IRList Digest Thursday, 10 December 1987 Volume 3 : Issue 47 Today's Topics: COGSCI - Mechanical design: meaning and representation - Reasoning under uncertainty, Universal theories of defaults CRTNET - Dissertation on paralanguage in electronic mail News addresses are Internet or CSNET: fox@vtopus.cs.vt.edu BITNET: foxea@vtvax3.bitnet ---------------------------------------------------------------------- Date: Wed, 2 Dec 1987 12:05 EST From: Peter de Jong Subject: Cognitive Science Calendar [Extract - Ed] Date: Sunday, 29 November 1987 15:07-EST From: Paul Resnick Re: AI Revolving Seminar; Allen Ward. NOTE UNUSUAL TIME Thursday 3, December 4:45pm Room: NE43- 8th floor Playroom The Artificial Intelligence Lab Revolving Seminar Series MECHANICAL DESIGN: MEANING AND REPRESENTATION Allen Ward Mechanical design has been much more resistant to automation than digital design, or mechanical analysis and drafting. Artificial intelligence methods offer both an explanation for and means to partially resolve this problem. Mechanical design programs, like all programs, work by changing arrangements of symbols. These symbol arrangements can be interpreted as representing sets of sequences of design and manufacturing operations, or equivalently the sets of physical artifacts those operations might produce. Mechanical design is hard to automate because these sets are diverse; because the traditional mathematical languages of engineers are unable to describe the sets in the quantitative detail required; and because the reasoning processes or design operations which change the symbol arrangements have not been clearly defined. I have developed languages for quantitatively representing specifications and components in a limited domain, that of power trains. I can use re-write rules to define a number of design operations. These operations can select components from catalogs, completing the design of power trains meeting an input set of specifications. ------------------------------ Date: Fri, 4 Dec 1987 12:31 EST From: Peter de Jong Subject: Cognitive Science Calendar [Extract - Ed] Date: Tuesday, 1 December 1987 16:11-EST From: Marc Vilain Re: BBN AI/Education Seminar -- Andee Rubin BBN Science Development Program AI Seminar Series Lecture REASONING UNDER UNCERTAINTY Andee Rubin Education Department, BBN Labs RUBIN@G.BBN.COM BBN Labs 10 Moulton Street 2nd floor large conference room 10:30 am, Tuesday December 8 Statistical reasoning is an important prerequisite for both ordinary and scientific thinking. Yet statistical reasoning is seldom taught to pre-college students, and when it is, the emphasis is often on formulaic manipulation, rather than on the concepts that are the foundation of reasoning about statistical matters. To address these concerns, we have developed, with funding from the National Science Foundation, a computer-enhanced curriculum in statistical reasoning called Reasoning Under Uncertainty that incorporates the ELASTIC (TM) software system. The course is designed to help high school students develop statistical reasoning abilities by using real world activities with which they have practical experience. The ELASTIC (TM) software, implemented on a Macintosh computer, is a tool for recording, representing, and manipulating statistical information. It has standard capabilities such as the ability to represent different types of variables and create appropriate graphs, including confidence intervals. Its most experimental features are three interactive programs: Stretchy Histograms, Sampler, and Shifty Lines, each of which allows students to interact directly with statistical graphics in order to achieve a deeper understanding of the underlying statistical concepts. The curriculum and software were field-tested in Belmont and Cambridge High Schools in the spring of 1987. The talk will describe and demonstrate the pedagogical principles underlying the course and software, some results of the field test, and our plans for future development. ================== Date: Thursday, 3 December 1987 10:30-EST From: Rosemary B. Hegg Re: Jon Doyle seminar on 12/8/87 SEMINAR SEMINAR SEMINAR DATE: Tuesday, December 8, 1987 TIME: Refreshments: 12:45PM Lecture: 1:00PM PLACE: NE43-8th floor playroom ON UNIVERSAL THEORIES OF DEFAULTS JON DOYLE Department of Computer Science Carnegie Mellon University ABSTRACT Though unifications of some of the numerous theories of default reasoning have been found, we bolster doubts about the existence of universal theories by viewing default reasoning from the standpoint of decision theory as a case of rational self-government of inference. Default rules express not only methods for deriving new conclusions from old, but also preferences among