Declaration
Chapter 9: Construction of the Information Personae
Download 0.56 Mb.
|
- Navigate this page:
- Autonomous Agents
Chapter 9: Construction of the Information PersonaeThe emphasis of our culture on the individual has produced a complex network that is at once interconnected and disconnected. The computer revolution turned its back on those tools that led to the empowerment of both co-located and distributed groups collaborating on common knowledge work. In light of this, we need to consider some early work of Douglas Engelbart’s research group at the Augmented Research Centre (ARC) at the Stanford Research Institute (SRI). In the early 1960s this group was developing systems that promoted collaboration among people doing their work in an asynchronous, geographically-distributed manner. At the time the project started, display technologies were extremely primitive—most people were still using punch cards and paper tape (Engelbart and Lehtman 247). This was probably the biggest deterrent to successful experiments in online collaborative work. In order to address this problem, Engelbart and his team developed the first Online System, or NLS terminals. The display consoles were equipped with typewriter-like keyboards, a five-finger key set (a five-finger equivalent of a keyboard) and a mouse that Engelbart invented. Engelbart explains that the mouse was just a demonstration of “augmenting knowledge workers.” The rest of the world was focused on the idea of “office automation,” believing that the “real user” of computers was a secretary who needed tasks automated”95 (Hunt 84). We have transitioned from a time when computer networks were designed primarily for machine-to-machine interaction, though in a fairly “dumb” fashion, to our present situation where computer networks largely facilitate human-to-human interaction. We are now approaching a time where machine-to-machine interaction will play a far more central role once again, though increasingly independent of us. In other words, we are entering an era of autonomously functioning software that performs work for us, what many now refer to as “agents.” Perhaps this notion is yet another human-machine fantasy, but it also offers the possibility of being a means of reclaiming our time and getting away from endless hours in front of the screen. The combination of ubiquitous computing with agent software promises to change drastically the way we work and socialise. This conclusion is not unique when one considers the enormous efforts at work in industry and academia to develop online agents. But designing agency is a task riddled with philosophical and ethical implications with many issues of identity and privacy at its centre.
Non human-agentsIt’s a hot topic because it sometimes seems that there are all sorts of non-human entities, such as cyborgs, intelligent machines, genes, and demons lose in the world. Along with ozone holes, market forces, discourses, the subconscious, and the unnameable Other. And, or so many claim, such non-humans actors seem to be multiplying. For if angels and demons are on the decline in the relatively secularised West, then perhaps robocops and hidden psychological agendas—not to mention unnameable Others—are on the increase. (Callon and Law 481) In 1997, world chess champion Garri Kasparov lost for the first time against “Deep Blue,” IBM’s speedy super-computer. The match was one of the most popular live events staged on the Internet. The web site set up received more than 74 million hits representing more than 4 million users visits from 106 countries during the 9-day event. Kasparov himself said that his greatest wish was to posses a combination of human intelligence and computer memory—not a simulation of human intelligence, but intelligent access to the archives (“Case Study: The Anatomy of a Chess-Cast”). Thus, it is no longer the storing and archiving of information, but the suppression of it that has become a central cultural technique of the information age. And for this, we turn to non-human agents—to search, filter, and select the information we need. According to traditional humanist notions, what marks human agency is “action” as opposed to the mere “reaction” of machines. Human action is perceived to be intentional, responsive, and “free.” Could non-humans ever be agents? This is a question that philosophers, socio-biologists, theologians, science-fiction writers, scientists, and those working with the technologies are grappling with and that is aptly posed by Michael Callon and John Law in “Agency and the Hybrid Collectif.” In their essay they challenge the notion of human and non-human agency and make an important point that it is not only the relations, but also their heterogeneity, that are important. Autonomous AgentsWhen I refer to “agents on the Net,” I mean software that is programmed to do specific tasks autonomously. In