Vision of the Knowledge Network Our basic goal is to develop an Intermental Knowledge Network (IKN) where the minds of collaborating individuals are linked between themselves and to distributed electronic information resources. This IKN will use a variety of ingenious sensors and actuators to enhance the rendering and expression of information to and from the collaborating minds. This will enable minds universal access to the IKN independent of most physical disabilities. In our Knowledge Network, computers play three distinct roles. Firstly there is the powerful but rather conventional role as Web and other servers dispensing and creating knowledge on demand -- they create an electronic information world with universal access in which fertile medium our IKN is built. Our IKN will specialize in bootstrap fashion on the study of information aimed at building IKN's of our type. Secondly computers serve the human minds in linking them to this world -- here we use our NeatTools software to rapidly prototype and optimize the universal access for particular individuals. Thirdly our IKN will enable more than the traditional incoherent asynchronous interactions of minds with web based information systems. Conventionally each client (human mind) links essentially independently to a single (web) server in a given transaction. The IKN’s Wisdom is obtained by the incoherent but interacting sum of individual contributions. As in a parallel computer, our proposed IKN will as necessary enhance this asynchronous activity with the coherent and synchronous linkage of minds together to tackle a single problem. This coherent linkage is achieved by developing our collaborative system TangoInteractive to support universal access. Already TangoInteractive supports both synchronous and asynchronous activities but this feature will be further extended in this proposal. Note that our concept is a major extension of the interesting and still developing shared immersive virtual environments. In the latter, one represents the world classically by the actions of other people on it. In contrast, an IKN directly represents other participants through a rendering of their perceptual state. Similarly an Intermental Knowledge Network reverses the traditional role of the human and computers. Rather than the human as the usually asynchronous viewer of the computer's possibly parallel computations, we consider the digital computer network as fertile medium supporting the really powerful computer corresponding to the synchronous interactive IKN of linked minds. Note how designing our universal access knowledge network has naturally led to the novel concept of an Intermental network which can be used in essentially all Knowledge Networks -- whether or not they feature the use or study of universal access. IKN Technology Summary There are three major distinctive technologies needed to implement our IKN. Firstly there are the sensors and hardware interfaces needed to implement the linkage of individual minds to the IKN with universal access. Secondly there is the software system NeatTools which implements a powerful rapid prototyping dataflow paradigm to integrate the hardware components into complete client systems. Finally we form the coherent synchronous IKN using the Tango Interactive collaboratory which formally supports the sharing of distributed objects which are either the linked minds or information "nuggets" from underlying web. TangoInteractive for the Universal Access IKN Currently TangoInteractive, like the pioneering Habanero system on which it was based, supports the powerful event sharing model for replicated objects which appears to most appropriate for any KN supporting synchronous object sharing. In this proposal, we extend Tango in two critical fashions. Tango is built around core client and server infrastructure, which is exposed to the user as a "control application". This graphical user interface allows overall control of and access to the linked users and applications. This information naturally must be universally accessible to all the minds involved in an IKN synchronous session. We will extend our current Java GUI to support universal access and believe this will be quite straightforward as this applet has rather limited and well defined scope. Universal access to other shared objects will be handled differently and will exploit Tango's ability to support the linkage of different views of a given object. Any object can be automatically mapped through filters embedded in Tango's core infrastructure and so customized separately for each client. We will design a suite of filters that will supply access to chosen classes of users. Note that support of different views is already implemented in Tango but for cases where clients have different roles (such as teacher and student) rather than different interface requirements. Note that support of different views requires cooperation between server and client. For instance a blind user would download an audio enriched version of the object while the deaf client would download the image rich version. On the client side, we would use NeatTools to interface the chosen object views with the available sensors. Note that we will develop some filters such as sonification for general HTML pages, which will be useable on a broad class of objects. Other filters will be specialized to particular objects such as a shared Java educational applet illustrating a physical simulation. Any KN will need both synchronous and asynchronous collaboration and Tango Interactive's capabilities will be extended in this respect as part of ongoing activities outside this proposal. Already Tango supports several versions of a shared Web browser and we are adding a database backend for recording synchronous sessions for later playback. We are also linking Tango with Lotus Notes so as to be able to exploit its well known workflow and asynchronous tools such as calendars and threaded discussion lists. Other relevant Tango extensions include support of a classic multi-room paradigm with possible persistent applications and users and appropriate security.