Abstract— Conversation is the most natural communication means for people to communicate with each other. I believe that conversation plays a critical role in realizing a paradigm of human-centered web intelligence in which web intelligence engines are grounded on the human society. We are currently building a computational framework for circulating information in a conversational fashion, using information packages called conversation quanta that encapsulate conversational scenes. Technologies are being developed for acquiring conversation quanta on the spot, accumulating them in a visually recognizable form, and reusing them in a situated fashion. Conversational Informatics, based on measurement, analysis, and modeling of conversation, constitutes the theoretical foundation for these applications. I will overview recent results in Conversational Informatics that will help achieve our vision. I will also discuss our approach in the context of Social Intelligence Design aimed at the understanding and augmentation of social intelligence for collective problem solving and learning. Index Terms— Conversational Informatics, Social Intelligence Design, Human-centered computing, Human computer interaction I. INTRODUCTION HE goal of Web Intelligence is to create a world wide wisdom web (w4) by integrating individual intelligences available on the global network using technologies such as web agents, web mining and farming, web information retrieval, web knowledge management, web intelligence infrastructure, and social network intelligence [1]. In order for Web Intelligence to be able to maximally benefit the human society, it should be well-interfaced to the human society so that each member of the human society can benefit from it without much difficulty and Web Intelligence can gain enough sources of knowledge from the human society. Web Intelligence will synergistically co-evolve with the human society if it is intimately embedded in the human society. One of the key issues in embedding Web Intelligence in the human society is information grounding, which roughly means that the information user is aware of the association between information and the real world. Web Intelligence need to provide information in such a way that people can readily Toyoaki Nishida is with Dept. of Intelligence Science and Technology, Graduate School of Informatics, Kyoto University, Yoshida-Honmachi, Sakyo-ku, Kyoto 606-8501, Japan (phone: +81-75-753-5371; fax: +81-75-753-4961; e-mail: nishida@i.kyoto-u.ac.jp). ground it on their daily life. Even though potentially useful information is provided with Web Intelligence, it might be useless if the information user fails to recover the reference to real world or reconfigure the image implied by the given statement. Unfortunately, information grounding is not easy to establish once information is isolated from the original situation it is created unless special care is taken at the moment information is created. We need to invent a technology that permits people to preserve cues for information grounding on the spot so that they can help ground the information later at the situations different from the original one. In this article, I focus on the role of conversation in information grounding and present a suite of technologies aimed at realizing a paradigm of human-centered web intelligence. Apparently, conversation is the most natural communication means for people to communicate with each other. A closer look reveals that various kinds of processes related to creation or recovering information grounding are in action in conversation. For example, pointing and gaze are basic forms of creating association with the real world and co-occurring propositions. Gestures and postures may suggest the scope and modality of the utterances. In addition, dialectic aspects of conversation help participants interpret the meaning of information through discussions. We are currently building a computational framework for circulating information in a conversational fashion, using information packages called conversation quanta that encapsulate conversational scenes consisting of participants' behavior, references to the environment, and meta-descriptions. Technologies are being developed for acquiring conversation quanta on the spot, accumulating them in a visually recognizable form, and reusing them in a situated fashion. Conversational Informatics is a field of research aimed at establishing the theoretical foundation for these applications, based on measurement, analysis, and modeling of conversation. The field exploits a foundation provided by Artificial Intelligence, Natural Language Processing, Speech and Image Processing, Cognitive Science, and Conversation Analysis. It is aimed at shedding light on meaning creation and interpretation resulting from the sophisticated mechanisms in verbal/nonverbal interactions during conversation, in search of better methods of computer-mediated communication, human/computer interaction, and support for knowledge Conversational Informatics and Human-Centered Web Intelligence Toyoaki Nishida, Member, IEEE T Feature Article: Toyoaki Nishida 19 IEEE Intelligent Informatics Bulletin November 2007 Vol.8 No.1