounting and management system, interactive map and multitude of other things. What sits at the backend of this, the second face of the product that faces away, is what is dictating the behavior and the agenda of the interface.

With Interface as with the Image, the user finds a public experience private. User consumes multiple images, shaping his /her idea of the Image and thus the motivation to produce one. The image thus produced is routed back to the automated funnel and finds itself being shown to another user. All Interfaces have been the point of access to the products/systems/services. Every interaction between the system and the user (and often between one component of the system and another) is recorded as a log or any other format often for functional, legal or operational reasons. This data was essentially stored in Data Warehouses(read large centralized servers) and was limited in volume. It is now much bigger considering millions of people on social networks interacting with each other, sensor environments generating huge amount of man-machine interaction data and the data that people voluntarily give to be a part of a certain digital ecosystem. Now handled by millions of distributed computers,this data has ceased to be a headache and has instead become something that can be sliced and diced to get very rich insights on how people interact with machines(physical or virtual). These insights are increasingly being automatically fed back into the system to train machines and make them more intelligent to manipulate user behavior for better results [19]. The Interface is now a constantly evolving highly customized artifact that learns for user behavior analyzed against the data generated by billions of similar users to get that one click that seals the deal. User is still happy to see a easily accessible thoughtful and resourceful slave in the Interface. The Interface is happy to see yet another statistical data point, or a cookie behaving the same way as other data points in a classified cohort.

This is not limited to the end consumer feedback loop. The Vector of the Interface is revealed in how corporations roll out APIs (Application Programming Interface) [20] that are used by software developers. These APIs have pre existing code libraries and a user interface that simplifies writing the code and debugging it. These APIs are a little ahead of the “Slave Master” battle in that sense that apart from manipulating user behavior, they also manipulate what will the end result of user behavior be. To explain it further, a developer has little agency in how the application/product will evolve when all building blocks and framework are pre-established to dictate the course of product evolution [21]. This is generally packaged and sold as an approach where the user must not reinvent the wheel and build over the existing logical and machine stack to create applications of extreme complexity and overarching utility.

Juxtaposing the virtual space against the physical space, and building over Michel De Certeau’s walking the city [22] (and in the act deconstructing its methods of production of space), Galloway talks about walking through the virtual space and brings out the idea of access. Interface presents a completely different face of the space, not just creating another presentation layer but also refusing access to anyone who does not trace the Interface path.

Like a double filter for the camera which edits time with a snap and space with a frame [23] the interface zooms at the user first through the usable and data driven interface exercising a very similar two level filtering process. At the end of that experience, user participates in giving more data, receiving a higher set of customized offers. In the process of consuming those offers user further gives in more data.

Similar to a Selfie, the “Slave Master Interface” mirrors every user intention keeping the advanced data driven segmentation away from the user’s thought and sight. Every new page on the browser that claims to set you free leaves a cookie at your desktop, innocently trying to customize its offering while secretly leaving a camera at your doorstep that you would willingly pose to, everyday.

Interface is the new Camera.

What follows

The next post will contain the analysis of interviews with three leading data scientists (one of them also being a image processing expert) and a user interface Designer as a part of primary research. Transcripts will be made available during the presentation.


[01] Peter Modenson. Using the Terminal. Ubuntu. Retrieved from

[02] Doblin, Jay. 1970. One Hundred Great Product Designs. Van Nostrand Reinhold Co.

[03] Things like washing clothes, driving, cooking etc which formed the routine ritual were now trusted to machines. Tasks like these appear simple from a human mental model but remain mechanically and electronically complicated. I call these tasks ritually simple.

[04] Castells, Manuel. 2011. The Rise of the Network Society: The Information Age – Economy, Society, and Culture, Volume 1. John Wiley & Sons.

[05] Galloway, Alexander R. 2012. The Interface Effect. Polity.

[06] Cooper, Alan, Robert Reimann, and David Cronin. 2007. About Face 3: The Essentials of Interaction Design. John Wiley & Sons.

[07] Drucker, Johanna. 2011. Humanities Approaches to Interface Theory. Culture Machine. 12(0). Pp. 1-20.

[08] See [5].

[09] See [6].

[10] Hirshfield, Leanne M., Erin Treacy Solovey, Audrey Girouard, James Kebinger, Robert JK Jacob, Angelo Sassaroli, and Sergio Fantini. 2009. Brain Measurement for Usability Testing and Adaptive Interfaces: An Example of Uncovering Syntactic Workload with Functional Near Infrared Spectroscopy. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. ACM. Pp. 2185-2194.

[11] Poole, Alex, and Linden J. Ball. 2006. Eye Tracking in HCI and Usability Research. In Encyclopedia of Human Computer Interaction 1. Pp. 211-219.

[12] Rosenfeld, Louis, and Peter Morville. 2002. Information Architecture for the World Wide Web. O’Reilly Media Inc.

[13] Wikipedia. Johanna Drucker. Retrieved from

[14] Wikipedia. Alan Cooper. Retrieved from

[15] Wikipedia. Jesse James Garrett. Retrieved from

[16] Koffka, Kurt. 2013. Principles of Gestalt psychology. Routledge.

[17] Roberts, Richard D., Helen C. Beh, and Lazar Stankov. 1988. Hick’s Law, Competing-Task Performance, and Intelligence. Intelligence. 12(2). Pp. 111-130.

[18] Wikipedia. Fitt’s Law. Retrieved from

[19] Russom, Philip. 2011. Big Data Analytics. TDWI Best Practices Report, Fourth Quarter.

[20] Wikipedia. API. Retrieved from

[21] See [5].

[22] De Certeau, Michel. 1984. The Practice of Everyday Life. Berkeley, USA: University of California Press

[23] Jha, Anand Kumar. 2014. The Camera Membrane. Sarai. August 01. Retrieved from

Share this:


Print Print

Categorised in: ,
Published on: September 5, 2014

No comments yet. Please leave a comment using the form below.

Leave a Reply

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

Comments Policy: We look forward to your comments on the posts published on this website. The comments should be constructive and conversational, as opposed to being offensive or merely critical. Personal attacks and rudeness are absolutely not tolerated. As the Editors of the website approve comments before they are published, there can be a slight delay in their appearance on the pages, especially during weekends. The comments are published under Creative Commons Attribution 2.5 India license. For any clarification, write to us at dak[at]sarai[dot]net.