Applesearch And The Internet

AppleSearch and the Internet January 1995

Contents Introduction

3

The Internet

4

Gopher ..................................................................................................................6 Wide Area Information Servers ( WAIS) ..................................................................7 World-Wide Web ( WWW) ........................................................................................7 AppleSearch and Internet Tools

10

AppleSearch and Gopher.......................................................................................10 AppleSearch and World-Wide Web ........................................................................11 AppleSearch and WAIS ...........................................................................................12 AppleSearch and Other Tools ................................................................................12 Conclusion

13

For Further Information

14

Introduction Connecting one computer to another, to hundreds of computers in a local or wide area network, or to millions of computers over the Internet has never been an easy task. In an ideal computing world, connectivity should be transparent, where reaching one or a thousand other computers is as easy as plugging in a cable and clicking on an icon. This ideal has guided Apple Computer in its efforts to make the Macintosh easy to connect in a local area network, from a remote location wirelessly, or over the Internet. Over the years, Apple has developed several products that further the “connectivity should be transparent” ideal, such as AppleSearch. An easy-to-use information search and retrieval application, AppleSearch is designed to be a personal assistant—a knowledge navigator that finds information relevant to users needs, prioritizes it, and keeps users updated if the information changes. AppleSearch integrates locally networked and Internet resources into the LAN environment. This application searches for specific information on local and wide area networks as well as data remotely dispersed on the Internet. On a local level, AppleSearch indexes information on servers and other networked storage devices. Information search requests are initiated by users via “reporters.” These reporters access information, retrieve and analyze the relevancy of their finds, and present the information in an easy-to-read format. The productivity benefits to organizations are immense. Archived electronic data, which was rarely accessible, becomes useful again to users on a LAN with AppleSearch. Users who need to track important changing data can do so with on-line news feeds and AppleSearch. And research-centric organizations can easily and cost-effectively search local servers, online data, and the Internet. On the Internet, AppleSearch works with Internet tools, such as Wide Area Information Servers ( WAIS), the World-Wide Web ( WWW ), and Gopher, in locating files and documents. AppleSearch works diplomatically with these Internet tools to bring Internet information into the local area network to be leveraged by all users. The following is an in-depth review of the Internet.*

* For those readers familiar with the Internet but not familiar with how AppleSearch interacts with it, the “AppleSearch and Internet Tools” discussion can be found on page 10.

Introduction

3

AppleSearch’s reporters harvest information from sources on local area networks or from the Internet, such as Wide Area Information Servers.

The Internet The Internet is a collection of computer networks, spanning over 80 countries around the world.1 There are over 36,000 computer networks connecting at least 2 million computers and some 25 million computer users to the Internet.2 It has been estimated that someone joins the Internet every 10 minutes.3 Internet users include faculty and students in colleges and universities; researchers in scientific laboratories; librarians; artists; teachers and pupils in grammar and high schools; journalists and other writers; and, bankers and employees of many commercial firms. In fact, about a fifth of the Fortune 500 companies have Internet access. Such a diverse audience means a wealth of ideas, a broad range of opinions, and differing levels of Internet experience and needs. With the population of the Internet nearly doubling every 12 months, the amount of information on the Internet is growing exponentially. How can we best measure this growth? Well, one measure of information on the Internet is called a packet. A packet contains a body of information being sent from one person to another over the Internet, enclosed in an electronic envelope with addresses for both sender and recipient. Packets can be up to 200 to 1,500 characters long.4 As you might suspect, nearly all files and messages mailed on the Internet are larger than just a single packet. Over the past decade, the number of packets mailed on just one part of the Internet, called the National Science Foundation backbone or NSFNet, has increased a great deal. In 1988, around 100 million packets were sent over this part of the Internet; by the end of 1993, over 35 billion packets or 9.2 terabytes were shipped. The total number of packets sent by the end of June, 1994 was in excess of 71 billion.5 Clearly there is a great deal of information on the Internet.

1

See “NSFNet Networks by Country,” August 1, 1994 in gopher://nic.merit.edu/Internet documents/connectivity/nets.by.country. Eighty-three countries are listed in this document. In Lawrence H. Landweber’s “International Connectivity,” version 11 of July 11, 1994, (gopher://nic.merit.edu/Internet documents/connectivity/world.list.txt) at least 152 countries enjoy connectivity; 86 do not.

