All You Ever Wanted to Know about WAIS -------------------------------------- From: composer@Beyond.Dreams.ORG (Jeff Kellem) Subject: WAIS, A Sketch of an Overview The following is something I threw together, in the time it took to type it, for some friends that had shown an interest in WAIS at one time or another. It may be of interest, for those that are curious about WAIS. Jeff Kellem Internet: composer@Beyond.Dreams.ORG WAIS, A Sketch Of An Overview ============================= Jeff Kellem, Beyond Dreams 23 September 1991 So, what is WAIS? ----------------- First off, WAIS stands for Wide Area Information Server. And, just what the heck is a "Wide Area Information Server"? It's a set of programs (or, more specifically, a protocol) that allows users to search and access different types of information from a single interface. This information can be practically anything, from text to sound to images to whatever you can think up. The information can reside anywhere and on many different computer systems. The WAIS protocol is an extension of the ANSI Z39.50 information retrieval protocol. (WAIS is pronounced "ways", BTW.) How does one access info from a WAIS server? -------------------------------------------- You'll need a WAIS client. There are a few sample clients (user interfaces) provided with the WAIS distribution available from Thinking Machines (TMC) [see below, "How do I get the WAIS distribution?"]: a GNU Emacs, a shell, and an X Windows interface. A Mac interface is also available from TMC. Other interfaces are currently "in the works", such as a Motif interface, one for the NeXT, a dumb terminal interface, among others. Once you start up the WAIS client, you specify what's called a source to search upon. [A `source' specifies a server of information and how the client can contact it, along with some other information.] You can ask multiple sources for information. Then, you ask the source(s) a question/query. A question consists of a phrase. With the current sample server/client implementations, this phrase is basically considered a set of keywords to search for, based on weights/percentages of each word in the documents (you're searching upon). But, since WAIS really just specifies the protocol for the client and server to use for communication, the underlying search on the server could just as well use various natural language queries upon its information. The information on the server could also easily be some arbitrary database, as long as the server knows how to understand a WAIS question and send back a response. The user never has to see this part of WAIS. All s/he see is the client side of things, the choice of a source(s) and the posing of a question. Once the user has found some relevant documents, s/he can ask the source(s) for other documents "similar" to the ones found. Well, what type of information is available? -------------------------------------------- All kinds and more are being added all the time. Here's a list of the current server names (taken from the release announcement of version 8 beta 3.1), organized by topic: Biology: biology-journal-contents.src biosci.src Molecular-biology.src Usenet and internet archives: comp.graphics.src comp.admin.src comp.db.src comp.emacs.src comp.multi.src comp.archives.src rec.pets.src comp.sources.src usenet-science.src user-contrib-cookbook.src usenet-cookbook.src homebrew.src info-mac.src sun-spots.src Frequently Asked Questions: NeXT.FAQ.src unix.FAQ.src ibm.pc.FAQ.src mac.FAQ.src Connection Machine info: CM-applications.src CM-tech-summary.src CM-fortran-manual.src CM-paris-manual.src CM-star-lisp-docs.src CMFS-documentation.src Books: jargon.src bible.src koran.src sample-books.src MIT-algorithms-bug.src MIT-algorithms-exercise.src MIT-algorithms-suggest.src Libraries: online-libraries.src tmc-library.src Misc: open_systems_calendar.src matrix_news.src astro-images-gif.src astro-images-fits.src nsf-bulletins.src midi.src unimelb-research.src (Australia!) supreme-ct.src NIH-Guide.src UNTComputerDoc.src US-Gov-Programs.src UiO_Publications.src (Norway!) cosmic-abstracts.src cosmic-programs.src directory-of-servers.src eff-documents.src eff-talk.src empire.src empire20.src internet-documents.src internet-drafts.src internet-resource-guide.src internet-rfcs.src patent-sampler.src poetry.src risks-digest.src sample-pictures.src unix-manual.src wais-discussion-archives.src wais-docs.src wall-street-journal-sample.src weather.src world-factbook.src How do I find out about these and other WAIS servers? ----------------------------------------------------- Thinking Machines runs a WAIS server, directory-of-servers.src, which is a "white pages" of other WAIS servers. You can query this WAIS server to find other servers and new ones that pop up. When someone creates a server that they want others to know about, s/he registers that server with the directory-of-servers. Directions on how to do this are given in the WAIS distribution from think.com. Okay, so how do I obtain the WAIS distribution? ----------------------------------------------- The WAIS distribution, currently at version 8 beta 3.1 is available via anonymous ftp from the: think.com:/public/wais directory. The filename for WAIS version 8 beta 3.1 is: wais-8-b3.1.tar.Z and, for the Mac client/interface: WAIStation-0-62.sit.hqx A bibliography of WAIS related articles is included below. If additions are made to this bibliography, you can obtain a fresh copy via anonymous ftp from: quake.think.com:/pub/wais/wais-discussion/bibliography.txt If you have further questions about WAIS, feel free to send me a note. What about mailing lists or newsgroups related to WAIS? ------------------------------------------------------- There are three mailing lists, maintained at think.com, regarding WAIS. They are: wais-interest@think.com announcements of new releases for the Internet environment. One message a month or so. wais-interest-request@think.com To join, send mail here. wais-discussion@think.com Moderated digest of mail from users and developers on Electronic Publishing and other WAIS related topics. Includes all from wais-interest. wais-discussion-request@think.com To join, send mail here. wais-talk@think.com unmoderated list for implementors and interactive discussions. A couple notes a week, depending on activity, of course. wais-talk-request@think.com To join, send mail here. As a reminder, requests for additions/deletions/etc. should go to wais--request@think.com where `' is one of `interest, `discussion', or `talk'. Archives are available from WAIS server, wais-discussion, or anonymous ftp from quake.think.com:/pub/wais/wais-discussion The USENET newsgroup, alt.wais, is also devoted to discussing WAIS and related items of interest. It is disjoint from the above mailing lists. If you have any further questions regarding WAIS, feel free to send me a note. I can be reached via e-mail at: composer@Beyond.Dreams.ORG Hope this was of some help. This document is an extreme rough draft, typing from a stream of consciousness. Jeff Kellem Composer of Dreams / Dream Maker Beyond Dreams Internet: composer@Beyond.Dreams.ORG [ The above document is Copyright (c) 1991 Jeff Kellem/Beyond Dreams, except for portions taken from other related documents, as noted. You have permission to freely redistribute this for non-commercial and non-profit purposes. It would be nice if you let the author know about any redistributions that are expected to reach more than a single person. :) Please contact the author if you wish to use this document in any other fashion. Most likely, there won't be a problem. ] WAIS Bibliography (attachment) ============================== Wide Area Information Servers (WAIS) Bibliography Barbara Lincoln (barbara@think.com) 8/23/91 ------------------------------------------------------------------ Hard copies of the following documents are available. Some documents are available electronically, as stated, but contain figures which are not reproduced in the ascii version. Email, fax, mail or phone your name, address and phone number to: Barbara Lincoln, Thinking Machines Corp., 1010 El Camino Real, Suite 310, Menlo Park, CA, 94025, phone: 415-329-9300, fax: 415-329-9329, email:barbara@think.com ------------------------------------------------------------------- WAIS Documents --------------- - "Wide Area Information Servers", (overview sheet) Brewster Kahle, Thinking Machines, April, 1991. One page overview of the Internet release of WAIS. Available via anonymous ftp: /pub/wais/wais-discussion/wais-overview.text@quake.think.com or WAIS server wais-discussion-archive.src - "An Information System for Corporate Users: Wide Area Information Servers", Brewster Kahle, April, 1991. Thinking Machines technical report TMC-199. Also in ONLINE Magazine Aug 1991. Report on the system constructed for Peat Marwick and other corporate users. Has screen shots of WAIStation. Available via anonymous ftp: /pub/wais/doc/wais-corp.txt@quake.think.com or WAIS server wais-docs.src - "WAIS Bibliography", Barbara Lincoln, Thinking Machines, August, 1991. Available via anonymous ftp: /pub/wais/wais-discussion/bibliography.txt@quake.think.com or WAIS server wais-discussion-archive.src - "Wide Area Information Servers Concepts", Brewster Kahle, November, 1989,Thinking Machines technical report TMC-202. Early draft of paper outlining how a Wide Area Information System could grow. Available via anonymous ftp: /pub/wais/doc/wais-concepts.txt@quake.think.com or WAIS server wais-docs.src - "Wide Area Information Servers: A SuperComputer on Every Desk", Brewster Kahle, Thinking Machines, August 1991. This is a set of overview presentation slides for WAIS. - "WAIStation, A User Interface for WAIS", February 1991, Thinking Machines technical report TMC-203. User interface documentation with screen shots. Available via anonymous ftp: /pub/wais/doc/wais_users_guide.txt@quake.think.com or WAIS server wais-docs.src WAIS-related articles/publications -------------------------------- - "For the PC User, Vast Libraries", New York Times, pg.C1 John Markoff, July 3, 1991. Overview of WAIS Internet experiment. - "Browsing Through Terabytes", Byte Magazine, Richard Stein, May 1991, pp157-164. Readable article on what a large WAIS system looks like. - "WAIS Promises Easy Text Retrieval", MacWeek, Henry Norr, May 14, 1991, pg22. Report on the Peat Marwick WAIS system. - "Release 1.0", Esther Dyson, April 1991, entire issue. In-depth article on commercial systems and protocols, featuring