A TECHNOLOGY PRIMER
by: Uday S. Murthy, Ph.D., ACA and S. Michael Groomer, Ph.D., CPA, CISA
 

Data Communications and Networking

 

Learning Objectives

After studying this chapter you should be able to:

The dramatic technological advances that swept the computer industry in the seventies and eighties resulted in the development of extremely fast and powerful personal computers. These personal computers made it possible to maximize individual productivity. However, most current hardware and software technological developments have been aimed at maximizing group productivity. Increasingly, personal computers are networked together to enable communication between users and to facilitate sharing of data and resources. This chapter is aimed at providing a basic understanding of a range of telecommunications concepts including local area and wide area networks. We also discuss some recent communications technologies affecting business such as client/server systems, the Internet, and electronic data interchange.  Almost all computer systems in organizations today are networked, and these networked computer systems invariably house a wealth of accounting information.  It is therefore important for accountants to have a working knowledge of data communications and networking concepts.

Telecommunications concepts

Telecommunications refers to the electronic transmission of information from a point of origin to a point of destination. A telecommunications network is composed of five components: (1) terminals and workstations, (2) transmission links, (3) transmissions methods, (4) nodes and switches, and (5) architecture and standards. Each of these components is now discussed in some detail.

Terminals and workstations

User interaction with a network is almost always via a terminal or a workstation. As shown in the following figure a terminal is a device with a screen and a keyboard but usually does not have any data storage or processing capability independent of its connection to the network. For this reason, such terminals are referred to as dumb terminals.

 

 

Workstations on the other hand do possess data storage and processing capability independent of the network connection. The term "intelligent workstation" is often used and the device is invariably a personal computer, as shown in the figure below. Even a point of sale cash register could serve as an end-user input/display device in a network.

 

 

 

 

Terminals and workstations are the devices the user interacts with. At the sending end, these devices convert text, graphics, and speech into electronic signals. These signals must eventually be reconverted from electronic form back to the original form (text, graphics, speech) if the destination is another user. If the destination is simply a computer or a server, then the signals remain in electronic form. Regardless of whether a terminal or a workstation is used, a network interface card (NIC for short) is required to physically connect the terminal or workstation to the network. In personal computers (the most common type of workstation), a network interface card is an expansion board that works with the network operating system to control the flow of information over the network. 

A recently emerging option, which is almost a hybrid between a dumb terminal and a workstation, is referred to as a "Net PC. An “internet appliance” is designed for network operation and eliminates features such as hard disks, disk drives, and CD-ROM drives. These appliances are “closed units” which cannot be expanded or modified by the end-user.  To facilitate network processing, they come with a network adapter built-in and software pre-loaded and configured for Internet access. Owing to their simplicity, they are less likely to experience software and hardware crashes (which most PC users are very familiar with!). Compaq currently offers such a home Internet appliance under its iPAQ brand name. 


Transmission links

Transmission links, the next component of a telecommunications network, are the links that actually carry the electronic transmissions and can be either physical or "through the air." Twisted pair telephone wire, coaxial cables, and fiber optic cables are examples of physical transmission links. The figure below shows these cable types.

 

 

Twisted pair wire is the copper wire used to connect telephone devices. The wire is twisted to enhance signal transmission relative to straight wire. Twisted pair wire can be either shielded or unshielded. Shielded twisted pair permits higher data transmission speed and is less susceptible to interference, but is also more expensive than unshielded. Unshielded twisted pair is more common in homes and businesses. Coaxial cables (or simply coax) are used for cable television and to connect a video cassette recorder to a television. Coax cables comprise a number of insulated wires inside a thick (usually black) insulated sheathing. Relative to twisted pair, coax cables are capable of far higher transmission speeds and are also less susceptible to electromagnetic interference. Fiber optic cables carry only digital signals and can carry signals at very high speeds. Data transmission on fiber optic cables is in the form of light pulses. Fiber optic cables can transmit data at very high frequencies thereby providing much larger transmission capacity, or bandwidth, than twisted pair wire or coax cables. Since transmissions along a fiber optic line occur using pulses of light, this medium is essentially impervious to interference from electromagnetic sources. However, fiber optic cables are more fragile than twisted pair and coax cables and are usually more expensive. 

