From: rick@cs.arizona.edu (Rick Schlichting) Subject: INDUSTRY: Kahaner Report: Sony Computer Science Lab Date: 25 Apr 93 03:52:20 GMT Lines: 746 Crossposted from comp.research.japan [Dr. David Kahaner is a numerical analyst on sabbatical to the Office of Naval Research-Asia (ONR Asia) in Tokyo from NIST. The following is the professional opinion of David Kahaner and in no way has the blessing of the US Government or any agency of it. All information is dated and of limited life time. This disclaimer should be noted on ANY attribution.] [Copies of previous reports written by Kahaner can be obtained using anonymous FTP from host cs.arizona.edu, directory japan/kahaner.reports.] From: Dr. David K. Kahaner US Office of Naval Research Asia (From outside US): 23-17, 7-chome, Roppongi, Minato-ku, Tokyo 106 Japan (From within US): Unit 45002, APO AP 96337-0007 Tel: +81 3 3401-8924, Fax: +81 3 3403-9670 Email: kahaner@cs.titech.ac.jp Re: Sony Computer Science Lab 24 April 1993 This file is named "sony-csl.93" ABSTRACT. Activities at Sony's Computer Science Laboratory (Tokyo). I wanted to hear about new activities at Sony's Computer Science Laboratory (CSL) since I last visited (see "sony", 30 May 1990). My host was CSL's Director. Dr. Mario Tokoro, Director Sony Computer Science Laboratory Takanawa Muse Building 3-14-13 Higashi-gotanda, Shinagawa-ku Tokyo 141 Japan Tel: +81 3 3448-4380; Fax: +81 3 3448-4273 Email: MARIO@CSL.SONY.CO.JP Tokoro has been CSL's Director since its founding in 1988. Before that he was a Professor at Keio University, and even now he spends four days each week at Keio and only one day each at CSL. With a small lab, 12-15 scientists, engaged in basic research he feels that amount of time is sufficient. The lab has expanded somewhat since I saw it last, but its general look remains the same, located in a building in the southern portion of central Tokyo. Sony has a strong presence in this area of the city, with a variety of labs and other facilities. CSL occupies one floor of a modern office building. Very Westernized individual offices and large common space for informal discussions typify CSL. I am told that it was a model for at least one other new Japanese lab, the Real World Computing Lab in Tsukuba. Tokoro was involved in shaping the original charter of CSL, which remains that of doing basic research in Computer Science. He explained to me that the staff interact frequently with Sony's other labs but that those are focused heavily on development. CSL's goal is to be a window on activities that will be of interest in the next century, and possibly achieving breakthroughs in computer development. For that reason he has tried to recruit outstanding scientists and encouraged them to work openly in an international arena by publishing in English in respected journals. CSL's corporate charter specifically states that "it is our policy to make public the results of our research and to avoid developmental activities that have as their objective increasing the profits of private corporations." (A list of recent CSL reports is attached to this report.) Tokoro pointed out that compensation for his scientific staff in entirely unrelated to seniority and directly coupled to performance. Each researcher sets his or her own research goals within CSL's basic research themes. The latter are focused on distributed operating systems, computer networks, programming languages, user interfaces, etc. These also correspond to some extent to Tokoro's own research interests; it is clear that he has had an important role in defining current activities at CSL. I asked him is there is a problem with such a loose linkage between CSL and the main-line Sony activities, either