Wide Area Information Servers

Archie will tell you where a file is, based on a name that you give it, but it can't help you search for information based on what's in the file. That's a job for an application called Wide Area Information Servers (WAIS, pronounced "ways"). WAIS was conceived by Brewster Kahle in the late 1980s, and was developed by Dow Jones, Thinking Machines, Apple Computer, and KPMC Peat Marwick as a joint project. Since then, Brewster has formed a company called WAIS, Inc., which is now developing WAIS as an information tool for corporations. But many WAIS databases and applications are still freely available on the Internet.

WAIS allows you to search for information in databases located on server computers. Think of WAIS as a sort of electronic reference librarian. When you ask it where you can get information on a certain subject, it searches databases and returns documents it thinks will help you. Now, the servers don't actually understand your question; they simply look for documents that contain the words and phrases you used. The documents can be pictures and sound as well as text. The neat thing about WAIS is that once you find some articles that fit the bill, you can ask WAIS to find more documents with those characteristics. WAIS tries to "listen" to the feedback you give it before continuing its search. If you're using a WAIS client, you can save your questions and ask WAIS to continue searching at regular intervals for updates to it, or only when you specifically ask for something.

The WAIS system is very powerful and covers a lot of territory. At least 600 databases (more being made available all the time) are on server computers all over the world. Here's just a small sampling of the information you have access to: poetry, sheet music indexes, science fiction reviews, journalism periodicals, and organic gardening. There are also archives of many mailing lists and USENET newsgroups searchable by WAIS.

You can query these servers several ways. The best way is to use an archie client program and specify the nearest server. For example, if you live in Sweden, use the archie.luth.se server. (See previous Archie section for instructions.)

If you don't have an archie client, you can telnet into most of these, and login as archie to use the service. Again, use the one closest to you. Once you're on, type help to get a list of commands. If you want to start searching for a file, type set search sub and then type find filename , where filename is the name of the file you're searching for. Archie will "think" for a while, and then produce a list of every place that has a file by that name. You can then have this list sent to you via email by typing mail your-email-address . When you're done searching, just type exit to get back to home base.

Location		Archie Server Name 
ANS server, NY, USA		 archie.ans.net   
Rutgers, NJ, USA		 archie.rutgers.edu   
AT&T, NY, USA			 archie.internic.net   
SURAnet, MD, USA		 archie.sura.net   
U. of Nebraska, NE, USA		 archie.unl.edu   
Australia			 archie.au   
Austria				 archie.edvz.uni-linz.ac.at   
Austria				 archie.univie.ac.at   
Canada				 archie.uqam.ca   
Finland				 archie.funet.fi   
Germany				 archie.th-darmstadt.de   
Italy				 archie.unipi.it   
Japan				 archie.wide.ad.jp   
Korea				 archie.sogang.ac.kr   
New Zealand			 archie.nz 
Spain				 archie.rediris.es    
Sweden				 archie.luth.se   
Switzerland			 archie.switch.ch   
Taiwan				 archie.ncu.edu.tw   
United Kingdom			 archie.doc.ic.ac.uk  

Accessing WAIS. As is the case for archie and Gopher applications, you can access WAIS either by using a client program running on your system or by remotely logging into a public client. There are client applications available for Unix systems, called swais and waisearch (xwais for X Windows systems).

You can try out a simple WAIS terminal interface by remotely logging in to sunsite.unc.edu; login as swais. When you login, it will ask for your terminal type; in most cases, you'll be emulating a VT100 terminal. Although this interface is very powerful, it's not very user-friendly. There are other options-- graphical clients are available for PCs and Macs. The Appendix lists where to obtain them and details how to get started in becoming a WAIS expert. In the meantime, you may have an occasion or two to search WAIS databases accessible via a Gopher or the WorldWideWeb interface (explained below).

