The eventual theoretical foundation of Internet Studies ™ combines the collapse of ontology with an integrated and consistent set of nudges and an active and self-sustaining community of spimes. Let’s call these the Three Pillars of the Internet Age. These pillars are bound together by what I will call a participatory framework. Internet studies differ from other “studies” disciplines (media studies, gender studies, etc) in that the protocols which govern the interactions between entities within a participatory framework are well-defined, and in most cases are explicit and formal (for instance, IP describes (at some level of analysis) the communication between all networked objects). Exchanges between entities within the framework are interactive, interoperable, and cooperative, and hence they are participatory. Internet studies is also far more interested with the possibilities made available by the infrastructure that supports the participatory framework, than in any particularly realization of those possibilities. For instance, Internet Studies is interested in the question, “what is a blog?”, and what kinds of communication, social organization, and information distribution possibilities that this kind of resource makes available, and is less interested in a question like “How has DKos changed the political climate in 2008?” which in some sense is merely a specific application of the more general social protocol.
I’ll talk just a bit more about the three pillars below.
Pillar One: Everything Is Miscellaneous
Collapsing ontology (more specifically, collapsing the distinction between data and metadata) as described by Weinberger and Shirky yields a minimalist ontology that is unsustainable by human minds alone. This in itself is nothing new; we have always used external frameworks for structuring our knowledge. The organization of libraries is a paradigmatic case of using external resources (shelves, numbering systems) to help structure and support our organizational techniques.
What changes with the a mature internet is that the methods for retrieving data from this organizational scheme is also unsustainable by humans. Shelves and numbering systems were optimally practical and effective for human researchers who have to dive through stacks of books to look for specific information, and were designed with human organization and retrieval in mind. On this system, the dramatic ontological leveling accomplished by the internet seems to pose a radical danger for the members of that system, since it yields no straightforward, intuitive paths for organization and retrieval, and therefore no methods for developing a mastery of the material so organized.
But in the Internet Age, humans are not the primary data retrievers. If everything is miscellaneous, only powerful search engines are able to sort through piles of information to retrieve the relevant data. Humans must depend on the ability of search engines to manage data retrieval in an ontologically minimal, densely interconnected semantic web, and the search engine must develop some criteria for determining relevance for a given search. Therefore, this is not a one-way dependence; it is dynamic in the sense that a search engine’s criteria of relevance will be based on feedback from users as they continue to update and enrich the structure of the semantic web. Here we see the first instance of the participatory framework necessary to support the Internet Age. Internet users and search engines dynamically depend on the cooperative interactivity of each other. The beginning of the Internet Age can roughly be dated to the popularization of Google.
Pillar Two: Nudges
The hallmark of the Internet Age is the automation of data retrieval, and consequently humans are no longer responsible for the organization of data. In a sense, this means humans are no longer the primary agents of scientific progress; although they play a central role, the primary work of structuring knowledge is handled by our machines. Humans therefore have two reduced but no less important roles to play in the Internet Age: they are contributors and end users. These roles are not unique to humans; all participants in the decentralized framework characteristic of the Internet Age can be understood in terms of these roles. The role of participants as contributors will be discussed in Pillar Three. As end users, participants are tasked with making use of the information made available by automatic data retrieval systems. That is, we have some task or project for which the retrieved data is useful. Human participants are unique from the perspective of the framework only in terms of the unique kinds of projects or tasks they bring to the table.
Just as the collapse of ontology renders data retrieval practically impossible for end users, it also becomes nearly impossible to know what to do with the information retrieved. Often, the end user’s project on its own is insufficient to determine how to use data in order to complete the task. So the second pillar of the Internet Age is developing and designing an interface between a network and its participants that presents information as primed for use. Nudges are design sequences or interface protocols that dispose the end user to treat information in particular ways, or make suggestions on how to use the information presented. Some nudges are intuitive, reflexive, and natural; others require habituation, training, and standardization among participants and infrastructure. Organizing information in a suggestive way in order to yield some desired behavior is also a very old technique (see road signs, advertisements), but is reinvented in the Internet Age to be flexible, customizable, and personalized through the feedback made possible by a participatory framework. TiVo’s suggestions are a familiar example of dynamic nudges.
It is important to note that this is a separate issue from the first pillar, although the two are intimately related. For the most part, the collapse of ontology is something end users do not have to worry about. It doesn’t matter how a search engine organizes things internally, as long as it can get you the information you need when you ask for it. To be sure, understanding how a search engine organizes information can help focus the search project, and (contra Dreyfus) using a search engine effectively is a skill that must be mastered. Furthermore, we should be interested in how our primary data retrieval systems go about their organizational processes, since its methods could have negative or unintended consequences on the nature of the framework itself. It is worth investigating possible biases in Google’s search results, for instance.