2 “NSFNet Networks by Country,” August 1, 1994, (gopher://nic.merit.edu/Internet documents/connectivity/nets.by.country) lists 36,153 nets around the world. 3

Stephen Steinberg, “Travels on the Net,” Technology Review, vol. 97, no. 5 (July 1994), p. 22.

4

Ed Krol, The Whole Internet User’s Guide & Catalog. 2nd ed. Sebastopol, Calif.: O’Reilly & Associates, 1994, p. 26. John Markoff estimated packets averaging 200 characters in length. See Markoff, “Traffic jam already on the information highway,” New York Times, Nov. 3, 1993, p. D9.

5

From the “Internet Monthly Report for July 1994,” available at gopher://nic.merit.edu/Internet documents/newsletters/.

4

AppleSearch and the Internet

Numbers of Internet Hosts

Traffic for One Month in: 3,800,000 2,056,000

1,136,000 617,000 213

562

1981 1983

1,962

28,174

1985 1987

159,000

1989

313,000

1990 1991

1992 1993

1994

The number of computers and computer networks on the Internet is growing at a rapid rate.

How is information exchanged on the Internet? Electronic mail — a message sent by one person to another—is one way in which the Internet is used. Electronic discussion groups on the Internet number in the thousands and cover academic, arcane, and popular subjects of almost every imaginable stripe. These groups are managed by software on host computers tracking subscribers, lost mail, and transmission problems, and in some cases by moderators—on-line editors who try to keep electronic dialogues focused. One estimate suggested that at least seven million Internet users are involved in these discussion groups, and that at least forty thousand messages are handled each day by these groups.6 A site that decides to carry these news groups may see up to 1.5 gigabytes in traffic a month. Transferring files—documents, programs, graphics, audio—from one computer to another is yet another prominent use of the Internet. In fact, it has been reported that half of the activity on the Internet involves the transfer of files from one electronic location to another, with electronic mail composing only 20% of the traffic; this is not surprising, given that there may be as much as 6 terabytes or 6,000 gigabytes of files on the Internet.7 Files on the Internet can take several forms. Static files are posted to computers for downloading and never change; they may be documents, databases, and applications ready to use once they’re transferred from one location on the Internet to a local computer. Dynamic files are quite different, changing as new information arrives. These files may be satellite images of weather, catalogs of university and college libraries, press releases, and phone books, and they change frequently. Keeping up with these sorts of files is difficult without some sort of monitoring tool that recognizes the needs of the user and responds to alterations as they occur.

6

Steinberg, op. cit.

7

Paul Gilster, Finding it on the Internet. New York: Wiley, 1994, p. 10.

The Internet

5

Years

Billions of bytes

Millions of pages

1991 1992 1993 1994

1,268 2,761 6,053 11,226

181 384 865 1,604

Blue Skies, software from the University of Michigan, provides a graphic interface to meteorological information from the U.S. Weather Bureau and other agencies.

Given that the annual growth rate of data transmission is 30% and that 100 million people will be connected to the Internet by 1998,8 it is not surprising that the current batch of Internet tools are already swamped and experiencing incredible difficulties— at times— in doing their basic tasks. These basic Internet tools—Wide Area Information Servers ( WAIS), the World-Wide Web ( WWW ), and Gopher—were invented for a less crowded and less complicated Internet. A new sort of Internet facilitator is needed to work as a catalyst with traditional Internet resources and tools, to assist WAIS, WWW, and Gopher in locating information in a timely and efficient manner. This facilitator is AppleSearch. But first let’s examine how some Gopher, WAIS and WWW servers work. Gopher Internet Gopher was developed in 1991 at the University of Minnesota as a means for the students, faculty, and staff to find and distribute information on the campus networks, without knowing arcane computer commands. Gopher rapidly spread beyond Minneapolis to become one of the most popular ways to find and distribute information on the Internet. Over 200 billion bytes of information a month— the equivalent of a seven million page newspaper—are distributed by Gopher from hundreds of Gopher servers around the world.9 Information is made available with menus and icons for documents; moving from one Gopher server on one continent to another on another continent is simply a matter of double-clicking on a menu with TurboGopher, the Gopher client for the Macintosh. No need for remembering electronic addresses or special commands to sign on to another system. A Gopher client does not keep a connection to a Gopher server open continuously. The client opens a connection only when a user decides to, and keeps it open only long enough to retrieve the desired information, disconnecting as soon as the data is safely transferred. For users and system administrators, Gopher is simple to use and relatively efficient.

