Chapter 44: Connecting Your Computer to a Network or Another Computer

Any time you have more than one computer in your business, household, or other location, you can see some significant benefits if you connect them together through a network, including file exchange, instant messaging, and shared access to printers, file servers, and the Internet. A computer connected to a network can be a very powerful communications tool.

This chapter explains some of the most important concepts related to computer networks and offers instructions for setting up the most important types of network connections on a computer running Windows.

How Networks Work

A computer network is a set of two or more computers that can exchange commands and data. Successful computer communication requires a physical connection among the computers, using wires, radio, light waves, or some other transmission media, and a common set of data formats, languages, and rules for signaling.

The worldwide community of computer designers and users has established a long list of networking standards for everything from direct cable connections to carrier pigeons. With a few unimportant exceptions, all the networks you're likely to use work in a similar manner-only the physical media that carry the signals are different.

Packet data

The first communications networks, including telegraph and early telephone systems, used direct physical connections between the origin and the destination. When you placed a telephone call, an operator at a switchboard (or later, an automatic switching system) would connect the wires from your telephone to the wires that led to the home or office of the person you wanted to call; for the duration of the call, there was a continuous link between the two telephone sets.

If the two parties were in the same town, or in the same neighborhood in a bigger city, the operator would plug the line from the calling party directly into the line to the person they wanted to reach. But if the two parties were not both connected to the same switchboard, the operator would connect to a second switchboard through one or more trunk lines, as shown in Figure 44.1. For as long as the call continued, there was a physical connection between the two parties' telephones, and the lines that connected them could not be used for any other call.

Figure 44.1: Early telephone service used a separate physical connection for every call.

CROSS-REF 

Chapter 5 talks about how your computer combines a series of bits into a sequence that corresponds to a specific letter, number, punctuation mark, or other character called a byte, and how it uses those bytes to move data around inside the computer's processor and chipset.

Note 

Long before computers, that same kind of bit sequence was used by news agencies to send text to Teletype machines through telephone and telegraph wires. That's why the Associated Press, Reuters, and the others are still called wire services.

Continuous connections work well for relatively slow, low-volume services, but as the demand for more connections increased, communications engineers developed methods for sharing a single circuit among many users. In a high-speed computer network today, the computer sending a stream of data breaks up the stream into smaller strings of binary data called packets, which the computer at the receiving end reassembles into a continuous message. Each packet moves through one or more switching centers that can hold it until it can establish a link to the destination (or to the next switching center). This method provides several benefits:

• It allows the receiving computer to examine the integrity of each packet and ask for another copy when it detects a problem.

• In a complex network with more than one possible route between the originating point and the destination, it permits the network to choose a different path if the first choice becomes unreliable during transmission.

• It can carry packets from more than one message through the same communications channel at the same time.

At any specific point in time, the signal on a communications circuit is either on or off. In order to use a circuit to transmit information, the people or machines at each end of the circuit must agree on these rules:

• The amount of time each bit occupies (the duration)

• The type of signal that is accepted as an "on" signal (also known as a 1 or 0, or a mark or space) rather than random noise

• The exact order of bits that corresponds to each letter, number, or other character

• A sequence of bits that identifies the beginning of each character string

It doesn't matter how you send those bits and bytes from one point to another. You can use flashing lights, smoke signals, two different audio tones, or a series of electrical impulses through a wire. As long as you use the same method to convert between useful information and data bits at both ends, you can use the channel to communicate information. When the communications channel is working properly, the information that goes into the transmitter is exactly the same when it comes out of the receiver.

Unfortunately, almost every kind of communications channel is susceptible to some kind of interference. It could be electrical noise caused by a lightning strike, interference from another communications circuit, dirt on an electrical contact, or fog between the smoke signals and the person reading them. When a modern communications circuit is moving tens of millions of bits every second, it doesn't take much to make some of the data unintelligible.

Therefore, each data packet usually includes some kind of error checking along with the original data. This is usually a standard string of bits called a checksum attached to each byte or each packet. If the checksum is not correct, the receiver notifies the transmitter, which sends the same byte or packet again.

In a complex network, each packet might move through several different communications channels. For example, you might send a file or an e-mail message from your laptop computer through a Wi-Fi link to the local area network (LAN) in your office, and from there through a series of switching centers on the Internet to the recipient's service provider and onward to their LAN and the recipient's computer. At each stage of this path, the network equipment might add or remove a string of routing information and error checking checksums to or from the beginning or end of each packet. Because the network has to carry all of this handshaking data along with the original message, the effective data transfer speed is always somewhat slower than the maximum capacity of a channel.

Fortunately, most of this process of dividing the data stream into packets, and adding error checking and other handshaking information occurs automatically, once you have configured your network interface, modem, router, and other network equipment. All you have to do is assign the right address to each file or message before you send it.