In contrast to the above physical transmission links, infrared, microwave and satellite transmissions also act as transmission links, however the transmissions are "through the air." The first of these wireless transmission links is infrared transmission, which involves the movement of extremely short wavelengths in a straight line. Infrared transmission, so called because its range is below the visible spectrum, is the transmission technology used for remote control units on audio and video equipment. As you may have experienced, infrared waves have a very limited range and cannot penetrate solid objects. Whereas infrared transmission is used only for very short distances, microwave and satellite transmission are used to cover long distances. The figure below depicts microwave and satellite transmission.

 

 

Microwaves are very high frequency (above 1 gigahertz) radio waves that travel along a "line of sight" path. Since microwaves travel in a straight line (hence the term "line of sight"), repeater stations must be placed every 50 miles or so to relay the signal from one point to the next. Microwave transmission is also very high bandwidth and carries over half of all telephone and television traffic in the U.S. The final "through the air" transmission link option is satellites. Radio signals are sent from ground stations to satellites which then transmit the signals to other ground stations at distant locations. The satellites are usually in a geosynchronous (or geostationary) orbit which means that they rotate at the same speed as the earth's rotation speed such that they always appear above the same point relative to the earth's surface. One communication satellite in a geosynchronous orbit can cover about 30% of the earth's surface. Needless to say, leasing time on a satellite link can be very expensive due to the high cost of launching and operating a communications satellite.

Transmission methods and modes

Transmission methods consist of either analog or digital transmission using either circuit switching or packet switching. Analog transmission is used for voice transmission by continuously varying an electrical signal to correspond with the variation in the sound wave produced by the speaker's voice. In digital transmission, a signal has only two states -- 0 and 1. Thus, all that is needed is a mechanism for distinguishing between 0s and 1s, and this is accomplished by sending a positive voltage for a 0 and a negative voltage for a 1. In analog transmission, speeding up the signal results in distortion of the voice signal, much like fast forwarding an audio tape results in a "squeaky" voice. Digital transmission in contrast results in no such distortion when the transmission speed is increased. Analog and digital transmission are depicted in the figure below.

 

 

Networks typically use either circuit switching or packet switching to transmit messages. In circuit switching, the method used for placing telephone calls, the signal is routed through various nodes from the point of origin to the point of destination, and the selected path is fixed during the duration of the session. The path is also dedicated -- once a session (phone call) is initiated then there can be no other users (callers) on the channel connecting the sender and receiver. In packet switching information to be sent is divided into small blocks called packets. Each packet is coded with the destination address and is sent separately over the network. Different packets can take different routes through the network to reach the destination node. As each node receives a packet, it checks the destination address on the packet and either accepts the packet or sends it further along if it is intended for another recipient. Upon arrival at the destination, the packets are assembled together in the correct order and presented at the receiver's terminal. Unlike circuit switching, packet switching does not tie up a channel since there is no dedicated path from the sender to the receiver. Packet switching thus results in more efficient use of a network than circuit switching. The method of transferring data over the Internet, the foremost "public" wide area network, is packet switching. When you send an email message over the Internet, the message is formed into packets and sent from node to node until it reaches the final destination. 

For the transmission methods discussed above, the transmission mode can be either synchronous or asynchronous. In synchronous transmission, a timing signal is used to keep the sending and receiving unit synchronized to one another.  Synchronous transmission requires expensive equipment but usually results in faster data transmission.  In asynchronous transmission, the sending and receiving units are not synchronized to one another.  Although asynchronous transmission is cheaper (no expensive timing devices are needed), data transmission is slower.  Most PC to mainframe communication is conducted in asynchronous mode.

