NYU ON THE WORLD-WIDE WEB

Where Are We Now, Toto? On the Internet? On the Web? In Gopherspace? Or Are We Back in Oz?

by David Frederickson

[Ed: Links to web pages and/or e-mail addresses which have become inactive since the publication of this article have been enclosed in curly brackets { }. Replacement links have been provided where possible.]

Here we are, in the middle of the Information Age, zipping about on the Information Superhighway, feeling a bit ill-informed. Or are we on the Web? Or the Internet? What are these networks we keep hearing about?

Here's what they look like from the vantage point of the nontechnical editor's desk. Some of the networks are more or less real, and some are definitely less real. Let's start with the Internet.

The Internet

The Internet is more or less real. It is a name applied to all of the interconnected networks in the world (including NYU-NET), and theoretically, if you are sitting at a computer connected to the Internet, you can communicate with any other connected computer. This is possible largely because people around the world have agreed to use a set of conventions called TCP/IP (Transport Control Protocol, Internet Protocol) for transferring information over the connected lines.

The physical structure of the Internet is fairly amorphous. Until recently, the National Science Foundation financed a backbone of several major high-speed lines that crossed the US, which carried a major portion of the Internet traffic, but below that level the net was always rather messy; since NSF support was withdrawn early this year, the picture has become even more complex. The major high-speed lines, and many of the smaller ones, are leased from companies that provide long-distance telephone service; local networks like NYU-NET tend to be separate sets of lines, individually administered. About all that can be said is that there are actual wires and fibers out there, busily carrying pulses of electricity and light (and air carrying microwaves, as well), and that if the pulses follow the right protocols, that's the Internet.

But neither you nor your computer has to know much about the topology; when you send e-mail to your friend on the West Coast, there are hundreds of routes the message can take -- in fact, different phrases of your message may take different routes before being reunited at your friend's computer. The fact that there are many possible routes makes the system more reliable, since it doesn't depend on any specific intermediate points.

Exchanging Electronic Files

E-mail is only one of the uses of the Internet, of course. Sending and receiving information in longer electronic files is another use -- the main one, of network volume. There are several ways this can be done; the most important -- again, in terms of volume -- are FTP, Gopher, and World-Wide Web. Let's take them in that order, which is the order in which they gained acceptance and use.

All three are client-server systems, in which a given host computer houses a certain set of files and makes them available with a server program, and a matching client program running on any other computer can ping the files to that second computer. And each system is governed by protocols that assure the system will work.

File Transfer Protocol

FTP (File Transfer Protocol) is by now an old standby that has allowed thousands of researchers to transfer millions of files over the Internet. As with e-mail messages, a given file is packaged in small packets that wend their way separately from computer A to computer B, where they are reassembled to form a duplicate of the original file. The most popular manifestation of this protocol is Anonymous FTP, which allows a user at the client computer to log onto a distant host computer anonymously for certain limited purposes: to read the directories of available files, and to download any of the files to the client computer. The files themselves don't have to be specially formatted in order to be transmitted via FTP.

This system works well, but it is not very easy to navigate to the host FTP site. Various programs, such as Fetch and WinQVT/Net, make the procedure simpler, but you still have to know the specific address of the host computer and the files on it.

Gopher

The Gopher protocol and programs, developed at the University of Minnesota, made the whole process easier for the user. For the past two years, the NYU CWIS (Campus-Wide Information System) has used Gopher hosts to make a wide variety of information available to people with ACF accounts -- and indeed, to anyone around the world with an Internet connection and a Gopher client program. The user sees a menu, selects an item, and is presented either with another menu or with a document that can be read on screen or downloaded to the user's account or computer. During this process, the user is unaware that the client software may be making connections with a variety of different hosts around the world -- all of that navigating is done behind the scenes. The navigation is easy because when you select an item on a menu -- by typing a number or clicking a mouse cursor, depending on the particular client software you're using -- the program does the work of making the connection to the new file, and if necessary to a new host computer.

The interface is simple, even bare-bones. It is designed to work on modest computers, and it transmits text files that are presented in a character-based mode. Because text files are inherently short (a novel takes about as many bits to transmit as a page-sized picture), the system is swift -- as swift as the network connections will allow. The disadvantage is that anything to be made available via Gopher must be reduced to the simplest ASCII text, and will be displayed pretty much that way.

But the ability to navigate so easily is a tremendous advantage. Users began to experience the territory as a separate realm of the Internet called Gopherspace, where a separate set of rules made it easier to find and use information. Gopherspace was (and is) a virtual network, a new almost-equal-opportunity society defined by the programs that make the connections and purvey the information.

The World-Wide Web

The World-Wide Web (or Web, or WWW, or W3), is another virtual network with no separate physical identity. It is made up of the computers on the Internet that use a more complex set of conventions embodied in HTTP (Hypertext Transport Protocol); to talk to each other, the computers run client-server software that embodies the protocols. The Web and its protocols were developed about the same time as Gopher, but in a different community and with different objectives. The international community of particle physicists centered on CERN in Switzerland; such scientists had access to the largest and most advanced computers, and needed to exchange and organize information. As with Gopher, navigation on the Web is all done behind the scenes by the client programs, called powsers. Unlike Gopher, most powsers can display images in a graphical environment, though the original hypertext browsers were text-based, and were actually inferior to Gopher.

Lynx is a more modern text-based browser from the University of Kansas that is included in the NYU-Internet account. It will not display images, though you can download image files for later viewing with other software on your own computer. It actually runs on the ACF machines called is, is2, etc., rather than on your own computer, which acts as a terminal only.

What really changed everything on the Web was the release of Mosaic. Graphical powsers such as Mosaic (originally from the National Center for Supercomputing Applications at the University of Illinois) and Netscape (a commercial product of a group including some of the original Mosaic developers, which quickly became the most popular powser) display images, and can call up other programs to display compressed files, movies, and so forth.

This is luxury navigating, Internet surfing with all the bells and whistles. And the bells and whistles don't come cheap. When Carol Hutchins first wrote about Mosaic and the Web in the November 1993 issue of this publication, it was clear that it would increase network traffic hugely when it caught on. Not if; when: the Web was immensely attractive, and people would want to explore it and use it.

The Web takes ease of navigation even further than Gopher: Material can be presented in menus, much as in Gopher, but hypertext links can be embedded in text and images as well. No longer are you restricted to navigating an essentially hierarchical menu structure. Now you can click on a highlighted phrase in the text itself and get more information -- perhaps a definition, or an associated picture, or a connection to a computer halfway around the world where the phrase is treated in depth. Click on the musical instruments in a picture and hear each instrument play. Click on a person's name and be offered a form that you can fill out to send e-mail to her.

A splendid feature of the Web protocol is that it embodies others: You can use a Web powser to explore Gopherspace, without necessarily being aware that you're now using Gopher. Call for a file to be sent to you, and you're suddenly using FTP. The process is not without hitches -- it's not perfectly seamless, connections can be refused because of overwhelming traffic, addresses have a maddening ability to go out of date as people change their directory structures. In most respects, the system works. [ C ]

David Frederickson was the editor of Connect at the time of this article's publication.
{david.frederickson@nyu.edu}

Posted 18 October 1995. Revised 30 January 2004.