Date: Sun, 15 Sep 85 18:58 EST To: irdis at vpi Subject: IRList Digest V1 #8 IRList Digest Sunday, 15 Sep 1985 Volume 1 : Issue 8 Today's Topics: EMAIL - Error in Issue 7, problems with some addresses Query - Is there a sudden interest in IR after years of lull? Announcment - Report on first annual AAAI-ED meeting - Smalltalk for the Macintosh Article - The Interactive Encyclopedia System (TIES) - (additional comment) Value of Touchscreen ---------------------------------------------------------------------- From: "Edward A. Fox" Date: Sun, 15 Sep 85 17:45:48 edt Subject: Errors - number of last issue, mail addresses 1) This is Issue 8 whereas the last one was 7. Please correct the second mention of Issue number in your last digest. 2) Some of you may not have been receiving IRList due to changes in CSNET site lists, handling of domain names, etc. Please advise if you missed issues and cannot obtain them from other more easily. ------------------------------ From: William HATFIELD <686@njit-eies> To: fox%vpi.csnet@CSNET-RELAY Subject: Question Sudden Interest in IR Message-ID: Ed I have the impression of a recent resurgence of interest in IR, and maybe also more work, papers, etc., in the last YEAR after a lull throughout the 1970s. And maybe even more off the wall: that there's been little fundamental research in IR since the mid-1960s. Do you know of any basis for this, have a similar (or different) impression? regards, wm%NJIT-EIES.MAILNET@MIT-MULTICS.ARPA [Comments: 1) See the short article by G. Salton in July issue of JASIS for a survey of some good work done during the time you indicate there was a lull. 2) I personally do feel there has been a lot of activity recently, and that the trend continue. IR and AI have found each other, and there are many cross interests with other areas too: office inf. systems, new processor and storage hardware, networks, etc. 3) The distribution list for IRList is still growing! Does anyone have firm evidence in terms of memberships, papers submitted, etc.? - Ed] ------------------------------ From: Mark Richer Date: Mon 2 Sep 85 16:15:26-PDT Subject: Report on first annual AAAI-ED meeting [Copied from AI-ED] Well it's about time I tell you something about the AI in Education meeting (AUG. 20TH, 1985) that was held at the IJCAI conference in L.A. (For those new to this list and unfamiliar with IJCAI there were several earlier announcements about a special meeting that WAS held at the International Joint Conference on AI). --- There were at least 50 people at the meeting --- Jeff Bonar from LRDC (Learning Research & Development Center, University of Pittsburgh) led a discussion on Intelligent Tutoring Systems. --- Gary Fine from Intellicorp was supposed to speak about commercial realities, but couldn't come because of a business meeting (perhaps that says something about current commercial realities itself). --- Jeff's talk raised a lot of interesting questions and discussion. It was a lively group with lots of heated debate on several topics: --- diagnosis of student errors (how much diagnosis is really needed, useful, or practical?; Are bug catalogs sufficient in most domains?) --- what are the relative merits of traditional CAI, ICAI, other alternatives ("tool kits") --- evaluation of AI/Education programs (or lack of evaluation so far) --- what should we be teaching? how can computers best be used in education? where does AI fit into the picture in terms of developing useful educational programs? Augmented CAI (ICAI), Socratic Tutor (ITS), microworlds (papert, diSessa, et al.)? --- general architectures for AI/Education technology; domain independent tools; methods for formally specifying a curriculum; tying instructional actions to diagnosis --- can we develop an ITS Authoring System? It seemed that the meeting was a definite success and there was a consensus that a meeting at AAAI 86 (annual conference of the American Association of AI) would be a good idea. IJCAI summary ----- ------- Advanced Computer Tutoring, Inc. announced a lisp tutor for sale (at the exhibition hall) which is based on ITS technology developed at Carnegie Mellon (Anderson, et al.) The IJCAI 85 proceedings (over two thousand pages) had 6 articles focusing on AI and Education (28 pages), which is not a very impressive amount but if you are interested check in your local library or contact Morgan Kaufmann Publishers, Inc., 95 First Street, Los Altos, CA 94022 (415-941-4960) [I don't know the price] Articles in the proceedings: (1) the geometry tutor, Anderson, et al. (2) dynamic student modelling in an int. tutor for LISP pro, Reiser, et al. (3) heuristic and formal methofs in automatic pgm debugging, Murray (4) grammars as user models, Schuster (5) An ES for understanding expressions from electric circuit analysis, Joobbani, et al. (6) Computer assisted reasoning with MIZAR, Trybulec, et al. mark ------------------------------ Date: Mon, 26 Aug 85 17:53:19 pdt From: Mark Lentczner Subject: Smalltalk for the Macintosh [Copied from INFO-MAC digest] Just thought I'd clear things up, Smalltalk is now available for the Macintosh (yes, not "MacLump", but real honest to god Macintosh, all you need is at least 512k, and of course it will run on the MacXL too). It costs only $50. The correct address to write to for an order form is: Smalltalk Request c/o Eileen Crombie Apple Computer, Inc. 20525 Marianni Avenue Cupertino, CA 95014 Please do not send anything to the other address posted until you have the order form. This is very important as the other address will not be able to send you order forms or send you Smalltalk without and [sic] order form. [We are not advertising here! Since much work on IR has focussed on browsing and human interfaces, it is important for us to all be aware of the price and power of current hardware and software. - Ed] If there is a call for it, I have a 102 line info file that I can post about Smalltalk on the Macintosh. -mark lentczner Smalltalk Group Apple Computer, Inc. 20525 Marianni Avenue, MS:22Y Cupertino, CA 95014 UUCP: {dual, nsc, voder}!apple!mark CSNET: mark@apple ------------------------------ Subject: From: Ben Shneiderman Date: Mon, 26 Aug 85 08:22:25 edt Hello Fox@vpi -- glad to hear about the idea of IRList. May I suggest your adding "Human-Computer Interaction" to your list of topic areas. [Good idea - I will edit the welcome message. Thanks, Ed] Would you like to include our notice about TIES?? [Sure - here it is in nroff output form. - Ed] The Interactive Encyclopedia System (TIES) --- ----------- ------------ ------ ------ Ben Shneiderman, Department of Computer Science Janis Morariu, College of Library and Information Services University of