Line sharing and switching devices

The fourth component of telecommunications networks are line sharing and switching devices whose function is to receive signals from sending nodes and efficiently direct those signals to the intended destination. These devices can be thought of as airline hubs or highway interchanges where traffic from all directions gets routed appropriately. Line sharing devices include multiplexers and front end processors. Switching devices include bridges, routers, and gateways. 

multiplexer (MUX for short) is a line sharing device that bundles small streams of slow-speed traffic together and places them on a single high capacity high speed line. At the receiving end, another multiplexer unbundles the traffic so that individual messages can be delivered to the appropriate locations. The sharing of a single line between many users results in cost savings since the multiplexer obviates the need for separate lines from the point of origin to the point of destination for each user. All that is needed is a separate line from each user to the multiplexer which then uses a single line to transmit messages originating from multiple sources. In this manner, multiplexers permit the efficient transmission of a series of messages while maintaining a separation between each of the messages. If a company has thousands of employees that need to be connected to the Internet, multiplexers can be used to reduce the number of physical connections directly to the Internet -- many users may simply be connected to the Internet through a multiplexer. 

Two common multiplexing methods are frequency division multiplexing (FDM) and time division multiplexing (TDM).  FDM is used with analog systems (rather than digital systems).  Signals from multiple devices are modulated using different frequencies and then stacked together for transmission over the single line.  TDM is used with digital systems.  In TDM, signals from each device are divided into one-bit time frames. These time frames from all devices are then interleaved together such that the individual signals are merged into a single multiplexed signal. 

front-end processor is a device that connects the multiplexer to the mainframe computer or server that is at the receiving end of the messages. While the multiplexer simply separates the bundled signals so that messages can be distinguished from one another, the front-end processor performs error checking and control functions that would otherwise have to be performed by the computer or server. Packets of data can be checked for errors, with messages that fail error checks being logged and stored in a separate file. The front-end processor can ensure that access to the computer/server is being gained only be authorized users. Finally, messages can be stored in a buffer until the computer/server is free and requests input of the messages. In essence, the front-end processor relieves the computer or server from performing these functions. The following figure depicts multiplexers and a front end processor.

 

 

Switching devices are hardware components that direct traffic over the physical network transmission links. They include bridges, routers, and gateways. A bridge is a device used for interconnecting local area networks (to be discussed a little later in the chapter) so that they can exchange data.  Bridges can connect networks even if they do not use the same wiring or networking protocols.  Imagine a large organization with several departments, with each department having a local area network. Bridges between the departmental networks will enable users in different departments to communicate with one another. A router is an intelligent connecting device that is used to move data from a source to a destination.  Routers can connect networks that use either the same protocol, or dissimilar but compatible protocols. As the name suggests, a router selects the most efficient path to the destination by looking up routing tables of all possible paths between the origin and destination networks. Routers are thus more intelligent than bridges because they can search for the best path from source to destination.  A brouter is a combination of bridge and router -- it performs both tasks.  A gateway is needed to connect networks using dissimilar protocols that are not compatible with one another.  The gateway handles the task of protocol conversion thus enabling communication between the sending and receiving networks. Gateways are thus more complex than routers because of the transformation that needs to be performed to the messages as they move across incompatible architectures.

Architecture and standards

We have already briefly introduced the concept of a protocol or a communication convention. The final component of a telecommunications network is its architecture and standards. This component addresses the transmission protocols used and the specific network architecture around which the various components are organized. A protocol defines the procedures to follow when transmitting and receiving data. It can be thought of as a "common language" for computers to talk to one another. Protocols define the format, timing, sequence, and error checking mechanisms that allow telecommunications to occur.

 

Open Systems Interconnection (OSI) model

While many hardware and software vendors have attempted to promote their own protocols, the International Standards Organization (ISO) has set forth a reference model for networking called the Open Systems Interconnection (OSI) model. The OSI model consists of seven interconnected layers, as shown in the figure on the next page.