in the form of pressure from the corporation for practical output, or from the researchers due to lack direction. He commented that CSL's President T.Doi is on the Board of Directors of the parent company and has provided very strong support for CSL's activities. Also, he admitted that a small lab is a very minor mark on Sony's balance sheet, and this was a good reason for not allowing CSL to become too large (e.g., expensive). On the last point he felt that selection of researchers with strong internal goals was essential, with some of the same self selection that occurs in university faculty. Also he encourages group collaboration to provide reinforcement. Finally, although the lab is a basic research facility, it is not difficult to see how some of their research projects (see below) could ultimately move into corporate development and in that sense CSL is somewhat integrated into the company's long term strategic planning. Tokoro explained three major projects. One is based on future user mobility. He feels that computing environments in the next decade will be widely distributed, constantly-changing, and ubiquitous. People will move with computers and will use them while moving. The simplest example would be a user who wants his computer environment to come up on a local workstation wherever he is, without having to perform a remote login to his "home" system. Similarly, electronic mail should follow users like a cellular telephone. Currently, a user who is working at the office and wants to leave will typically save editing files and stop applications. Then these can be copied from an office workstation to a laptop and started up there. The next day he will reverse the process. (Removable hard disks, or Apple Mac's Duo are approaches to this issue.) Tokoro believes that Unix will be unable to support this level of mobility while maintaining reliability, security, and flexibility, and that other approaches are required. For more than five years Tokoro, students, and colleagues have been developing a fully object-oriented operating system that is focused on object mobility. Here an object should be free to move around a distributed environment, just as portable/mobile computers are moved frequently from one place to another. (Existing systems assume stability of computers and objects.) Mobile objects should also adapt to their execution environment, and be highly dependable, e.g., objects should be trusted to provide services so that tasks with deadlines will be finished on time. Tokoro has proposed a model (computational field model) that deals with object mobility. Since the semantics and properties of an object change while it is running or when it moves, this approach separates objects and metaobjects, the former being a "container" for information, the latter defining the semantics. Each object has its own group of metaobjects and in this sense a group of metaobjects is a virtual machine. Since a metaobject is also an object there can be metaobjects for it, etc., leading to the notion of a metahierarchy. The notion of migration of an object means that an object changes its group of metaobjects. Since portable/mobile computers are frequently moving around networks, an object communicating with these computers has to change its communication protocol from local to interconnected. This can occur when the object migrates to a group of metaobjects which have suitable protocol modules to communicate with these computers. Each object can have its own properties. Some are temporary. Others, such as a name server or a directory service are shared by many client (objects). Some objects rely on the underlying hardware, including device drivers. The properties