Maryland, College Park, MD 20742 June 1985 The Interactive Encyclopedia System (TIES) has been under development at the University of Maryland since Fall 1983. It allows novice users to explore information resources in an easy and appealing manner. They merely touch (or use arrow keys to move a light bar onto) topics that interest them and a brief definition appears at the bottom of the screen. The users may continue read- ing or ask for details about the selected topic. An article about a topic may be one or more screens long. As users traverse articles, TIES keeps the path and allows easy reversal, building confidence and a sense of control. Advanced features include the ability to view an index of articles or print out arti- cles of interest. TIES is attractive for instruction (and entertainment) because the author's ideas and writing style are the focus of attention. Through careful human factors design, the computer aspects have been trimmed to let the author com- municate to the students and to allow the students to control their learning. TIES consists of: - Novice user browsing software - Authoring software for composing new articles and editing - Databases for browsing. Databases already created: Austria and the Holocaust (109 articles) Adele Stamp Student Union (42 articles) Introduction to TIES (18 articles) Introduction to Authoring on TIES (16 articles) FULCRUM Project (30 articles) Online Maintenance Manual (52 articles) TIES is appealing to authors because of the explicit instructional model, the reduction of computer-related concepts, the focus on content, and the lively user interface. It is an engaging challenge to reformulate pedagogic plans into the network of related articles that TIES supports. There is a great sense of satisfaction in composing articles and seeing the linkages come to life as they are used by students in novel ways. TIES allows authors to create a network of conceptual knowledge in which con- cepts are linked associatively and the learner is free to explore pathways based on their needs and interests. Potential applications include tutorials about software products, self-study guides for any discipline, travel guide- books, annotated Shakespeare or the Bible, and maintenance manuals for equip- ment. Each visitor suggests intriguing and novel applications. The TIES authoring software guides the author in writing a title, brief defin- ition (5-25 words), text (50-1000 words, typically), and synonyms for each article title. The author marks references in the text by surround them with a pair of tildes. TIES collects all references, prompts the user for synonym relationships, maintains a list of articles, and allows editing, addition, and deletion of articles. A simple word processor is embedded in the authoring software, but users can create articles on their own word processor, if they wish. There are no commands to memorize, every operation is done by selection from options on the screen. TIES runs on a standard IBM PC (monochrome or color) and on IBM PCs equiped with touchscreens. We are attracted to the possibility of eliminating the keyboard while still providing substantial exploratory power. TIES was first written in APL and has been rewritten in the C programming language. A brief user's guide and a more extensive author's manual are available. Dan Ostroff, a graduate student in computer science, did the implementation and a major portion of the user interface design. Three experimental studies have been conducted to test out certain design alternatives (such as demonstrating the advantage of arrow keys over the mouse for this system) and observe user behavior. More than 160 subjects partici- pated in these controlled experiments. In addition, more than a hundred novices and experts have tried and commented informally on the current design. In the study comparing the arrow keys (maybe better termed "jump" keys because the cursor would jump to the closest target in the direction pressed) to the mouse, the arrow keys proved to be and average of 15% faster and preferred by almost 90% of the subjects. We conjecture that when there are a small number of targets on the screen and when jump keys can be implemented, they provide a rapid, predictable, and appealing mechanism for selection. In a second study using the TIES technique, subjects traversed a database with 42 articles about the University of Maryland Student Union. The embedded menus technique reduced the number of screens viewed when compared with an explicit menu strategy. There were significant reductions in the times for task per- formance, and the subjective preference was strongly for the embedded menus. The embedded menus idea was also used for two experiments with online mainte- nance manuals for electronic equipment. A tree structured and linear form of a 52 page maintenance manual was prepared for screen presentation and in paper form. Experimental subjects had to perform 12 tasks using one of the manuals. Significant differences were found showing that time was reduced using the paper versions. No significant differences were found between the tree and linear versions for speed or error rates. When a pruning algorithm was applied to the text to allow users to trim text unrelated to their task, the time was cut in half. This latter experiment used only the computer condition and demonstrated one of the advantages of screens over printed text. This is important, since for many applications printed manuals are still easier to use and approximately 30% faster to read than computer displays. TIES is complete, but there are many refinements and many potential extensions which we would like to pursue. The software needs further documentation and the authoring guide could be expanded. Major extensions include support for graphics, videodisc, or integration with other software. Further testing is needed to select the optimum touchscreen or other input devices and to test alternate screens. We are also interested in further testing with the screen mounted horizontally inside a nicely built wooden table. We hope to attract the large fraction of the population that is anxious about using computers and would like to conduct field trials. ------- ------------------------------ From: Ben Shneiderman Date: Mon, 8 Jul 85 10:23:45 edt Subject: Re: TIES etc. Hi again...glad my info was useful. We did find touchscreen was easier to teach people than arrow keys, and that arrow keys were a solid winner against the mouse when there are a small number of large targets on the screen. Please let me know what you are up to. -- Ben Shneiderman ------------------------------ END OF IRList Digest ********************