WorldWideWeb

WorldWideWeb (WWW or W3) is a browsing and searching system originally developed by the European Laboratory for Particle Physics (also known as CERN). It allows you to explore a seemingly unlimited worldwide digital "web" of information. The WorldWideWeb is built upon the concept of hypertext and hypermedia, which links independent interrelated documents and pictures into a three-dimensional cyberspacious world. Almost every piece of information you look at provides you with pointers, or hooks, into other documents on related subjects. And these documents aren't just text--they can also be sound and images--so the WorldWideWeb is really a hypermedia information retrieval system.

The Web is a continuous distributed-information construction project; tens of thousands of people are adding knowledge to it daily by bringing up their own web servers, which provide content and links, or bridges, between documents. Servers are also referred to as pages or home pages; for example, you may hear someone say, "Visit the Dr. Fun home page located at SunSITE." Think of home page as a front door to the WWW. At the time of this writing, there were 2,500 web servers, and that number is growing every day.

The Web lets you embark on digital journeys travel information links by simply clicking (or selecting) highlighted words or phrases. Once you make a selection, a hyperlink is followed to the destination, a related document, which may also contain links to other documents (and so on, and so on). WWW does more than just let you browse--it also allows you to search for key words in certain documents.

This hypertext Web environment in a way mimics your thought processes. We don't think or learn in a linear fashion; most of our thought processes can be pretty random at times. For example, while you're driving to work, the song on the radio reminds you of a party you were at several months ago, and then you start thinking of Joe, who you saw at that party. You haven't seen Joe in a long time, and you decide to give him a call. A simple song on the radio led you to Joe's doorstep. So who knows what the Web will lead you to?

The WWW uses some special protocols. One of these is HTTP, which stands for HyperText Transfer Protocol. This protocol simply allows very quick network file transfer, and it's used in WWW browsers as a faster alternative to FTP. HTML, or HyperText Markup Language, is a very simple language used for basic formatting and presentation of hypermedia documents. You can use HTML to specify the format of the document, where the hypertext links go, where images and sounds go, and so on.

Getting Access to the WWW. There are a number of ways to plunge into the WorldWideWeb. The programs that provide an interface to the WWW are also known as browsers, which come in two flavors: terminal (text-based) and graphical. Your options depend on how you're connected to the Internet. If you're dialing in and using a slow modem, you're limited to the terminal clients; two well known ones are the CERN LineMode Browser and Lynx. If you've got a high-speed dedicated connection, you can probably use powerful graphical client applications, such as Mosaic, Cello, and TKWWW. The CERN browser, Lynx, and Mosaic are explained in this chapter. If you're interested in learning more about other browsers and servers, explore the Web for information about the Web (here's a URL to start: http://info.cern.ch/hypertext/WWW/status.html), or obtain the WWW FAQ (see the Appendix).

My wake-up call to online publishing came back in the Spring of 1989, when I first received a forwarded posting from dissident students in Tiananmen Square. Their first-person accounts represented my initiation to the instantaneous, unmediated communication the Internet makes possible. Used as a global publishing machine, this medium presents us with a fundamentally different, reader-driven and boundless means of reciprocal recorded communication which promises to redefine the one-way street we used to call "publishing."

Laura Fillmore, President, Online BookStore (OBS)

Public Telnet Clients. If you don't have a WWW client program installed on your computer, there are public browsers you can access via Telnet. You can try connecting to the granddaddy of them all, the WWW server located at CERN in Geneva, Switzerland. Just type telnet info.cern.ch (this is a really popular site, so you may have problems getting through). Another one that's publicly available is located at the University of Kansas. Type telnet ukanaix.cc.ukans.edu, and login as kufacts. Or you can try one located at the Hebrew University of Jerusalem in Israel: telnet vms.huji.ac.il, login as www. (The WWW FAQ lists some other public sites.)

The public telnet clients are good for evaluating the WWW, but they're heavily visited sites, so it's better to use a local client browser instead. To find out what's available on your system, ask your Internet provider.