of an object changes as it evolves. The concrete implementation of these ideas is in the Apertos (Open) operating system. The original name was Muse, but this was discarded because of naming conflicts with an existing, unrelated, product. Apertos is completely object-oriented, with a few twists for practicality. For example, it is difficult to inspect the internals of an object, because an object is protected against access from other objects, so objects must have a method provided for exposing its internals for debugging, etc. Object management (such as invocation or scheduling) is also difficult, as access to an object's state or other information is needed. Tokoro believes that separating objects from metaobjects allows these difficulties to be overcome. There are certainly other projects in object-oriented distributed operating systems (Amoeba, Chorus, Choices, etc). Tokoro believes that Apertos has certain advantages in the areas of encouraging concurrent object-oriented programming, as well as increasing maintainability and reusability, freeing systems programmers from kernel programming chores such as synchronization. Further, Apertos is specifically designed for a large-scale open, distributed system featuring mobile computing. The current status of Apertos is as follows. It has been implemented on a Sony News Workstation using C++, and it is being implemented now on a MIPS R3000 workstation. Implementation on a 68000-based or similar system is claimed to be fairly easy. Tokoro told me that a "hacker" could do this in at most two months. A researcher at CSL, Yasuhiko Yokote (Email: YKT@CSL.SONY.CO.JP) told me that they hope to port the system to a PC in the near future. Apertos is thought of as being at the "beta" test level. Tokoro is interested in locating colleagues who want to experiment, and will make Apertos available to them in source form for use in nonprofit institutions. However, my inclination would be to let CSL staff port to the system to a workstation that is more widely available in the West. (In this context, I asked Tokoro what features Sony felt it could bring to the workstation arena to encourage sales of its News or follow-on products, given that Sun and HP dominate the workstation market in Japan so heavily. He felt that people might prefer Sony's product on the basis of size, silence, etc. (e.g. human interface), and of course interesting applications. They certainly will have a tough battle in this highly competitive sector.) Another major project falls in the general area of facial displays. The background for this is Tokoro's concept of an "intimate computer". He envisions a small computer that can accept verbal commands, and respond not only verbally, but also with an appropriate facial display of a person (wife, friend, teacher, etc) known to the user, and with expressions that are meaningful to the discussion context. Of course, this is a very advanced idea and CSL staff have a long way to go. Also, advanced interfaces are being studied by many other research labs. But this particular project seems to me to be quite consistent with Sony's consumer product and market experience, more so than the workstation. One step in this direction was Sony's Palmtop, pen-based system. This was cute and interesting, but Tokoro and I agreed that typing was more comfortable for us than writing. Nevertheless, the concept of a small, really personal, computer is unassailable. Work on facial displays is being led by Dr. Akikazu Takeuchi (Email: TAKEUCHI@CSL.SONY.CO.JP) who showed me several of the research activities. Research papers on this subject are often presented at human-computer-interface conferences, so I only will summarize the key points