IRList Digest Saturday, 30 April 1988 Volume 4 : Issue 22 Today's Topics: Query - Libraries of the future - Sample queries needed for thesis - Distributed information systems Email - Roy Rada address and workshop on Computer and Writing Discussion - Computers and Writing, IRList and Hypertext - Retrievable hypertext report Interest - Speech recognition and language modelling Call for Papers - AI and Hypertext: Issues and Directions News addresses are Internet or CSNET: fox@vtopus.cs.vt.edu BITNET: foxea@vtvax3.bitnet ---------------------------------------------------------------------- From: well!sac@lll-crg.llnl.gov (Steve Cisler) Subject: Libraries of the Future Keywords: library, information retrieval, future Date: 13 Apr 88 12:32:42 GMT An Inquiry Into Libraries of the Future ----------------------------------- I am going to be working on a number of projects dealing with the concept of libraries of the future. I would like to have readers of this newsgroup discuss their ideas about what libraries will and will not be. If you post a response, it would be helpful if you state how you currently use or don't use libraries. Some information junkies satisfy their habit without ever stepping foot inside a library; others have people to find the information for them. Many of us visit and make use of libraries several times a week. To start this off I'll introduce myself: I am automation manager in a large county library in the San Francisco Bay Area, but I will be moving to a corporate library of a computer firm and will be dealing with both engineers and librarians involved in projects that will affect the future technology in libraries of all sorts: academic, school, public, private, and government. I also publish a newsletter called Connect: Libraries & Telecommunications. I am convinced there will be extraordinary gains in the areas of information retrieval, telecommunications, and expert systems. But with terabytes of text, images, and sounds available through fiber optic cable and manipulated by 'software agents' what will librarians be doing? How far down the economic ladder will these technologies penetrate? Will we maintain the idea of Free Public Libraries in the 21st century, or will it be pay-as-you go? Feel free to focus on your own special interest or raise more questions about the big (and still fuzzy) picture. Steve Cisler 625 Barnacle Way Suisun, CA 94585 7070 422 5089 ------------------------------ Date: Thu, 21 Apr 1988 22:38-EDT From: Michael.Mauldin@NL.CS.CMU.EDU Subject: Need sample queries for my thesis For my PhD thesis I have implemented an information classification and retrieval system (called FERRET) that reads (among other things) the SCI.ASTRO newsgroup. For a comparison study, I need sample user queries to this system, and I am appealing to the newsgroup readership for help. In return for a few minutes of your thinking time you will earn my undying gratitude and a mention in my acknowledgements section. The study will compare the retrieval effectiveness of my program against a boolean keyword retrieval for a database composed of StarDate radio scripts previously posted to this newsgroup. Thus general questions about astronomy (especially of spacefilght and the local solar system) are appropriate. I'd like about 5 different questions from each contributor, plus a list of keywords you'd try to use to find the answer from a regular keyword based/free text searching system. For example: 1. What comets come near the earth ? [ comet near earth ] 2. What space probes were sent to Jupiter ? [ jupiter (OR probe spacecraft satellite) ] 3. Who discovered Neptune ? [ neptune discover ] Don't worry too much about whether the questions you ask make sense for the database -- I need samples of questions that you would like an astronomy database to be able to answer. And 5 questions is not a maximum, you're more than welcome to inundate me with sample questions, as simple and as complex as you'd like. If you *don't* want me to mention your name in the acknowledgements section, please indicate that in your reply. Thank you very much for your help. Michael L. Mauldin (Fuzzy) Department of Computer Science Phone: (412) 268-3065 Carnegie Mellon University ARPA: Michael.Mauldin@NL.CS.CMU.EDU Pittsburgh, PA 15213-3890 UUCP: ...!ucbvax!nl.cs.cmu.edu!mlm ------------------------------ From: dartvax!eleazar!ericb.UUCP@seismo.css.gov (Eric J. Bivona) Subject: distributed information systems Summary: what are people doing? Date: 27 Apr 88 00:32:39 GMT Organization: Dartmouth College, Hanover, NH I'm curious about what work people are doing concerning various information systems. To start things off, we here at Dartmouth are looking to do a pilot project to create a standard Macintosh interface that can access a variety of Information Resources that might exist on the campus network. Resources we are considering: Online card catalog Encyclopedia, dictionaries, thesaurus Sociology/Census Campus events Campus directory Full text of articles or abstracts Gateways to other services (off campus) Many, many others... We are planning a Server/Client model, with a protocol to provide data representation (perhaps XDR) and query functions over a reliable byte stream, or a transaction based protocol. I'd be very interested in discussing the functionality necessary for this protocol to work, and be robust. This robustness is due to the fact that our model demands that the server translate the request into the local DBMS query language. It is of course possible for the Mac/workstation to talk directly to the DBMS, if necessary, in which case the user interface (client) has to translate from mouse clicks/text directly to the DBMS language. I've looked at some of the ISO protocols for distributed database access (that have been proposed), but I've yet to see anything deal with data representation. I'm curious to know about current work in user interfaces to databases, general query languages (are there such things?), what other projects are working towards, etc. Mail or followups welcome! -Eric eric.bivona@dartmouth.edu ------------------------------ Date: Mon, 11 Apr 88 14:18:34 BST From: roy Subject: new address Thank you for sending the