But usually, end users don’t care how a search engine organizes things, they just want access to the things organized by whatever methods are available. But once those things are made available, end users DO care how to make use of those things. So beyond organizing data, search engines (and all contributors) are charged with the additional task of presenting that data in convenient and efficient ways to maximize the instrumental benefit to the end user. This is especially important in light of the collapse of ontology. Since the Internet Age has abandoned the essentialism of the Aristotlean categories, then the content of the information retrieved can no longer inform the proper use of that information. Therefore, carefully designed nudges involved in the presentation of information are critically responsible for the flourishing of the Internet Age. The explicit form/content distinction that was lauded in the Web 2.0 hype back in 2005 put the focus of web design directly on facilitating a participatory interactive network, which include nudges toward commenting, aggregating, and tagging information to encourage useful presentation of material to all end users. Failing to design interfaces that encourage these behaviors will lead to the collapse of the participatory framework of the Internet Age.
Pillar Three: Spimes
The only way to increase the value of the network is to add more participants. Participants are both contributors and end users, and though there is a recognized disparity between those who contribute and those who merely ‘leech’, the behavior of both contributors and leechers strengthen the connection weights of the semantic web, and thus all participants add value to the web. Although nudges are designed to shape, manage, and encourage certain kinds of behaviors within the system, ideally it manages behaviors without compromising freedom of choice, and so the data points observed from actual usage will yield interesting results that will shape the organization of information and ultimately the development of finer and more efficient nudges.
Even still, the contribution and use of information by human participants alone cannot account for the wealth of information made systematically available in the Internet Age. Much of the information made available on the web is broadcast and integrated by objects themselves. In other words, objects themselves become contributors and users in the overall network. This move to Object Participants is characteristic of the decentralized networks of the Internet Age; in particular, such networks are no longer centered on humans are the primary agents of the system. Humans are just one kind of participant in an framework that accepts (and encourages) a variety of participants, each with unique projects and protocols for interfacing with the overall network.
I hesitate to call contributions of object participants ‘automatic’ (in the way that Google’s data retrieval is automatic), since the way an object broadcasts information might very well depend on human intervention and propagation. When an object is autonomously responsible for broadcasting its own information in order to integrate into an Internet of Things, when we have a proper Spime in Bruce Sterling’s sense. At the moment, we have lots of semi-automatic protospimes that, with the aid of humans, broadcast limited information about themselves to largely firewalled databases. The paradigmatic protospimes are packages sent through FedEx or UPS, which periodically update the web with info about their current location (or when they were last seen by a scanner) and estimates for when they will arrive at their destination. Since humans typically operate the scanners as they sort and transport these packages, the system is not fully automatic. Ubiquitous internet and technologies like RFID tags strive to make this processes more automated. When our world is sufficiently spimey, we will have built the third and final pillar of a mature Internet Age.
The collapse of ontology means that objects don’t fall into neat preestablished categories. Nevertheless, there are certain datapoints (and metadatapoints) that are of general interest for all objects, insofar as we want to keep track of them, and are therefore part of a reduced, minimalistic ontology of things. These include things like current location in space and time, proximity to other relevant points of interest, and metadata like place of origin, destination, constitution, and so on. Call this “situational data”. Some of this situational data, like spatial location, only hold for real, physical objects. Digital information has its analog in tools like linkbacks, hyperlinked references, and footnotes that track the sources and histories of information, and which give some noncommercial motivation for being concerned with intellectual property. In fact, maintaining the integrity of situational data for digital information is absolutely crucial for search engines hoping to cope with unorganized, miscellaneous data. While the collapse of ontology was essential for sorting through the huge amounts of digital data without the physical constraints of obsolete organizational schemes, the Internet Age ironically encourages projecting this new minimalistic ontology back onto physical things in order to create an Internet of Things, where not just data but objects themselves become searchable and retrievable by the same basic participatory methods.
Spimey objects, like all participants in the network, are responsible for their own continued integration with the rest of the network. This means objects are responsible for broadcasting relevant information in a form that is interoperable with other protocols governing the interactions of participants, and for demanding an interface that encourages such interactions. Sterling argues that accessing this information for all objects is critical for solving the sustainability problem, and that the only practical way of accessing this information is by having the objects themselves manage and broadcast this information to the rest of the network. In other words, make the objects themselves part of the participatory framework. Undoubtedly this is an important piece of the puzzle. But even if spimes aren’t a solution in themselves, their contributions to the network are valuable on their own, as is any participant.
The Three Pillars of the Internet Age are all mutually supporting and mutually dependent, and cannot be understood or implemented without the well functioning of the other parts. The Internet Age is still in its infancy (or more accurately, in its adolescence), and we are currently witnessing the bootstrapping of this infrastructure, which is why the system remains somewhat unstable and many everyday technologies (like TV) have yet to fully make the transition to the internet.
There is more to say about each of these pillars, an in fact I am just organizing the themes of some noteworthy recent books without really doing anything new, but I think making this general structure explicit helps to understand the role each of us (including our machines) play in the confusing, unimaginable present.