Network connections

Your computer can use several methods to join a network:

• You can connect to another computer as a remote terminal

• You can exchange data directly with a second computer (peer-to-peer networking)

• You can connect to a LAN and exchange data with all of the other computers connected to the same network

• You can connect through a LAN (a local area network in which all the computers and other connected devices are usually close together, and in which all the connected devices can exchange data with one another) to a wide-area network (WAN, a network in which the connected devices are connected through telephone lines or other communications channels) or the Internet

When most people talk about networking, they probably mean connecting a computer to a new or existing LAN, and through that LAN to the Internet. However, remote terminal connections and peer-to-peer networks still have their uses.

Remote terminals

Before everyone had personal computer, most computers used one or more terminals, usually connected to them through a serial data port. A terminal typically included a keyboard that sent data to the computer's input, and either a video display or a printer that received data from the output. A terminal that was located in a different room from the computer was called a remote terminal. If both the computer and the remote terminal were located in the same building, they were usually connected through a serial data cable. But it was also possible to use a pair of modems to connect the terminal to the computer through a telephone line.

A remote terminal can work with just about any kind of computer and operating system. Even though your own computer uses Windows, you can send commands through a communications link to machines running Unix, Linux, and most proprietary operating systems; as long as the host recognizes the commands, it accepts and responds to them.

CROSS-REF 

See Chapter 33 for more on Unix and Linux.

Today, remote terminals are mostly used for access to mainframe computers and time-sharing systems (a shrinking number of library catalogs also require remote terminal access). Most connections to remote computers use the World-Wide Web instead of a remote terminal connection, but you might still need to use a remote terminal when the system doesn't offer any alternative connection method.

When you type a command on the terminal's keyboard, the terminal sends the command to the host computer. For example, Figure 44.2 shows a login screen at The Well, a virtual community that operates on a Unix host. The host treats a connection through a telephone line or through the Internet as if it was receiving commands and returning data to and from a local terminal.

Figure 44.2: The Well accepts logins from remote terminals.

If the host computer is connected to a LAN, you can also use your remote terminal connection to gain access to connect the host to other computers through that network. If the host has an Internet connection, you can run many Internet programs (such as an e-mail client or an FTP file transfer program) from your remote terminal. Because you are using the command-line interface, or shell, to use the Internet, this kind of access is called a shell account.

Windows uses a utility called a terminal emulator to connect to a distant host and present the same appearance as a remote terminal. Most Windows terminal emulators include these features:

• Emulation of common terminal types, including VT-100, Wyse60, and IBM 3278, along with the generic ANSI terminal, among others

• Support for several file transfer protocols that specify the rules for moving data between computers, including Xmodem, Ymodem, Zmodem, and Kermit

• Automatic control of a modem

• Telnet support through a TCP/IP (Internet) connection, explained later in this section

• Preset connection profiles for quick links to frequently used hosts

• Access to the Windows Clipboard to cut, copy, and paste text between the terminal emulator and a text editor or other program

• A method for saving the text of a terminal session as a file or sending the file to a printer

The HyperTerminal program (Start All Programs Accessories Communications HyperTerminal) supplied with Windows is entirely adequate for simple connections, text entry, and file transfers. However, there's a better version available at no cost from Hilgraeve, the developer that created the program for Microsoft. You can download HyperTerminal Private Edition from http://www.hilgraeve.com/htpe.

If you're using a distant computer for more complex work (such as remote access to a Unix host or some other mainframe system), you might want to consider a third-party terminal emulator program with additional features, including scripts and emulation of the specific characteristics of more makes and models of terminals (such as codes produced by function keys and on-screen text in more than one color).

Telnet (short for telecommunications network) is the set of rules that the Internet and other networks use for text-based communication between a terminal and a host computer, or between two computers. When your computer is already connected to the Internet, you can use a telnet utility to log onto a remote host as a terminal, as shown in Figure 44.3. There's a very primitive telnet utility supplied with Windows; most terminal emulator programs (including HyperTerminal) include telnet services and allow you to use them as your default telnet program. To set up a telnet session, open Start Run and type telnet [name of host].

Figure 44.3: A terminal emulator (top) normally connects directly to a host computer. A telnet program (bottom) connects to a host through the Internet.

Most terminal emulator programs include an option that makes that program the default program that runs when you enter a telnet command. If you can't find that option, you can follow these steps to set your favorite terminal emulator as the default telnet program:

• Open My Computer and select Tools Folder Options from the menu.

• In the Folder Options dialog box, open the File Types tab. The window shown in Figure 44.4 opens.

  
  Figure 44.4: Edit the URL: Telnet Protocol setting to set a new default telnet program.

• Select the (NONE) URL:Telnet Protocol file type from the list. Click the Advanced button near the bottom of the dialog box.

• In the Edit File Type dialog box, select Open in the list of Actions (it's probably the only entry in the list). Click Edit to open the Editing action dialog box.

• In the Editing action dialog box, use the Browse button to open a window where you can select the new default program. Hyper Terminal is located at C:\Program Files\Windows NT\hypertrm.exe. Most other terminal emulator programs are located in a subfolder with the program name, within the Program Files folder. If you can't find the program, use the Windows Search tool to locate it.