here. A subject's facial image is input, either by video camera, photo, etc., and then a wire frame model is built via 500 polygons. The face can be rendered using a skin-line surface material with Gourand shading or by applying a texture map taken from a video frame or picture. This can be tuned manually (via typical surface fitting shape functions) to allow a three dimensional model that can be rotated and animated. Facial display is realized by local deformation of the facial polygons, using an underlying muscle model from K. Waters (16 muscles and 10 parameters controlling mouth opening, jaw rotation, eye movement, eyelid opening, and head orientation). Various expressive facial movements can then be built up by moving eyebrows, mouth, head, etc., to embody emotion such as agreement, anger, surprise and others. At run time, the animation subsystem awaits a request from the speech subsystem, which specifies values of the 26 parameters, based on a simple linear differential equation specifying the rate of change of each of the parameters. About 20-25 frames per second can be realized on a SGI workstation. I found the face construction aspect of this project fascinating, and in fact Takeuchi has built up a lovely interface that he calls a "face construction lab", and an "interactive facial animation testbed". These would be of great interest to anyone working in the area of facial processing. (Some of the underlying modeling is collaborative with Keith Waters. Also, Steve Franks, who had a visiting internship at CSL last year, helped Takeuchi with this project and jointly implemented the interactive facial animation testbed.). For the speech subsystem, voice input is acoustically analyzed by a sound processing board. A speech recognition module outputs word sequences that are assigned higher scores by a probabilistic telephone model. These sequences are syntactically and semantically analyzed and disambiguated using a loose grammar and restricted domain knowledge. >From the semantic representation of the input, a plan recognition module extracts the speaker's intention. ("I am interested in Sony's workstation" leads to "he wants to get precise information about Sony workstations".) The intention output generates a response that is output by a voice synthesis module. Each module except the very last, can send messages to the facial animation subsystem about which facial display should be generated. In situations where speech inputs are not recognized, or otherwise invalid, a display corresponding to "not confident" is displayed, a rather quizzical look on the display face. At the moment various experiments are being conducted to measure the success of the system. It is a very interesting project. But even with a great deal of effort it still has only a modest vocabulary and clearly there is much more opportunity to develop it further. Another project (that I did not see demonstrated) was network protocols to support a moving host. This is a network layer that has been slid above IP and below TCP, all sitting on top of Unix providing for host migration transparency via the concept of a virtual network. Director Tokoro, because of his strong academic connections, has a very open view of the research at CSL, and has encouraged visits by Western scientists, and free discussion. CSL has had a large number of short term visitors and at least a few long term Western visitors. Although for the latter to be successful, Tokoro emphasizes that it is important that there be a specific plan of collaboration. He definitely wants to encourage such planned visits, as opposed to just a "drop-in" sabbatical. ----------------------------------------------------------------------------- List