 

 

The application layer is the highest layer in the OSI model and is what the user sees (at both the sending and receiving ends of the communication network). It defines the way the user's application program interacts with the network. The application program could be either electronic mail, database management, or a terminal emulation program (for connecting to a mainframe computer system). It is in the application layer that the user's message is converted from human readable form to computer readable form with the message header indicating the sender and intended receiver of the message. The presentation layer defines the way that data is formatted, presented, converted, and coded. In essence, the presentation layer ensures that the message is transmitted in a language that the receiving computer can understand (often ASCII -- American Standard Code for Information Interchange). If necessary, or as directed by the user, the message is also compressed and encrypted at this stage. The session layer coordinates communication between the sender and receiver. In essence, this layer maintains the session for as long as it is needed, performing security, logging, and any administrative functions that are needed. It is in the session layer that the mode of communication is established -- either full duplex where both parties in the communication can send and receive messages simultaneously, or half duplex where the parties must take turns communicating. All of these details are recorded and placed into a "session header" for the session. 

The transport layer, layer 4, defines protocols for message structure and supervises the validity of the transmission by performing error checking. In effect, the transport layer protects the data being transmitted. The protection comes from checksum tests -- mathematically calculated sums based on the contents of the data being sent. The "transport header" records each segments checksum and its position in the message. The third layer, the network layer, defines protocols for data routing to ensure that the data arrives at the correct destination node. It is the network layer that essentially selects a route for the message, using protocols such as TCP/IP (transmission control protocol/internet protocol), or IPX/SPX which is the protocol for Novell networks.  Routers, discussed above, are used at the network layer.  The data are formed into packets and a header is added that contains the sequence and number of packets and the network address of the destination. Layer 2, the data-link layer, validates the integrity of the flow of data between nodes. This validation is performed by synchronizing blocks of data and controlling the flow of data. In this manner, the data-link layer supervises the transmission. It confirms the checksum and then addresses and duplicates the packets. The data-link layer keeps a copy of each packet until it receives confirmation from the next point along the transmission route that the packet has been received. Bridges, discussed above, are used at the data-link layer.  Finally, the physical layer, is the actual transmission hardware or link along which the messages physically pass. It is only along layer 1 that messages physically move from the origin to the destination. If phone lines are being used, then it is the physical layer that actually converts the digital signals into analog signals so that they can be carried on the phone line. Intermediate nodes along the transmission path verify the checksum and might reroute the message in light of congestion in the network. 

At the receiving end, the message passes through the same seven layers, in reverse. The physical layer reconverts the analog signals into digital form (bits). The data-link layer recomputes the checksum, confirms arrival, and logs in the packets. The network layer recounts each packet for security and billing purposes. The transport layer again recalculates the checksum and rebuilds the message segments. The session layer holds the parts of the message until the message is complete and then sends it to the next layer. If the message was compressed, the presentation layer expands it, and if the message had been encrypted it is decrypted at this stage. Finally, the application layer reconverts the bits into readable characters and directs the data to the appropriate application (e.g., email). 

A few comments about the seven layer OSI model are appropriate. First, each layer is independent allowing protocols for each layer to be defined and developed independent of other layers. Second, communication is possible only between adjacent layers. Layer 3 can communicate only with layer 2 and layer 4. Finally, the model makes a distinction between data communication tasks and data manipulation tasks. Layers 1 through 4 (physical through transport) perform data communications tasks which interact primarily with hardware devices. Layers 1 and 2 interact only with hardware devices, whereas Layers 3 and 4 interact with both hardware and software devices. For data transmissions over the Internet, it is layers 1 through 4 that handle the necessary tasks for physically transmitting packets of data over the Internet. As we will see a little later, there can literally be hundreds of "intermediate nodes" that come into play when data is transmitted over the Internet. Layers 5, 6, and 7 which perform data manipulation tasks interact primarily with software -- the operating system and the specific application program being used. Thus, when you use a Web browser such as Netscape, layers 5 through 7 are responsible for the necessary translations between the network transmission (e.g., Novell Netware or Windows NT), the operating system (e.g., Windows 9x), and the browser (e.g., Netscape 4.x).

Network types

Let us now turn to a discussion of the broad categories of networks -- local area networks and wide area networks. It should be noted that networks -- both local area and wide area -- can be either private or public and can carry either voice or data or both. Given our focus on computer-based accounting systems, we focus only on data oriented local area and wide area network, although the concepts also apply to voice networks.