irlist. My new email address is sq45%ibm.liverpool.ac.uk@cunyvm.cuny.edu; snailmail is Dept. Computer Science, Univ. Liverpool, Liverpool L69 3BX, England; and phone is 44-51-790-6022 ext 2854. I'm participating in a workshop April 15 and 16 entitled "Computers and Writing" at Sheffield City Polytechnic, England--there will be a published report about it. ... Yours, Roy Rada ------------------------------ Date: Thu, 28 Apr 88 22:17:04 BST From: Roy Rada Subject: hyperdocument note A newsletter-type message with the following parts: a) recent meeting on computers and writing; ... d) IRList. a) An April 15-16 workshop entitled "Computers and Writing" was held in Sheffield, England. Attendees included Delin, Goodlet, Hardman, Rymaszewski, and Sharples. Interest in starting an electronic newsletter to continue the workshop spirit was expressed, and I volunteered to float a trial balloon. ... d) The IRList Digest often carries good hypertext news. Ed Fox, as editor of that Digest, what do you think of an electronic newsletter devoted to hypertext or an IRList supplement on hypertext ... or ...? Roy Rada BITNET/EARN/NORTHNET/GULFNET/ASIANET Address: SQ45@IBM.LIVERPOOL.AC.UK Internet (First choice) Address: SQ45%IBM.LIVERPOOL.AC.UK@CUNYVM.CUNY.EDU Internet (Second choice) Address: SQ45%IBM.LIVERPOOL@NSS.CS.UCL.AC.UK JANET Address: SQ45@LIVERPOOL.IBM UUCP Address: ....!MCVAX!UKC!LIV!SQ45 [Note: I am happy to include hypertext discussions in IRList - just send in comments. If someone decides to form a separate group, that is fine, but my definition of "information retrieval" includes hypertext and hypermedia as sub-fields. I would like to see more hypertext and hypermedia activities under the aegis of ACM SIGIR, and will certainly be helping get out products to ACM members in connection with the ACM Database Products effort. - Ed.] ------------------------------ Date: Mon, 11 Apr 88 03:24:32 PDT From: :@gnu.ucsf.edu (John Gilmore) Subject: Is a *retrievable* HyperText report available? Date: Mon, 29 Feb 88 15:22:49 DNT From: Jakob Nielsen Tech Univ of Denmark Subject: HyperCard stack with report on HyperTEXT workshop available > My report on the recent HyperTEXT workshop is now available > in a hypertext version in the form of a 400 K HyperCard stack. > To read it, you will need a Macintosh and APple's HyperCard program. Has anyone extracted the useful information from the gizmos and gadgets and produced a convenient text version of this document? I don't see the sense in mailing floppies around the world, and having to dig up a Mac and run a proprietary program, to read a simple report on a topic of interest. Please remember when producing materials for the worldwide network, that binary dumps of Mac "hype-card" stacks are not generally useful. At least produce something in a format that can be read on more than one manufacturer's machines -- or a format that is documented, so others can implement it on their machines. [Note: Jakob sent me ascii files as well, which I can resend. Yes, we need standards! Now it is my turn to give you a hard time: Speaking of retrievable, I wonder if your email address would let me send files to you. You would be amazed how difficult it is to reply back to IRList inquirers - I doubt my mailer can handle it if your user name is simply ":" as is suggested by the address above. Please make it easy on mail recipients to reply by "signing" submissions as I request in the Welcome message, with an address labelled as "internet" or "bitnet," in canonical form. Thanks, Ed.] ------------------------------ From: Bitnet%MSSTATE::LBOGGESS 15-APR-1988 15:51 To: FOXEA Subj: interests Hello - My name is Lois Boggess (Bitnet address LBoggess@MSState). For some time Tom English and I have been doing research on speech recognition and language modelling (a lot of it probabilistic), with with strong interests in natural language processing and dictionary analysis, as well as applications for the speech handicapped. ... Thanks. ------------------------------ Date: Fri, 8 Apr 88 10:35:43 EST From: Bob Allen Subject: Workshop on AI and Hypertext AAAI-88 WORKSHOP: AI AND HYPERTEXT - ISSUES AND DIRECTIONS Tuesday, August 23, 1988, St. Paul, MN OBJECTIVES The development of practical hypertext systems has evoked new interest in hypermedia through much of the AI community. Concurrently, progress in knowledge representation, user models, natural language synthesis and understanding, and in qualitative reasoning all promise to enhance the scope and utility of hypertext documents. Research into development and utilization of massive knowledge bases is of intense interest to both disciplines. The AAAI-88 Workshop on Artificial Intelligence and Hypertext will explore novel and controversial issues at the frontier of AI and hypertext research. Suitable topics include, but are not limited to: * automatic creation of hypertext from linear documents * user models and adaptive documents * integrating hypertext and heuristic systems * truth maintenance, argumentation, and collaborative writing of text and of programs * design, development and utilization of large knowledge bases and docuverses This half-day workshop is intended to promote interaction among leading researchers and practitioners. Several brief position statements will introduce central issues, to be followed by extensive general discussion. ATTENDANCE To promote lively and candid interchange, workshop attendance will be limited to 35 participants. Invitations to participate in the workshop will be extended on the basis of a position paper, outlining the writer's relevant work in, and positions on, the hypertext/AI frontier. ORGANIZING COMMITTEE Mark Bernstein, Eastgate Systems, Inc. K Eric Drexler, Stanford University Steven Feiner, Columbia University REQUIREMENTS The deadline for submitting position papers is May 1, 1988. Online submissions will not be accepted; hard copy only, please. Position papers should not exceed four pages. Send position papers to: Mark Bernstein Eastgate Systems, Inc. PO Box 1307 Cambridge, MA 02238, USA (617) 782-9044 Invitations to participate will be extended in early June. ------------------------------ END OF IRList Digest ********************