8

The figure of 100 million Internet users in the near future is cited in several sources; among them is Paul E. Hoffman, Internet instant reference. San Francisco: Sybex, 1994, p. xi.

9

Markoff, op. cit.

6

AppleSearch and the Internet

TurboGopher, the specific Macintosh client for Internet Gopher, was developed at the University of Minnesota. It takes full advantage of the Macintosh interface, so that users open menus by pointing and clicking, find menus and read dialog boxes, just as with any Macintosh-compatible program.10 TurboGopher also supports Gopher+ extensions, which allow for formatted files, security, and other features. Gopher’s success has meant that tens of thousands of files are available on over thirteen hundred registered Gopher servers (these servers might also be referred to as Gopherspace). Locating a specific document or server can be difficult. Some solutions to this problem have been developed. Bookmarks in Gopher clients help users locate useful menus without digging through layers of folders and documents and act as signposts, taking users back to resources when you need them. Veronica, another way to navigate Gopher servers, was invented in 1992 as a way to search all Gopher servers. Veronica indexes titles of files found on all of these servers on a weekly basis, providing a search engine in Gopher to its myriad contents. However, the rich abundance of information on Gopher servers around the world begs for another solution. Wide Area Information Servers (WAIS) WAIS was invented by the joint collaboration of Thinking Machines Corp., Apple Computer, KPMG Peat Marwick, and Dow Jones. WAIS is a tool that makes the contents of documents searchable over networks from a wide range of computing platforms and systems. It provides a single interface to search for information on diverse computers over the Internet. At least 700 WAIS servers provide indexed information to WAIS clients on everything from recipes to poetry. World-Wide Web (WWW) The World-Wide Web was created at the European Laboratory for Particle Physics (CERN), located just outside of Geneva, Switzerland, as a means to make terabytes of data, reports, scientific results, and surveys available to physicists and other researchers. World-Wide Web is the first truly global and successful hypertext application, linking information together across political borders, oceans, and even computer platforms on some 5,000 servers. To find information, you use a Web browser to access a server. A browser is a Web client program that displays information that may be text, graphics, or video from Web servers; it also provides links to other Internet resources. With a Web client, you access a document or file, read it, and watch for highlighted portions. These highlighted portions are embedded links that take you to other information in the Web or the Internet. If a portion of highlighted text is of interest, you just click on it and move immediately to another document or file. In the Internet world, these highlighted words are called “hypertext.”

10 TurboGopher can be downloaded from boombox.micro.umn.edu in the directory /pub/gopher/. For more information on TurboGopher, see Michael Fraase, The Mac Internet tour guide. Chapel Hill, N.C.: Ventana Press, 1993, pp. 185–209.

The Internet

7

TurboGopher, from the University of Minnesota, provides an organized, hierarchical, and graphic view of Internet information. Bookmarks save references to any number of the thousands of Gopher servers around the world.

Veronica, created at the University of Nevada, allows you to search the names of files and directories on Gopher servers. It will provide a list of matches and take you directly to servers which contain matched files and directories.

Wide Area Information Servers or WAIS provides access to hundreds of servers with indexed information, searchable in ordinary or Boolean language. These servers are available, as in this example, via TurboGopher or AppleSearch.

With the World Wide Web browser Mosaic, Le Louvre in Paris is accessible via the Internet.

Web servers provide views of bouquets from florists; rock stars; the Goodwill Games from St. Petersburg, Russia; southern California traffic; baseball; zebrafish; and movies of celestial collisions with the planet Jupiter. Traffic to these servers can be enormous; WebLouvre displays paintings from one of the world’s most famous art museums, providing over a million Web accesses to documents a month. Information on the Web is coded by URLs or Uniform Resource Locators beginning the HyperText Transport Protocol (or http). An URL is an address for a resource on the Internet, just like a street address refers to a home or business; with this address, a Web browser can find a file by simply clicking on a highlighted portion of text embedded with the proper URL. The first part of a URL describes how a file will be accessed; further information in the URL identifies a specific file and/or document. Links—the ways in which highlighted text in a Web document connects to an Internet resource that may be a text file, a QuickTime movie, or a Gopher—are created with HyperText Markup Language or HTML.

At the Jet Propulsion Laboratory at Pasadena, Calif., a Web server provided information from the Lab, NASA, and astronomical organizations around the world about the collision of Comet Shoemaker-Levy with the planet Jupiter.