Local area networks

When two or more computers in close physical proximity are linked together using physical connectors the result is a local area network (LAN). The main components of a LAN are (1) computers and other devices such as printers, (2) a relatively powerful computer which functions as the network "server," and (3) a communications channel connecting the computers and devices. Each computer and device to be connected on the LAN must have a network interface card (also referred to as an adapter). The three main types of cable physically connecting the computers together are coaxial, twisted pair, and fiber optics. As the name suggests, a server is any node on the network that provides a service such as shared access to a printer, a disk drive, or other devices such as modems.

Network topologies

The three network topologies are bus, ring, and starBus networks, in which each device is connected to a common "backbone," are by far the most popular. The actual connection is by means of a "tap" which allows the device to listen to the transmissions along the bus. Bus networks use the ethernet protocol which is capable of transferring data at the rate of 10 megabits per second (mbps) to 100 mbps. Devices connected via ethernet links use a technique called carrier sense multiple access/collision detection (CSMA/CD) to communicate. This convention allows multiple access to the bus network such that any node can essentially broadcast a message over the bus to another node. Every node constantly monitors the bus to detect messages intended for itself (hence the terms "carrier sense" and "multiple access"). If two nodes attempt to transmit messages simultaneously to each other or to the same node a collision occurs. The nodes involved in a collision wait for a random period of time until one of them tries to send the message again. A bus network is depicted in the figure below.

 

 

In a ring LAN, each device is connected to a common circular loop (cable). The physical connection to the ring is by means of a "repeater," as depicted in the figure below. The ring network is somewhat like a circular rapid transit system in which the train arrives at each station along the ring, dropping off and picking up passengers at each station. In similar fashion, each node in a ring network is polled to determine whether it has any messages to send. Messages typically move in one direction around the ring. When a message arrives at a node it is checked to determine whether it is intended for that node. If it is not intended for that node the message is simply regenerated and passed on to the next node. When the message reaches the node it is intended for, it is accepted and not passed any further. Since messages are regenerated at each node, ring networks can cover greater distances when compared to bus networks. This process of passing messages is called "token passing" and such networks are commonly referred to as "token ring networks." The node that originates a message creates a "token" which is what gets passed from node to node to the ultimate recipient of the message. A node that also has a message for the recipient of a token can simply add to the message and modify the token to indicate that an addition has been made. When the token reaches its final destination the receiving node returns the token to the sender thus signaling that the message was received in tact.

 

 

The final LAN configuration is the star in which one computer acts as the "host" computer to which all other computers and devices are connected. The central host computer is also referred to as the "hub." All messages are routed through the host. Messages are channeled by the host to the intended node in a process referred to as network switching. Star networks use a significant amount of cable since each node must have its own direct connection to the central host computer. The most critical link in a star network is the central host computer; if it breaks down then the entire network becomes inoperative. By contrast, bus networks are the most fault tolerant because a defective device does not cripple the entire network. Depending on the location of the defect, a ring network may become partially or completely inoperative.

 

 

Two very popular local area network operating systems (NOS) are Novell NetWare and Windows 2000 server. Both of these NOS are high-performance scalable operating environments that provide the powerful, reliable processing for services such as file and print sharing. An advantage of Windows 2000 Server is that it comes with Internet, intranet, and communications services built in. In this manner, the Windows 2000 Server NOS makes it easy to integrate the Web right into an organization's local area network. You are encouraged to visit these Web sites to further explore the features of these network operating systems. 

The three LAN configurations are summarized in the following table. As noted earlier, bus networks are by far the most popular and star networks are the least popular. However, hybrid networks, combining two or more of the above topologies are often the norm. For example, a large network could combine star and bus topologies, with several bus networks all connected to a central mainframe in a star configuration.

 

Local Area Networks

LAN

Description

Message passing

Ring

Each device connected to two other devices with no central host. 

Token passing, with each device passing a token to the next device. 

Star

Each device connected to a central host computer. 

Networking switching; all messages must pass through the host computer. 