CERN's LineMode Browser. This application is a very basic interface to the WWW for those users without graphical client capabilities. If it's available on your system, you simply type www and a start-up (home) document will be shown on your screen. You'll see numbers scattered throughout the document; these specify the links. To select a link, type the corresponding number.

At the bottom of the screen is a status line that tells you what type of document you're looking at--if there's a number range (for example, 1-15 equals 15 links), it means a hypertext document with links to other documents. If there are a lot of links in the document, you'll have to page through each screen to see all of them (since they can't all fit on the same screen). If you press {RETURN} (perhaps several times, if there are several screens), you'll see the rest of them. A help command lists all of the commands and a brief explanation. The last command option allows you to quit.

The very first thing you should do, of course, is type help to get a list of commands. There are some keywords that will help you navigate; for example, you can type home to return to the home (startup) screen, and back will put your browser in reverse, navigating backward on your Web path. If you follow a lot of links (by typing their corresponding numbers), and get yourself deeper into the Web, you can then back up to each link you referenced by typing back.

The other commands, top, up, and bottom help you navigate when you're looking at a document with a large number of links (150, for instance). At any time you can type up to look at the previous screen, to look at the next screen, top to return to the beginning of the document, and bottom to jump to the end.

You will definitely want to learn how to jump to other Web servers when you hear about them. The command to do that is go URL, where URL specifies the desired server. For example, if you wanted to peruse Wired magazine's web server, type: go http://wired.com.

Lynx. Lynx is available on Unix and VMS systems. Lynx differs from the CERN Browser in that it's a full screen-oriented application--you can use your arrow keys to position your "pointer" on the different links in the document. If Lynx is available on your system, you can start it up by typing lynx at the command prompt. You'll see a screen with text, a "home page," containing words and phrases that stand out (bolded or highlighted) from the rest of the text. Each of these phrases provides a hyperlink to another document, which may contain more links, and so on.

The three lines at the bottom of the screen contain helpful instructions. Here's a summary of some of the commands:

H)	Help		G)	Go to URL 
O)	Options		M)	Go to Main (Initial) Screen 
P)	Print		Q)	Quit

If you find a URL for a great Web server that you'd like to test-drive, you can jump directly to that also. Using the Wired magazine example above, you would type g http://wired.com.

Mosaic

The terminal client browsers will give you an idea of what's available in the Web, but keep in mind when you're using these that you don't have access to sounds, images, and movies, only to text-based documents. There are other ways to access the WWW besides a text-based terminal browser. Weighing in on the Client Mega Scale is the Mosaic hypermedia distributed information discovery and retrieval browser, developed by the National Center for Supercomputing Applications (NCSA) in Champaign, Illinois. The Mosaic client is a graphical interface to the WorldWideWeb of information, and is one of the neatest applications to make the Internet scene. An Internet jack-of-all-trades, Mosaic also has hooks and gateways into Gopher, USENET, WAIS, archie, and other front ends. It does just what its name implies--that is, show the Internet as a world made up of colorful, varied, and interesting pieces of information, including text, images, sound, and movies. Taking a Mosaic journey is a real trip--you can indulge all sorts of online whims and curiosities.

Many people confuse Mosaic with the WWW and vice versa. They are two separate but related entities; just remember that the WWW is a global hypertext world of information, whereas Mosaic provides an interface into this world (as well as into the other applications mentioned above).

When you fire up Mosaic, it presents you with a start-up screen, also known as a home page. As is the case for LineMode Browser and Lynx, this home page is actually downloaded from a server that is located somewhere on the Internet. If this is your first time, your home page may be downloaded from NCSA. Be sure to select the "demo document" to check out recommended must-see items. Keep in mind that your Mosaic client program is customizable; you can change your front door to open anywhere on the Internet. To begin your search, all you have to do is point and click on underlined (or highlighted) words and phrases. Every time you do this, a related document (which can be text, sound, or pictures) is downloaded and presented to you. Mosaic provides a "bread crumbs" feature, known as a hotlist (similar to Gopher bookmarks), that you can use to find the places you're interested in quickly without having to go searching through layers and layers of the Web every time you fire up the application. Mosaic will also let you specify a URL to visit directly--a useful feature when someone announces the availability of a new resource (look for a menu item that lets you "Open URL").