of Sony CSL Technical Reports (in BIBTeX form). Latest reports are at the end. (Articles written in Japanese are omitted) @TECHREPORT{ TR-88-1, AUTHOR = "Yasuhiko Yokote and Fumio Teraoka and Mario Tokoro", TITLE = "{Object Management in the Muse Operating System}", NUMBER = "SCSL-TR-88-001", INSTITUTION = scsl, NOTE = "To appear in 5th Conference Proceedings of Japan Society for Software Science and Technology, in Japanese", MONTH = "September", YEAR = 1988} @TECHREPORT{ TR-88-2, AUTHOR = "Fumio Teraoka and Yasuhiko Yokote and Mario Tokoro", TITLE = "{Inter-Object Communication in the Muse Operating System}", NUMBER = "SCSL-TR-88-002", INSTITUTION = scsl, NOTE = "To appear in 5th Conference Proceedings of Japan Society for Software Science and Technology, in Japanese", MONTH = "September", YEAR = 1988} @TECHREPORT{ TR-88-3, AUTHOR = "Mario Tokoro", TITLE = "{Issues in Object-Oriented Concurrent Computing}", NUMBER = "SCSL-88-003", INSTITUTION = scsl, NOTE = "To appear in 5th Conference Proceedings of Japan Society for Software Science and Technology, in Japanese", MONTH = "September", YEAR = 1988} @TECHREPORT{ TR-89-1, AUTHOR = "Yasuhiko Yokote and Fumio Teraoka and Mario Tokoro", TITLE = "{A Reflective Architecture for an Object-Oriented Distributed Operating System}", NUMBER = "SCSL-TR-89-001", INSTITUTION = scsl, NOTE = "Also appeared in Proceedings of European Conference on Object-Oriented Programming '89", MONTH = "March", YEAR = 1989} @TECHREPORT{ TR-89-3, AUTHOR = "Fumio Teraoka and Yasuhiko Yokote and Mario Tokoro", TITLE = "{Muse-IP: A Network Layer Protocol for Large Distributed Systems with Mobile Hosts}", NUMBER = "SCSL-TR-89-003", INSTITUTION = scsl, NOTE = "To appear in Proceedings of the 4th International Joing Workshop on Computer Communications", MONTH = "June", YEAR = 1989} @TECHREPORT{ TR-89-4, AUTHOR = "Shinji Kono and Shigeru Watari and Mario Tokoro", TITLE = "{Object Storage System and Programming Transparency}", NUMBER = "SCSL-TR-89-004", INSTITUTION = scsl, NOTE = "Technical Report 89-SF-30-2, Information Processing Society of Japan (in Japanese)", MONTH = "September", YEAR = 1989} @TECHREPORT{ TR-89-5, AUTHOR = "Shigeru Watari and Shinji Kono and {Ei-Ichi Osawa} and Rik Smoody and Mario Tokoro", TITLE = "{Extending Object-Oriented Systems to Support Dialectic Worldviews}", NUMBER = "SCSL-TR-89-005", INSTITUTION = scsl, NOTE = "To appear in Proceedings of Advanced Database System Symposium", MONTH = "December", YEAR = 1989} @TECHREPORT{ TR-89-6, AUTHOR = "Ei-Ichi Osawa and Shigeru Watari and Nobuhisa Fujinami and Mario Tokoro", TITLE = "{Users' View of Objects in the Muse Operating System}", NUMBER = "SCSL-TR-89-006", INSTITUTION = scsl, NOTE = "To appear in 6th Conference Proceedings of Japan Society for Software Science and Technology (in Japanese)", MONTH = "October", YEAR = 1989} @TECHREPORT{ TR-89-9, AUTHOR = "Mario Tokoro", TITLE = "{MD-based Computing: Toward a Computing Model/Methodology for Open Distributed Environment}", NUMBER = "SCSL-TR-89-009", INSTITUTION = scsl, NOTE = "presented at UK-Japan Computer Science Workshop, Oxford", MONTH = "September", YEAR = 1989} @TECHREPORT{ TR-89-10, AUTHOR = "Yasuhiko Yokote and Fumio Teraoka and Masaki Yamada and Hiroshi Tezuka and Mario Tokoro", TITLE = "{The Design and Implementation of the Muse Object-Oriented Distibuted Operating System}", NUMBER = "SCSL-TR-89-010", INSTITUTION = scsl, NOTE = "To appear in Proceedings of First Conference on Technology of Object-Oriented Languages and Systems", MONTH = "October", YEAR = 1989} @TECHREPORT{ TR-89-11, AUTHOR = "Mario Tokoro", TITLE = "{On Object-Oriented Distributed Computing}", NUMBER = "SCSL-TR-89-011", INSTITUTION = scsl, NOTE = "To appear in 6th Conference Proceedings of Japan Society for Software Science and Technology (in Japanese)", MONTH = "October", YEAR = 1989} @TECHREPORT{ TR-90-2, AUTHOR = "Mario Tokoro", TITLE = "{Toward a New Computing Model for an Open Distributed Environment}", NUMBER = "SCSL-TR-90-002", INSTITUTION = scsl, NOTE = "To be appeared in Lecture Note in Computer Science", MONTH = "April", YEAR = 1990} @TECHREPORT{ TR-90-3, AUTHOR = "Shigeru Watari and Ei-Ichi Osawa and Yasuaki Honda and Mike Reeve and Mario Tokoro", TITLE = "{Towards Music: A Description Language for the Muse Object Model}", NUMBER = "SCSL-TR-90-003", INSTITUTION = scsl, NOTE = "To appear in Proceedings of The Second International Conference on Technology of Object-Oriented Language and Systems", MONTH = "May", YEAR = 1990} @TECHREPORT{ TR-90-4, AUTHOR = "Rik Fischer Smoody", TITLE = "{JIBUTSU ($B<-J*(J), DICTIONARY OBJECTS}", NUMBER = "SCSL-TR-90-004", INSTITUTION = scsl, NOTE = "To appear in Proceedings of The Second International Conference on Technology of Object-Oriented Language and Systems", MONTH = "May", YEAR = 1990} @TECHREPORT{ TR-90-5, AUTHOR = "Fumio Teraoka and Yasuhiko Yokote and Mario Tokoro", TITLE = "{Virtual Network: Towards Location Transparent Communication in Large Distributed Systems}", NUMBER = "SCSL-TR-90-005", INSTITUTION = scsl, NOTE = "To appear in Proceedings of 5th International Joint Workshop on Computer Communications", MONTH = "June", YEAR = 1990} @TECHREPORT{ TR-90-6, AUTHOR = "Mario Tokoro", TITLE = "{Computational Field Model: Toward a New Computing Model/Methodology for Open Distributed Environment}", NUMBER = "SCSL-TR-90-006", INSTITUTION = scsl, NOTE = "To appear in Proceedings of 2nd Workshop on Future Trends in Distributed Computing Systems, Cairo, Sep. 1990.", MONTH = "June", YEAR = 1990} @TECHREPORT{ TR-90-8, AUTHOR = "Fumio Teraoka and Yasuhiko Yokote and Mario Tokoro", TITLE = "{Communication Protocols for Supporting Mobile/Portable Hosts}", NUMBER = "SCSL-TR-90-008", INSTITUTION = scsl, NOTE = "To appear in 7th Conference Proceedings of Japan Society for Software Science and Technology, in Japanese", MONTH = "September", YEAR = 1990} @TECHREPORT{ TR-90-9, AUTHOR = "Yasuaki Honda and Shigeru Watari and Ei-Ichi Osawa and Mario Tokoro", TITLE = "{A Model for Growing Objects}", NUMBER = "SCSL-TR-90-009", INSTITUTION = scsl, NOTE = "To appear in 7th Conference Proceedings of Japan Society for Software Science and Technology, in Japanese", MONTH = "September", YEAR = 1990} @TECHREPORT{ TR-90-10, AUTHOR = "Kiyoki Ohkubo", TITLE = "{RemoteDoesNotUnderstand: solves several problems}", NUMBER = "SCSL-TR-90-010", INSTITUTION = scsl, NOTE = "To appear in 7th Conference Proceedings of Japan Society for Software Science and Technology, in Japanese", MONTH = "September", YEAR = 1990} @TECHREPORT{ TR-90-11, AUTHOR = "Ei-Ichi Osawa and Mario Tokoro", TITLE = "{SocioAgent: A Society of Rational Speech Actors}", NUMBER = "SCSL-TR-90-011", INSTITUTION = scsl, NOTE = "To appear in 7th Conference Proceedings of Japan Society for Software Science and Technology, in Japanese", MONTH = "September", YEAR = 1990} @TECHREPORT{ TR-90-12, AUTHOR = "Yasuaki Honda and Shigeru Watari and Ei-Ichi Osawa and Mario Tokoro ", TITLE = "{Compositional Adaptation: A New Method for Constructing Software in Open-Ended Systems}", NUMBER = "SCSL-TR-90-012", INSTITUTION = scsl, NOTE = "Submitted to the European Conference of Object-Oriented Programming ", MONTH = "June", YEAR = 1991} @TECHREPORT{ TR-90-13, AUTHOR = "Ei-Ichi Osawa and Mario Tokoro", TITLE = "{Collaborative Plan Construction for Multiagent Mutual Planning}", NUMBER = "SCSL-TR-90-013", INSTITUTION = scsl, MONTH = "December", YEAR = 1990} @TECHREPORT{ TR-90-14, AUTHOR = "Shinji Kono and K. Tatsukawa and T. Aoyagi and Y.Kohda H. Tanaka ", TITLE = "{Object Oriented Load Distribution in DinnerBell}", NUMBER = "SCSL-TR-90-014", INSTITUTION = scsl, NOTE = "Presented at TOOLS Pacific 90 Sydney Australia ", MONTH = "November", YEAR = 1990} @TECHREPORT{ TR-91-2, AUTHOR = "Yasuhiko Yokote and Fumio Teraoka and Atsushi Mitsuzawa and Nobuhisa Fujinami and Mario Tokoro", TITLE = "{The Muse Object Architecture: A New Operating System Structuring Concept}", NUMBER = "SCSL-TR-91-02", INSTITUTION = scsl, NOTE = "Also appeared in Operating Systems Review Vol.25, No.2", MONTH = "February", YEAR = 1991} @TECHREPORT{ TR-91-3, AUTHOR = "Shinji Kono", TITLE = "{Recoverable Concurrent Objects}", NUMBER = "SCSL-TR-91-03", INSTITUTION = scsl, NOTE = "Japanese Database Workshop '90", MONTH = "February", YEAR = 1991} @TECHREPORT{ TR-91-4, AUTHOR = "Fumio Teraoka and Yasuhiko Yokote and Mario Tokoro", TITLE = "{A Network Architecture Providing Host Migration Transparency}", NUMBER = "SCSL-TR-91-004", INSTITUTION = scsl, NOTE = "To appear in Proceedings of ACM SIGCOMM'91", MONTH = "September", YEAR = 1991} @TECHREPORT{ TR-91-5, AUTHOR = "Mario Tokoro", TITLE = "{Toward Operating Systems for the 2000's}", NUMBER = "SCSL-TR-91-005", INSTITUTION = scsl, NOTE = "To appear in the Workshop on OS for the 90th and Beyond July 1991, Kaiserslautern, Germany. Also appears in a volume of LNCS", MONTH = "May", YEAR = 1991} @TECHREPORT{ TR-91-8, AUTHOR = "Ei-Ichi Osawa and Mario Tokoro", TITLE = "{Collaborative Plan Construction for Multiagent Mutual Planning}", NUMBER = "SCSL-TR-91-008", INSTITUTION = scsl, NOTE = "To appear in Proceedings of The Third European Workshop on Modeling Autonomous Agents and Multi Agent Worlds, Kaiserslautern, Germany", MONTH = "August", YEAR = 1991} @TECHREPORT{ TR-91-9, AUTHOR = "Yasuhiko Yokote and Atsushi Mitsuzawa and Nobuhisa Fujinami and Mario Tokoro", TITLE = "{Reflective Object Management in the Muse Operating System}", NUMBER = "SCSL-TR-91-009", INSTITUTION = scsl, MONTH = "September", YEAR = 1991} @TECHREPORT{ TR-91-10, AUTHOR = "Chisato Numaoka", TITLE = "{Conversation for Organizational Activity}", NUMBER = "SCSL-TR-91-010", INSTITUTION = scsl, NOTE = "To appear in Proceedings of The Third European Workshop on Modeling Autonomous Agents and Multi Agent Worlds, Kaiserslautern, Germany", MONTH = "August", YEAR = 1991} @TECHREPORT{ TR-91-11, AUTHOR = "Shigeru Watari and Yasuaki Honda and and Mario Tokoro", TITLE = "{Situated knowledge in open systems: Morphe's approach}", NUMBER = "SCSL-TR-91-011", INSTITUTION = scsl, MONTH = "October", YEAR = 1991, NOTE = "To appear in the Seventh Brazilian Symposium on Artificial Intelligence, November 1991"} @TECHREPORT{ TR-92-3, AUTHOR = "Fumio Teraoka and Kim Claffy and Mario Tokoro", TITLE = "{Design, Implementation, and Evaluation of Virtual Internet Protocol}", NUMBER = "SCSL-TR-92-003", INSTITUTION = scsl, NOTE = "in Proceedings of the 12th International Conference on Distributed Computing Systems", MONTH = "June", YEAR = 1992} @TECHREPORT{ TR-92-4, AUTHOR = "Nobuhisa Fujinami and Yasuhiko Yokote", TITLE = "{Naming and Addressing of Objects without Unique Identifiers}", NUMBER = "SCSL-TR-92-004", INSTITUTION = scsl, NOTE = "in Proceedings of the 12th International Conference on Distributed Computing Systems", MONTH = "June", YEAR = 1992} @TECHREPORT{ TR-92-5, AUTHOR = "Shigeru Watari and Yasuaki Honda and Mario Tokoro", TITLE = "{Morphe: A Constraint-Based Object-Oriented Language Supporting Situated Knowledge}", NUMBER = "SCSL-TR-92-005", INSTITUTION = scsl, NOTE = "in Proceedings of the International Conference on Fifth Generation Computer Systems 