Subscribe Us

Get free daily email updates!

Follow us!

Wednesday, September 21, 2011

The Smart Internet Vision


The notion of a smart internet requires a transformation in our understanding of the web and its architecture – a complete change of perspective, from a server-centric understanding to a user-centric one. This change will be much like the Copernican revolution, where the presumed structure of the solar system changed from an Earth-centric one to a Sol-centric one.
Three major extensions are called for in this transformation. First, a new “Copernican” user model for the web is needed that is centered on the users’ concerns and cognition. Second, a new kind of session concept is required that centers on the user’s perspective and her situation rather than the server’s perspective of user interactions.

Thirdly, the concept of dynamic social binding of web interactions, to turn what is currently a single user web interaction model into multi-users’ collaborative web interactions under the user’s control, is also needed. In essence, the smart internet supports an instinctive user model of the web, one in which the discovery, aggregation and delivery of services and resources results in rendered content that is optimal for each user or group’s situation. The smart internet will be an evolving extension of the internet in which online services and resources are discovered, aggregated and delivered dynamically, automatically and interactively in response to a user’s or group’s evolving concerns and situations, which may involve real time or proactive performance of tasks that address users’ goals. All aspects of this interaction must be conducted with awareness of, and adaptation to, the user’s personal and group context, task requirements and characteristics. The resulting aggregation of resources and content will be delivered in a manner appropriate to the user’s current concerns or situation, and as a unified entity abstracting relevant content and services from a single site, or from multiple sites and organizations. To be practical, the smart internet must be an evolution and extension of the current internet, building on the existing basic architectural elements of HTML, URLs and HTTP, while hiding these techno-centric elements behind objects and interaction that are more appropriate and intuitive, tailored to the end user’s current domain, goals and concerns. Rather than the user initiating interactions to accomplish tasks themselves, the smart internet should allow the user’s current concerns to drive the implicit discovery and aggregation of services and resources to serve the user and support the user’s cognition and action. There are three distinct principles of smart internet that set it apart from the internet today: an instinctive user model, sessions for users and their matter of concerns, and collective and collaborative web interactions.
User-Centric Model for Instinctive Interaction:

Identifying and applying appropriate user models is essential, in keeping with the requirements of user-centered design. Instead of being user-centered, the user model of the internet today is by and large techno-centric, exposing the fundamental components of the web architecture, resulting in a “one HTML page at a time” interactive model convenient for the server. The widespread use of the Internet should not be taken as proof that the techno-centric user model is sufficient. Users adapt to fit what the web has to offer, and in order to use valued content and services, they are willing to connect to myriad web sites, filtering the relevant information for their own context in order to address the task at hand. But this causes a great deal of inconvenience and wasted effort. What we seek instead is an alternative user-centric model that leads to interactions that are instinctive for the user, rather than being fitted to the server and awkward for the user.

Session for Users and Their Matters of Concern:

Today, the notion of session keeps track of the user and their interactions from the perspective of the server. The session ends when the user stops interacting with the site. Traditionally, a user session is defined as “a series of requests issued by a user to a web site in a single visit to the site” [9]. Technically, user sessions are HTTP sessions used to preserve the conversational state between a given server site and connection with a browser instance of a client device. Important session information such as user account and password are preserved and associated with the corresponding client, avoiding the need to ask for the same required information in a given request-response dialogue, resulting in better user experience. The existing concept of session, (i) is associated with one particular server, and (ii) is bounded by user’s real time synchronous interactions. When the web’s center of gravity is re-focused on the user, the concept of session must be extended beyond the server site view of user initiated real time synchronous interaction. In the smart internet, sessions are oriented to the perspective of users and their matters of concern, rather than simply being states that the server site wants to keep track of. Collaborative and Collective Web Interactions The third principle of the smart internet that distinguishes it from the current internet is that it explicitly supports close collaborate between users to resolve shared matters of concern. This principle has the following implications.

1. Dynamic Social Binding Dynamic social binding is defined as the capability to select other users dynamically to share interaction elements for different levels of interactions associated with mocs Shared interactions will occur at different levels of intensity, ranging from sharing of views as read-only, to co execution or delegation of tasks and sub-tasks of mocs, thereby turning web interactions from solitary undertakings to multi-user collaboration. Online shopping can be used to demonstrate the application of this principle. Suppose a user, A, has started a matter of concern relating to online Christmas shopping for his children. He places multiple items from the catalog into the shopping cart. Using dynamic social binding, user A selects his wife, user B, to co-execute different elements of interaction for the online Christmas shopping task as a moc. Now user B is enabled to participate and collaborate in operations of user A’s moc such as adding items to the shopping cart. Once the collaborative work has been completed, User A can transfer the session to user B or end the dynamic social binding session and continue to checkout himself.

2. Collective Intelligence In Smart interaction, matters of concern become the major drivers of activity, explicitly centering the processing on user needs and interests. In keeping with this focus on mocs, text and semantic search in the smart internet should return search results in units of mocs as a (pre-set; or ready-to-use as-is or with minor modification) purpose-built composite collection of related services and resources instead of being simply a list of unrelated single hyperlinks as in today’s internet. We should envision a new kind of search interface that enables users to locate, customize, consume, rate and review. Tools should also be available for users to personalize their own search interfaces based on their mocs. Queries such as “what do people with similar profiles to mine do in similar matters of concern (moc)?” should also be answered by the collective intelligence provided by other users with similar mocs. This functionality can be provided by harvesting statistical data concerning historical behavior, user ratings, user reviews and feedback presented to users. Proactive analysis on such collective intelligence can be done as batch processing so that it is readily available at runtime to support new user interactions.

0 Responses to “ The Smart Internet Vision ”

Post a Comment