During a Web session, it's likely that you will jump around quite a bit from one home page to another. Mosaic keeps track of where you've been in a particular session, so you can easily navigate backward and forward. Look for arrow buttons on your screen that, when selected, will help you retrace your steps. There's also a menu that lists all the places you've been. To go back home, simply click on the "home" icon or select "home" from the menu. When you quit a session, all of these history features will be lost, so be sure to add the most interesting places to your hotlist.

What's New, Mosaic? Groping your way around the WorldWideWeb can be akin to getting dropped off in a strange city with no directions or maps. There are several resources that can help you find what you want in Webspace. A popular one is the "What's New with NCSA Mosaic" page, a regularly updated source of announcements of new Web servers and the latest WorldWideWeb developments at NCSA. This server is available via the NCSA Home Page (the default home page for many Mosaic applications), or by jumping directly to http://www.ncsa.uiuc.edu/SDG/Software/Mosaic/Docs/whats-new.html.

If you need to find a specific resource, you should check out the Internet Resource Meta-Index page, accessible via the NCSA home page, or by jumping directly to http://www.ncsa.uiuc.edu/SDG/Software/Mosaic/MetaIndex.html. Included in this page are pointers to subject catalogs (for WWW, Gopher, WAIS, and Telnet) and searchable indexes of WorldWideWeb servers.

Unfortunately, Mosaic is one of those "power user" applications. In other words, if your Internet access is obtained by a PC or Mac and a dial-up line, you may not have the patience to sit around while it inhales the huge image, sound, and movie files off the net for your viewing and listening enjoyment. If you've got a really fast modem (14.4Kbps), and you're using SLIP or PPP (see Chapter 6 for more information), you can participate, but it's pretty slow, and it's easy to start imagining the application huffing and puffing while it's lifting large documents off the net. Your Mosaic application has an option that lets you turn off automatic image-download feature--in this case, you'll just download text without any graphics, which is much faster, especially for dial-up SLIP/PPP links. In order to hear and see everything, Mosaic requires a lot of "pieces/parts"--meaning, you need "external" (to Mosaic) viewer and player applications installed on your workstation to see and hear what's out there. And if that isn't enough, your workstation needs a pretty big engine, ample disk space, and a lot of memory. Assembling and installing all these parts, if they're not already available on your workstation, will take a while and might cause you some frustration.

Mosaic has been hailed as the "killer app" (killer, or way-cool, application), a portent of Internet interfaces to come. Many businesses are considering using it to provide easy access to their company product information and services, or for providing commercial information services. They won't have to convince people. Mosaic is so popular right now that it's being blamed for traffic jams and bottlenecks on the Internet.

Mosaic Clients and Web Servers. If you do have all the right stuff and want to test-drive Mosaic, you can download the client applications via anonymous FTP from the ftp.ncsa.uiuc.edu host in the Mosaic directory. Clients are available for Unix workstations running the X Window system, Apple Macintoshes, and PCs running Microsoft Windows.If you're interested in publishing (serving) information that other people can access, there are introductory documents available on this very subject via the WWW. To get started, check out the WorldWideWeb Initiative Page: http://info.cern.ch/hypertext/WWW/TheProject.html. To learn more about publishing your own info, start with the WWW and HTML Developer's JumpStation: http://oneworld.wa.com/htmldev/devpage/dev-page.html.

As you roam the Internet, you'll definitely get the sense that there's a culture and a shared history--things that people just "know." So that you don't feel left out, Chapter 5 will give the flavor of Internet culture, review some of what's gone on before you made the scene, share some "insider" information about security, and tell you where in the network world you can go to get help.

Copyright © 1994 by Tracy LaQuey and Editorial Inc.