sets of possible conclusions. Conflicting default rules, which form the central difficulty in the theories, represent inconsistent preferences about conclusions. These conflicting rules arise naturally in practice, especially in databases representing the knowledge of several experts. We formally compare these theories of rational inference with theories of group decision making, and develop doubts about universal theories of the former by considering well-known negative results about the latter. HOST: Prof. Peter Szolovits ------------------------------ Date: Mon, 7 Dec 87 12:01 EST From: Tom Benson 814-238-5277 Subject: CRTNET 114 [Extract - Ed] Date: Sun, 6-DEC-1987 12:38 EST Janet F. Asteroff Subject: Computer-mediated Communication . . . I recently completed my dissertation on paralanguage in electronic mail, the abstract of which is appended to this posting. I found, among the 16 people I studied, many forms of "extra expression" in the form of "paralanguage." Ultimately, I documented enough differences between writing on the computer and writing through other media to identify it as "electronic paralanguage" with its own formal definition. Many people believe that face-to-face communication is the richest form of communication because of the variety of signals and channels, and as well the potential for channel redundancy. I have no problem with this assumption. I do, however, take issue with comparing other forms of communication to face-to-face and then judging any other medium as "information poor." Some scholars of computer-mediated communication carry this negative frame of reference over to their own work. While the computer may not provide as many channels as face-to-face communication, and the channel itself may be somewhat more limited, there is considerable research to indicate that computer users have done some interesting things to convey their meaning and message. Since I am not a fan of clogging up bboards with long messages, anyone interested in my work can contact me directly at Asteroff@cutcv1.bitnet and I will be happy to send you more material. The dissertation is also available through University Microfilms: Janet F. Asteroff, "Paralanguage in Electronic Mail: A Case Study." Teachers College, Columbia University, May, 1987. /Janet (Asteroff@cutcv1.bitnet) ABSTRACT PARALANGUAGE IN ELECTRONIC MAIL: A CASE STUDY Janet F. Asteroff This study explores the use of paralanguage in electronic mail communication. It examines the use of paralanguage according to the electronic mail and computing experience and technical expertise of 16 library science graduate students who fall into two groups by rank of experience, novice and advanced. These respondents used electronic mail in a non-elective and task-related situation to communicate with their instructor. This case study is based on a multi-level qualitative content analysis of the electronic mail exchanged between the respondents and the instructor, and the attitudes and experiences of the respondents about their use of electronic mail and computers. This research interprets the roles and functions of paralanguage in computer-mediated communication and explores the phenomenon as an indicator of certain kinds of expression. Paralanguage is a component of spoken, written, and electronic communication. It gives to what is being communicated a character over and above that which is necessary to convey meaning in the linguistic or grammatical sense. Paralanguage in electronic mail is positioned between spoken and written paralanguage in its visual and interpretive structures. Electronic paralanguage, a term developed to describe paralanguage in computer-mediated communication, is defined as: features of written language which are used outside of formal grammar and syntax, and other features related to but not part of written language, which through varieties of visual and interpretive contrast provide additional, enhanced, redundant or new meanings to the message. Electronic paralanguage is revealed to be a component of communication which in some situations showed substantial differences by the rank of the respondent, as well as differences in individual behaviors. Novice respondents used more paralanguage in more types of messages than did advanced respondents. Electronic paralanguage also provides a robust picture of the character of communication. The use of exclamation points by novice respondents in task-related messages showed that electronic paralanguage can in certain cases be a general measure of stress and experience, and as well is a precise indicator of different kinds of positive and negative psychological stress. ------------------------------ END OF IRList Digest ********************