• Use the Open button in the Open With window, and the OK buttons in the other open dialog boxes to close the dialog boxes. Save your choice.

After you have assigned a new default telnet program, that program opens and sets up a connection when you enter telnet [address] in the Start Run window.

TCP/IP connections

Unless some other kind of network is specifically identified, just about any time people talk about connecting a computer to a network, they're probably using the TCP/IP (Transmission Control Protocol/Internet Protocol) standards that are the core of the Internet and most modern LANs. There are some other network standards out there, but because everybody eventually wants to connect to the Internet, they probably use TCP/IP.

Most Windows connections to a LAN use either a wired connection through an Ethernet adapter and cable, or a wireless connection through a Wi-Fi access point.

Before you start to set up your network connection, you must obtain the following information from your network manager, your Internet service provider, or the installation manual for the network hub or router. Don't worry if you don't know what one or more of these items represents; it's essential to have the names and numbers correct, but the significance of each one doesn't matter unless you're a network manager:

• The name assigned to your computer

• The name of the workgroup that your computer will join

• The name of your domain, if any

• If the network uses static IP addresses, the numeric IP address assigned to your computer

• If the network does not assign them automatically, the numeric addresses of your network's Internet Domain Name Servers

• The numeric address of your network's default gateway to the Internet

• The numeric subnet mask address

If you're connecting to an existing network through an Ethernet cable, follow these steps to set up a network connection:

1. If a network interface adapter is not already in place, turn off the computer and install one in your computer. If you're using a desktop computer, use a network card on an internal expansion card or connect an external adapter through a USB port. If you have a laptop, use either a USB adapter or an adapter on a PC Card. Use the software supplied with the adapter to install the device driver.

2. When you turn on the computer after installing a network adapter, Windows should automatically detect the new hardware and install any device drivers or other software that isn't already in place. If Windows doesn't automatically find the device driver, use the software supplied with the adapter. If the adapter came with additional software, install it now.

3. Connect an Ethernet cable from the network interface adapter to the network switch, hub, or gateway router that connects to the other devices in your network.

If you're connecting through a Wi-Fi link, follow these steps:

1. Consult the manual or quick start guide supplied with your wireless adapter to learn whether you should install the control software before or after connecting the wireless interface adapter. If the manual instructs you to load the software first, install it now.

2. If the wireless network adapter is not built into your computer, install the adapter on a PCI expansion card, a PC Card, or a USB connection. If the computer is not already on, turn it on now.

3. Windows automatically detects the wireless adapter and loads the appropriate device driver. When the Windows installation is complete, load any Wi-Fi software supplied with the adapter that you haven't already installed.

4. Restart the computer. If the network interface detects one or more Wi-Fi access points nearby, Windows and the proprietary software asks if you want to connect.

If your network manager or Internet service provider has provided specific instructions for connecting your computer, follow those instructions. If not, run the New Connection Wizard from Control Panel Network Connections to assign the necessary addresses.

Hubs, switches, and routers

Your computer connects to the network through either a hub, a switch, or a router.

A hub is the least complex type of network connection. All of the networked computers and other devices (such as printers) connected to a hub are linked directly through a common connection point. When a hub receives a data packet from one of the devices connected to it, it sends a copy of that packet to all of its data ports, so every packet that comes into a hub goes back out through all of the hub's ports. The destination device accepts packets that contain its address, and all the other network devices ignore them. Figure 44.5 shows a hub connection.

Figure 44.5: A hub is a passive network connection point.

A switch is a more sophisticated connection point. The switch examines the address of each incoming packet and copies the packet to the port connected to the device with that address. Figure 44.6 shows a data switch.

Figure 44.6: A switch sends incoming packets to specific destinations.

In a low-volume network, there's not much difference between the performance of a hub and a switch. However, when there's a lot of network traffic, the switch moves data through the network more efficiently. The difference in cost between a hub and a switch is often insignificant, so a switch is usually the better choice.

A router is a connection point or interface between two networks, such as a LAN and an Internet connection. On each side of the router, it has a different address that identifies it as a member of the network connected to that side. So, for example, it could connect all the computers on a LAN to the Internet through a shared broadband connection to an Internet provider's WAN.

The computers and other devices on the LAN side of the router have addresses that are visible only to other devices on the same LAN. All the devices on the LAN appear to outsiders on the other side of the router with the router's address.

A router that connects a LAN to the Internet is often called a gateway router. Figure 44.7 shows the sequence of network devices in a typical small home or office network. A larger network is similar, but it probably has more nodes. The individual computers and printers connect to one or more switches (or hubs) to form a LAN, which connects to an Internet service provider's (ISP) WAN through a router. The WAN connection moves data through a modem that drives either a DSL line or a cable TV line. At the ISP, the WAN connects to the Internet through yet another router. Each switch, router, and modem is a separate logical device, but the switch and router or the router and modem are often combined into a single physical device. Other routers are combined with Wi-Fi access points to allow both wired and wireless connections, or with a DSL or cable modem.

Figure 44.7: A small network usually includes a switch, a router, and a modem.