HyperText Markup Language or HTML is a collection of markers or labels for text, graphics, and other digital elements. It allows a computer and a Web browser such as Mosaic to translate marked data into an electronic document which can be displayed and manipulated. This home page, viewed with NCSA Mosaic, describes the art of HTML file creation. It was created by Peter Flynn at the Computer Centre of the University College at Cork, Ireland (http://www.ucc.ie/info/net/html.html).

8

AppleSearch and the Internet

Browsers Browsers— client programs that create an interface to Web and Internet resources—exist for a number of platforms; two browsers for the Macintosh include MacWeb and Mosaic. Mosaic is justifiably the most popular implementation of the Web. It was created at the National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign; over a million copies of this program have been distributed by NCSA.11 Mosaic, on appropriately equipped computers, displays graphics, points out hypertext with color, plays audio files, and runs QuickTime movies. All of this dramatic multimedia comes at high expense in that it uses enormous bandwidth on the Internet. The vast quantity of information on the Web makes it difficult to locate specific files and documents. The European Laboratory for Particle Physics (CERN), near Geneva, Switzerland, maintains a directory of sorts called the WWW Virtual Library to information on the Web, but it is not all inclusive. NCSA issues a digital description of new servers and software called What’s New with NCSA Mosaic, which is useful but not completely comprehensive. There have been some efforts to create programs to search Web servers, applications known as robots or spiders.12 These robots create indices of Web servers. WebCrawler, for example, indexes the contents of documents while World-Wide Web Worm indexes title pages and URL contents.13 Standards are being developed for these tools, but at best, they provide only an incomplete view of World-Wide Web.

11 NCSA Mosaic was awarded one of the 1994 MacUser Shareware Awards. See Gregory Wasson, “The 1994 MacUser Shareware Awards,” MacUser, vol. 10, no. 10 (October 1994), p. 136. 12 For a general review of Web spiders, robots, or wanderers, see David Eichmann, “Ethical Web agents,” pp. 3-13, In: The Second International WWW Conference ‘94: Mosaic and the Web. Advance Proceedings, vol. 1. Chicago, 1994. 13 To use WebCrawler, connect your Web browser to http://www.biotech.washington.edu/WebCrawler/WebCrawler.html

The Internet

9

NCSA Mosaic is one of the most popular tools to access information on the Internet. It provides a view of text, graphics and video as well as a way to hear audio over the Internet.

AppleSearch and Internet Tools

AppleSearch connects to nearly seven hundred WAIS databases, and makes information from these databases available via the AppleSearch server to local AppleSearch reporters on the network. Reporters search selected WAIS databases for specific information with queries addressed to the AppleSearch server, its indices, and the remote WAIS databases.

When compared to traditional Internet tools, AppleSearch takes a different approach to selecting Internet resources. First, it is designed for a local area network, not an individual user. Second, AppleSearch has its own WAIS Gateway. The AppleSearch WAIS Gateway permits WAIS servers to be shown to users as AppleSearch information sources, just as if they were locally available hard disk or CD-ROM drives. These WAIS servers are available for search queries on demand or on schedule. AppleSearch administrators can select which WAIS servers will be made available to queries, based on the demands of the local networked community. Third, AppleSearch’s unique, personalized “reporters” actively look for information for LAN users. AppleSearch is unique in that it does not take the entire information universe on the Internet and index it for access. Operating locally in the LAN, AppleSearch examines only a relevant slice of the Internet. This unique approach results in efficiency and cost effectiveness for LAN administrators. AppleSearch is designed to examine files indexed on a WAIS server, but it also works with digital information created with Internet resources such as the Internet Gopher and the WorldWide Web. This capability in AppleSearch is possible thanks to extensions developed with AppleSearch’s Developer’s Kit. GopherSurfer, designed at the University of Minnesota, and AppleWebSearch, created at the University of Texas Health Science Center in Houston, are examples of the use of the Developer’s Kit to take advantage of the strengths of specific Internet tools. Let’s examine these tools and a few examples. AppleSearch and Gopher AppleSearch operates with the University of Minnesota’s Internet Gopher software by indexing documents organized in a specific Gopher server. AppleSearch provides a search engine through Gopher’s structure of menus and submenus to locate specific information within the text of documents quickly and effectively. For Gopher users, AppleSearch performs a special service by providing quick access to content. Even remote users of an AppleSearch-indexed Gopher server can enjoy the benefits of AppleSearch’s engine to find information buried in a Gopher structure. These remote users can use any computing platform to access AppleSearch and Gopher. For complicated Gophers with many documents and layers, this ability means that information can be found more quickly than ever before. Here are several examples of how Gopher and AppleSearch work together.