Chapter 3 (“Telematic Culture”), I briefly discussed bots, a type of primitive agent that emerged in the early days of the Net. It is important to note, however, that there is not just one accepted definition for an agent. Stuart Russel, Professor of Computer Science at UC Berkeley, and Peter Norvig, Computational Sciences Division Chief at NASA, authors of a widely used text book, Artificial Intelligence: A Modern Approach, give a very general definition of an agent as anything that can be viewed as perceiving its environment through sensors and acting upon the environment through effectors. Autonomous agents originate primarily from artificial intelligence research. Object-oriented programming, human-computer interface design, control theory, cognitive psychology, and robotics have all contributed to the development of artificial intelligence developments. Thus, by definition, it is a highly interdisciplinary field. Michael Knapik and Jay Johnson have gathered various definitions of agents in their book, Developing Intelligent Agents for Distributed Systems.96 One generic operational feature of agents, they state, is their autonomy. Autonomous agents can operate without immediate intervention of humans and can have some kind of control of their internal state. They bring up an important feature—the social ability of agents. Earlier agent-oriented approaches have often focused on a single agent with simple knowledge and problem solving skills or a single agent with a general knowledge that performs a wide range of user delegated tasks. A centralised approach requires huge amount of information and processing power. Such centralised approaches fail in software systems just as they do in political and organisational systems—and it follows that they may well be the wrong approach for developing networked social spaces.
Since the introduction of the Web and its consequent commercialization, one of the problems universally facing developers is how to deal with information overload. Search engines became the information bots and they are widely developed and used on the Internet. But as the search agents become more powerful, they overload us once again with too much information. An intelligent agent (or simply an agent) is a program that gathers information or performs some other service without the immediate presence of the user and on some regular schedule. Typically, an agent program, using parameters provided by the user, searches all or some part of the Internet, gathers information the user is interested in, and presents it on a daily or other periodic basis. The next generation of agents focused on classification and narrowing down information, returning us to issues of categorising that I discussed earlier when considering databases. That navigating information would be the primary problem with the explosion of computer networks soon became evident to developers working in this environment. In the early 1990s, Internet pioneer Brewster Kahler developed the Wide Area Information Server (WAIS), the first system for publishing quantities of data in a searchable form on the Internet. WAIS helped bring commercial and government agencies onto the Internet by selling Internet publishing tools and production services to companies such as Encyclopaedia Britannica, the New York Times, and the Government Printing Office (Kahler, “Archiving the Internet”). As I mentioned in Chapter 5, Kahler is now archiving the Internet. More recently, with the rapid expansion of the Web and the near frenzied rate of Internet stocks, the stakes have been raised tremendously. Browser and search firms are purchasing technology that improves Web navigation. For instance, search company Lycos bought WiseWire, which automatically organizes Internet content into directories and categories. In 1997, Microsoft bought Firefly, which recommends content to Web surfers based on profiles they submit (“Microsoft Catches Firefly”). Directory: publications publications -> Acm word Template for sig site publications -> Preparation of Papers for ieee transactions on medical imaging publications -> Adjih, C., Georgiadis, L., Jacquet, P., & Szpankowski, W. (2006). Multicast tree structure and the power law publications -> Swiss Federal Institute of Technology (eth) Zurich Computer Engineering and Networks Laboratory publications -> Quantitative skills publications -> Multi-core cpu and gpu implementation of Discrete Periodic Radon Transform and Its Inverse publications -> List of Publications Department of Mechanical Engineering ucek, jntu kakinada publications -> 1. 2 Authority 1 3 Planning Area 1 publications -> Sa michelson, 2011: Impact of Sea-Spray on the Atmospheric Surface Layer. Bound. Layer Meteor., 140 ( 3 ), 361-381, doi: 10. 1007/s10546-011-9617-1, issn: Jun-14, ids: 807TW, sep 2011 Bao, jw, cw fairall, sa michelson Download 0.56 Mb. Share with your friends:
|