Bus

Devices connected to a bus channel, typically Ethernet. 

Carrier sense multiple access/collision detection; each device can "broadcast" a message to another device on the bus. 

 

 

Wide area networks

As discussed above, local area networks involve computers that are in close geographical proximity. In contrast, networks that connect users over vast geographical distances are referred to as wide area networks (WAN). The communication links connecting nodes in a wide area network are either phone lines, microwave links, or satellites. For a computer to communicate with another computer over phone lines, a device called a modem is required. Modems can be internal or external. A modem (an abbreviation of modulator-demodulator) converts the computers digital signals into analog signals for transmission over phone lines (a process called modulation). The modem at the receiving end reconverts the analog signals back into digital signals that the receiving computer can understand. The speed of modems is expressed in terms of the number of bits per second (bps) that can be transferred. A speed of 28,800 bps is considered to be a bare minimum for modems today, while 56 kbps modems are fast becoming standard issue on most home computers these days. While these recent developments in modem technology are encouraging, it should be noted that the quality of phone lines in most areas will only rarely allow transmissions in excess of 28.8 kbps. 

Modems are capable of both data and fax transmission (hence the term "fax modem"). However, the speeds of transmission for data and fax could differ. A modem rated at 28,800 bps for data may not necessarily transmit faxes at the same rate. In evaluating modems, a consideration is the CCITT standard to which a modem conforms. The CCITT (Consultative Committee on International Telephone and Telegraph) is responsible for setting international standards for networking and data communications. The CCITT V.34 standard requires data transmission speed of 28,800 bps. Thus, the manufacturer of a V.34 modem is asserting that the modem transmits data at 28,800 bps. A recently issued standard, V.90, applies to 56 kbps modems (there were previously two competing standards -- referred to as K56Flex and x2 technology).  There are some other features of modems that warrant discussion. Current modem technology allows the modem to automatically differentiate between data, voice, and fax. Voice mailboxes and speakerphone options are also commonly available in present day modems. 

In addition to conventional phone lines which can carry only analog signals, recent technology called integrated services digital network (ISDN) allows both analog and digital signals to be carried on the same phone lines. The bandwidth of an ISDN line is 128 kilobits per second (kbps) - considerably faster than 28.8 kbps modems. A significant advantage of ISDN is that data and voice transmissions can occur simultaneously by splitting the 128 kbps line into two 64 kbps lines, one for carrying the data and the other for handling the voice call. ISDN is gaining in popularity, but significant barriers to their widespread use are high cost and limited availability. ISDN modems cost approximately $400, and the monthly charge for an ISDN line ranges from a fixed expense of $30 to $60 per month and a variable expense at some amount per minute of use. Since the switching equipment must be updated to handle ISDN, phone companies have been slow to upgrade all their lines to ISDN. 

Recently, a number of alternatives to ISDN have become available. Local phone companies in a number of metropolitan areas are DSL (digital subscriber lines) service, which is a technology for bringing high-bandwidth information to homes and small businesses over ordinary copper telephone lines.  Typical downlink (downloading) speeds for DSL vary between 1.54 mbps and 512 kbps, in contrast with 64 to 128 kbps for ISDN. The uplink (uploading) speed of DSL, however, is typically 128 kbps which is similar to ISDN. In addition to local phone companies, cable television providers are also getting into the Internet connectivity game. Cable modems connect to the user's home computer using a special adapter through the regular cable TV connection. Cable modems have proven to be extremely fast -- offering download speeds up to 10 mbps and upload speeds up to 768 kbps.  Typical downloading speeds vary from 128 kbps to 1 mbps, while uploading speeds vary between 64 kbps and 128 kbps depending on the level of service.  A key advantage of cable modems is that they provide a constant connection to the Internet (unlike conventional dial-up modems) and offer significantly faster download speeds relative to dial-up modems. 