10

AppleSearch and the Internet

National Center on Adult Literacy (NCAL), University of Pennsylvania At NCAL, a FileMaker Pro database provides information on software related to adult literacy. This database points to 190 products available in the NCAL Gopher. Information was exported from this FileMaker database, and edited into individual records for AppleSearch. AppleSearch gives users an easier path to details on literacy software, supplementing both Gopher and FileMaker resources. Rather than search laboriously through long files, AppleSearch pulls up specific documents on demand.

AppleSearch is also being used at NCAL to sift through electronic mail related to literacy from Internet discussion groups and newsgroups. QuickMail sends messages from these groups to an archive folder, which in turn is indexed by AppleSearch. Users prefer AppleSearch as a tool to examine mail for specific topics in discussion groups rather than reading every piece of mail and commentary. Finally, AppleSearch is being used to create an index to research reports prepared by NCAL. University of Illinois at Chicago, Library and U.S. Department of State A test Gopher is being developed at the Library of the University of Illinois at Chicago for the U.S. Department of State14. This Gopher provides access to press briefings, speeches by President Clinton and Secretary of State Christopher, background notes, and other documents. Altogether, nearly 6 megabytes of information in nearly 150 documents are available in this Gopher. AppleSearch was added as an additional enhancement with Gopher Surfer on the Power Macintosh 7100 server in this experimental program. AppleSearch is readily available (at the top of the Gopher menu), permitting searches of the various files such as the State Department “Dispatches.” AppleSearch searches its indices of the Gopher files for terms that are entered into a standard dialog box of a Gopher+ compliant client. As the files on this Gopher grow, AppleSearch will continue to provide a quick way to locate new information and find archived documents.

AppleSearch and World-Wide Web Given that tens of thousands of documents are available in the Web, it is not surprising that there have been several approaches to examine and search these files. One of the problems in accessing Web documents by title is that 20% of items on the Web lack titles, while others contain titles that do not accurately reflect content.15 One approach to this dilemma has been developed at the University of Texas Health Science Center where AppleWebSearch provides indices of text in the Web.16 One specific application of AppleWebSearch was developed by Nick Arnett during his tenure at Multimedia Computing Corp. in Campbell, California. His server provides information about libraries and the Internet.17 His experiments with AppleSearch and the Web are inspiring an Internet-based library for Sarajevo, with support from the Soros Foundation. As with AppleSearch and Gopher, AppleWebSearch expands access to documents on the Web by providing an AppleSearch engine for searching. 14

See Edward J. Valauskas, “Creating Internet-accessible databases with AppleSearch and Gopher Server,” Database, vol 17, no. 5 (Oct.-Nov. 1994), pp. 99-102. 15

From Brian Pinkerton, “Finding what people want: experiences with WebCrawler,” p. 826, In: The Second International WWW Conference ‘94: Mosaic and the Web. Advance Proceedings, vol. 2. Chicago, 1994. 16

Contact AppleWebSearch’s author, Chuck Shotton, at [email protected] for further details. Subscribe to the LISTSERV http-talk for more information on AppleWebSearch development and usage. 17

Nick’s server can be reached at http://asearch.mccmedia.com/ Mr. Arnett now is the WorldWide Web product manager at Verity Inc. in Mountain View, California and can be reached at [email protected]

AppleSearch and Internet Tools

11

AppleSearch works as a search engine not only with WAIS databases but also with information on Gopher and Web servers. A Gopher server for the U.S. Department of State uses AppleSearch as a means to locate text in speeches, reports, and other documents related to foreign policy. The server can be found by pointing your Gopher client to dosfan.lib.uic.edu AppleSearch can be used on this server by any Gopher client operating on any computer platform. The server itself is a Power Macintosh connected to the Internet at the Library of the University of Illinois at Chicago.