1992", MONTH = "June", YEAR = 1992} @TECHREPORT{ TR-92-6, AUTHOR = "Fumio Teraoka", TITLE = "{Host Migration in Virtual Internet Protocol}", NUMBER = "SCSL-TR-92-006", INSTITUTION = scsl, NOTE = "in Proceedings of the 1st International Networking Conference", MONTH = "June", YEAR = 1992} @TECHREPORT{ TR-92-7, AUTHOR = "Katashi Nagao and Ei-Ichi Osawa", TITLE = "{A Logic-Based Approach to Plan Recognition and Belief Revision}", NUMBER = "SCSL-TR-92-007", INSTITUTION = scsl, NOTE = "Submitted to COLING-92", MONTH = "May", YEAR = 1992} @TECHREPORT{ TR-92-8, AUTHOR = "Katashi Nagao", TITLE = "{A Preferential Constraint Satisfaction Technique for Natural Language Analysis}", NUMBER = "SCSL-TR-92-008", INSTITUTION = scsl, NOTE = "To appear in Proc. of ECAI-92", MONTH = "May", YEAR = 1992} @TECHREPORT{ TR-92-10, AUTHOR = "Akikazu Takeuchi and Steven Franks", TITLE = "{A Rapid Face Construction Lab}", NUMBER = "SCSL-TR-92-010", INSTITUTION = scsl, NOTE = "To appear in Proceedings of The Third European Workshop on Modeling Autonomous Agents and Multi Agent Worlds, Kaiserslautern, Germany", MONTH = "May", YEAR = 1992} @TECHREPORT{ TR-92-14, AUTHOR = "Yasuhiko Yokote", TITLE = "{The Apertos Reflective Operating System: The Concept and Its Implementation}", NUMBER = "SCSL-TR-92-014", INSTITUTION = scsl, NOTE = "Also appeared in the Proceedings of the Conference on Object-Oriented Programming, Systems, Languages, and Applications 1992", MONTH = "October", YEAR = 1992} @TECHREPORT{ TR-92-15, AUTHOR = "Takao Tenma and Yasuhiko Yokote and Mario Tokoro", TITLE = "{Implementing Persistent Objects in the Apertos Operating System}", NUMBER = "SCSL-TR-92-015", INSTITUTION = scsl, NOTE = "Also appeared in the Proceedings of Object Orientation in Operating System 1992", MONTH = "September", YEAR = 1992} @TECHREPORT{ TR-92-16, AUTHOR = "Yasuaki Honda and Mario Tokoro", TITLE = "{Soft Real--Time Programming through Reflection}", NUMBER = "SCSL-TR-92-016", INSTITUTION = scsl, NOTE = "Also appeared in the Proceedings of IMSA '92 International Workshop on Reflection and Meta-level Architecture", MONTH = "November", YEAR = 1992} @TECHREPORT{ TR-92-17, AUTHOR = "Ei-Ichi Osawa", TITLE = "{Rational Choice of Actions in Agent Collaboration}", NUMBER = "SCSL-TR-92-017", INSTITUTION = scsl, NOTE = "Also in the Proceedings of The Second Pacific Rim International Conference on Artificial Intelligence (PRICAI'92)", MONTH = "August", YEAR = 1992} @TECHREPORT{ TR-92-19, AUTHOR = "Akikazu Takeuchi and Katashi Nagao", TITLE = "{Communicative Facial Displays as a New Conversational Modality}", NUMBER = "SCSL-TR-92-019", INSTITUTION = scsl, NOTE = "Also appearing in the Proceedings of ACM/IFIP INTERCHI, 1993 (Amsterdam, The Netherlands, 24 April - 29 April, 1993) ACM, New York, 1993", MONTH = "September", YEAR = 1992} @TECHREPORT{ TR-92-20, AUTHOR = "Chisato Numaoka and Akikazu Takeuchi", TITLE = "{Collective Choice of Strategic Type}", NUMBER = "SCSL-TR-92-020", INSTITUTION = scsl, NOTE = "To be presented in International Conference on ``Simulation of Adaptive Behavior (SAB92)'' at Hawaii in 1992", MONTH = "November", YEAR = 1992} @techreport{ TR-93-1, author = "Fumio Teraoka and Mario Tokoro", title = "{Host Migration Transparency in IP Networks: The VIP Approach}", number = "SCSL-TR-93-001", institution = scsl, note = "ACM Computer Communication Review, Vol.23, No.1", month = "January", year = 1993} @TECHREPORT{ TR-93-2, AUTHOR = "Yasuaki Honda and Mario Tokoro", TITLE = "{Object-based Concurrent Reflective Architecture for Time Dependent Computing}", NUMBER = "SCSL-TR-92-016", INSTITUTION = scsl, MONTH = "January", YEAR = 1993} -------------------------------END OF REPORT----------------------------