WAIS. - "Network to Unite Data Bases", San Jose Mercury News, John Markoff, July 21, 1991, page 1F. Rewriting of the "For the PC User, Vast Libraries," New York Times article with emphasis on Apple component. - "Designing a Desktop Information System: Observations and Issues", Thomas Erickson & Gitta Salomon. Human Factors in Computing Systems, CHI '91 Conference Proceedings (pp. 49-54) April 1991, New Orleans. New York: ACM, 1991. Early paper on the Apple interface for WAIS. - "Z39.50-1988: Information Retrieval Service Definition and Protocol Specification for Library Applications". National Information Standards Organization (Z39), P.O. Box 1056, Bethesda, MD 20817. (301) 975-2814. Available from Document Center, Belmont, CA. Telephone 415-591-7600. - "Z39.50-1991 Version 2", May 1991. Electronic version of the working copy of the Z39.50 implementors group. Anonymous FTP: /pub/wais/doc/z3950-spec.txt@quake.think.com or WAIS server wais-docs. - "The Z39.50 Information Retrieval Protocol: An Overview and Status Report", Clifford Lynch, Computer Communication Review ACM SIGCOMM Introduction to the the protocol of WAIS. - "WAIS Interface Protocol, Prototype Functional Specification", April 1990. Our Changes to the Z39.50-1988 specification. Thinking Machines Technical Reports ----------------------------------- - "Parallel Free-Text Search on the Connection Machine System", C. Stanfill and B. Kahle, 1986. Thinking Machines technical report TMC-72. Communications of the ACM, Volume 29, Number 12, December 1986. How a signature system can be used on a parallel machine. - "Massively Parallel Information Retrieval for Wide Area Information Servers", C. Stanfill, Thinking Machines, October, 1991. Paper presented at the IEEE International Conference on Systems, Man, and Cybernetics. Charlottesville, Virginia. Technical paper on Connection Machine servers on the WAIS system. Postscript copy available via anonymous ftp: /pub/wais/doc/cm-servers.ps@quake.think.com or WAIS server wais-docs. - "A Parallel Indexed Algorithm for Information Retrieval", C. Stanfill, R. Thau, December 1990. Thinking Machines technical report TMC-67. Appeared in Proceedings of the 12th International Conference on Research and Development in Information Retrieval SIGIR-89. - "Extracting Content Bearing Terms in Parallel on the Connection Machine", S. Smith, 1987. Thinking Machines technical report TMC-71. Submitted to ACL. - "Very Large Database Applications of the Connection Machine System", D. Waltz, C. Stanfill, S. Smith, R. Thau, 1987. Thinking Machines technical report TMC-70. Appeared in AFIPS/1987 NCC Proceedings, July 1987. - "Parallel Computing for Information Retrieval: Recent Developments", C. Stanfill, January, 1988. Thinking Machines technical report TMC-69. - "An Analysis of the Effects of Data Corruption on Text Retrieval Performance", S. Smith, C. Stanfill, December 1988. Thinking Machines technical report TMC-68. - "Information Retrieval on the Connection Machine: 1 to 8192 Gigabytes", C. Stanfill, R. Thau, December 1990. Thinking Machines technical report TMC-66. To appear in Information Processing and Management, 1991. - "WAIStation, A User Interface for WAIS", February 1991, Thinking Machines technical report TMC-203. User interface documentation with screen shots. Available via anonymous ftp: /pub/wais/doc/wais_users_guide.txt@quake.think.com or WAIS server wais-docs.src - "An Information System for Corporate Users: Wide Area Information Servers", Brewster Kahle, April, 1991. Thinking Machines technical report TMC-199. Also in ONLINE Magazine Aug 1991. Report on the system constructed for Peat Marwick and other corporate users. Has screen shots of WAIStation. Available via anonymous ftp: /pub/wais/doc/wais-corp.txt@quake.think.com or WAIS server wais-docs.src - "Wide Area Information Servers Concepts", Brewster Kahle, November, 1989,Thinking Machines technical report TMC-202. Early draft of paper outlining how a Wide Area Information System could grow. Available via anonymous ftp: /pub/wais/doc/wais-concepts.txt@quake.think.com or WAIS server wais-docs.src Electronic Services ------------------ - wais-discussion@think.com: Weekly digest of mail from users and developers on Electronic Publishing. Anonymous FTP access to archives: /pub/wais/wais-discussion/*@quake.think.com - wais-interest@think.com: announcements of new releases for the internet environment. A message every month or so. Anonymous FTP access to archives: /pub/wais/wais-discussion/*@quake.think.com - wais-talk@think.com: interactive list of developers. A couple notes a week. - WAIS Unix Release: Freeware release of client code (X and gnu-emacs), protocol code, and server code for accessing existing servers or starting new servers. Available via anonymous FTP: /public/wais/wais-8-b2.tar.Z@think.com - WAIStation Macintosh program: Freeware client program for accessing servers over tcp/ip. See WAIStation documentation above (included in the release). Available via anonymous FTP: /public/wais/WAIStation-0-62.sit.hqx@think.com - Macintosh demonstration screen-movie: Steve Cisler put together a short screen-recorder movie for seeing some of what WAIStation does. Available via anonymous FTP: /public/wais/WAIStation-Canned-Demo.sit.hqx@think.com