AppleSearch and WAIS AppleSearch permits up to 100 of about 500 WAIS servers to appear as searchable sources, along with locally stored files on local and wide area networks. An AppleSearch administrator can add or delete WAIS servers from WAIS’ Directory of Servers and WAIS’ references in AppleSearch.18 AppleSearch and WAIS both use similar indexing strategies, and report the results of their results with relevancy rankings. AppleSearch’s reporters can schedule their sessions with WAIS servers at non-peak times, or interactively as the need arises. AppleSearch is configured with information on selected WAIS servers, making it possible for AppleSearch administrators to customize AppleSearch for the Internet needs of its users. AppleSearch’s unique approach to WAIS servers means that local users can schedule reporters on routine daily or weekly scans of specific servers. It also means that users are not confronted with a first step of searching WAIS’ Directory of Servers for likely topical servers, to be followed with a more lengthy second detailed search. This process of searching the Internet saves time and computing resources, and gives an opportunity to schedule Internet usage through a gateway during off-peak hours. In addition, one connection to the Internet services the entire LAN. AppleSearch and Other Tools There are other tools that filter through information such as gigabytes of articles from newsgroups. MacSlurp, a batch news transfer program for example, will download articles from specific newsgroups for local processing.19 MacSlurp works with ToadNews in this process; ToadNews unbatches news and files it away for reading with a text processor or a newsreader. ToadNews provides control for users over the flow of information from newsgroups as well.20 Another tool is InterNetNews or INN, a means to handle newsgroups and files for distribution. It operates in UNIX and Apple’s A/UX to help local users handle some of the 40,000 messages posted daily on 2,000 newsgroups, in a similar fashion to MacSlurp and ToadNews.21 Although these tools are useful in managing the quantity of articles from newsgroups, they do so without a query-based intelligence to sort and sift articles for importance, as in the application of AppleSearch to WAIS.

18

AppleSearch points to three directories for WAIS servers — quake.think.com; wais.com; and, cnidr.org

19

For further details on MacSlurp, contact Tom Davies at [email protected]

20

Contact John P. Mah at [email protected] for information on ToadNews 1.1.

21

An INN FAQ (Frequently Asked Questions) is available by sending an electronic message to [email protected] with the note ‘get file faq-inn-1’ on the first line.

12

AppleSearch and the Internet

Conclusion AppleSearch provides LAN users with several cures for the abundance of local electronic information as well as information on the Internet. It helps LAN administrators, with the WAIS Gateway, to bring appropriate information into LAN. And it helps users get the information they need. AppleSearch’s ease of use as a search engine and Internet resource locator make it attractive to a broad spectrum of users from the novice to the user who is completely familiar with UNIX or Boolean searching. For those users familiar with Internet information and searching techniques, AppleSearch will not disappoint them in its speed and functionality. Finally, AppleSearch provides good value for its cost, reducing staff time in learning the ways of Internet tools, in locating files and text, and sorting through search results. For administrators and systems operators, its value comes in reducing the need for excessive training and support on the Internet and switching traffic on Internet connections to off-peak hours. AppleSearch’s complementary style of searching and relevancy ranking, a heritage born in WAIS, will enhance the current crop of Internet tools. The use of AppleSearch’s reporters and indexing opens up the AppleSearch experience to non-Macintosh platforms, bringing the Apple advantage to many Internet users. AppleSearch may be one of the most important Internet developments in recent memory, bringing Internet administrators and users a powerful engine to cooperatively work with other tools to make the Internet more accessible.

Conclusion

13

For Further Information Jill H. and Matthew V. Ellsworth. The Internet Business Book. N.Y.: Wiley, 1994. 376 p. Susan Estrada. Connecting to the Internet. Sebastopol, Calif.: O’Reilly & Associates, 1993. 170 p. Michael Fraase, The Mac Internet Tour Guide. Chapel Hill, N.C.: Ventana Press, 1993. 288 p., with diskette of software. Paul Gilster. Finding It on the Internet. N.Y.: Wiley, 1994. 302 p. Paul Gilster. The Internet Navigator. 2nd ed. N.Y.: Wiley, 1994. 590 p. Ed Krol. The Whole Internet User’s Guide & Catalog. 2nd ed. Sebastopol, Calif.: O’Reilly & Associates, 1994. 544 p. Bryan Pfaffenberger. Mosaic User’s Guide. N.Y.: MIS:Press, 1994. 274 p.

14

AppleSearch and the Internet

Apple Computer, Inc. 1 Infinite Loop Cupertino, California 95014 © 1995 Apple Computer, Inc. All rights reserved. Apple, the Apple logo, AppleShare, AppleTalk, and Macintosh are registered trademarks of Apple Computer, Inc. AppleSearch, PowerShare, and PowerTalk are trademarks of Apple Computer, Inc. PowerPC is a trademark of International Business Machine Corp. MS Windows is a trademark of Microsoft Corp. L00850-A