Microwave and satellite transmissions are becoming increasingly popular in wide area networks. The same direct-broadcast satellite technology that brings television to homes is also capable of providing a 400 kbps link to the Internet. However, there is no ability to upload data to the Internet using the satellite link; uploading would require a conventional modem and would thus be restricted to 28.8 kbps of uploading speed. The direct satellite Internet link does provide a viable option for users primarily interested in browsing the Internet which involves only downloading of data. The direct satellite Internet linking system may not however be available in all areas. 

The discussion above has focused on the uploading and downloading speeds for modems, ISDN, DSL, cable modems, and satellites. As a point of comparison, it is worth noting that a dedicated connection to the Internet can be either a T1 or a T3 link. A T1 link has an uploading and downloading speed of 1.54 mbps, whereas a T3 link has a capacity of about 45 mbps. Most internet service providers (ISP) have one or more T1 links, more than sufficient to handle Internet traffic for hundreds of concurrent users. 

Depending on the volume of transmission, and the resources available to an organization, phone connections may be made either using switched lines, WATS (wide-area telephone service), or leased lines. Switched lines are the least expensive option when the nodes need to communicate only occasionally and when the volume of data to be transferred is relatively small. A charge is assessed for each phone call placed to link computers on the WAN. Leased lines are the most expensive, but are often economical if the volume of transmission is very high. A fixed monthly fee, which is quite high, is charged for the leased line regardless of the level of usage. WATS is an option that is in between the two extremes. The WATS option involves a fixed monthly fee plus a lower cost for each individual call relative to purely switched lines. Thus, WATS may prove to be a cost-effective option when the transmission volume is high but not high enough to warrant a leased line. The various communication technology options for WANs are summarized below.

 

Communications Technologies

Technology

Description

Modem

Modulator demodulator. Uses conventional phone lines. Sustained speeds of about 33.6 kbps. Modems theoretically capable of 56 kbps. 

ISDN

Integrated Services Digital Network. Requires special adapters and digital modems. Speeds up to 128 kbps. Line charges and modem costs higher than conventional modems.  Fading in popularity with the emergence of DSL and cable modems.

DSL

Digital Subscriber Lines. Emerging as an alternative to ISDN, with downloading speeds up to 6 mbps.  Requires proximity to phone company switching station.

Satellites

Downloading only. Speeds up to 400 mbps. 

Cable modems

Constant Internet connection; speeds up to 10 mbps for uploading and downloading.  Requires cable TV connection.

 

 

Types of wide area networks

There are two broad types of wide area networks: private and public. Companies can set up their own wide area networks by leasing transmission lines from one of the providers of these lines (e.g., MCI or Sprint). It should be noted however, that it can be extremely expensive to set up a proprietary wide area network. The second option is to simply use a public wide area network. The Internet is in essence the world's largest public wide area network. We will first discuss the options for private wide area networks, after which we will discuss the Internet. 

When computer networking first became feasible, larger organizations established centralized data processing networks, where several remote data entry terminals (dumb terminals) were connected to a single centralized host computer. All data processing was done at the central location; the remote terminals were used simply to input data and retrieve information. The remote terminals could be connected to the central computer via one of three options: (1) point-to-point lines, (2) multidrop lines, and (3) multiplexed lines. These three options are depicted in the figure on the next page.

 

 

Point-to-point lines represent the most expensive option since each remote terminal must have its own direct connection to the central computer. However, each terminal has its own connection and will therefore experience faster response times. Moreover, failure in any one line affects only one remote terminal. Multidrop lines involve several remote terminals being connected to the central computer using a single line. This option is less expensive than point-to-point lines. However, response times are slower because of the sharing of the common line. Each remote terminal must contend with the other terminals for use of the shared line -- there can be only one terminal connected at any time. Further, if the shared line fails then all remote terminals suffer. The third option, multiplexed lines, represent an improvement over multidrop lines because of the use of a multiplexer. As already discussed, a multiplexer combines signals received from several terminals and transmits them over a single line. Using multiplexed lines, the remote terminals do not have "wait their turn;" they can simultaneously transmit data since the multiplexer is capable of handling multiple messages received simultaneously. 

The second major type of WAN is a distributed data processing (DDP) network. As the term suggests, in a DDP network, data processing is distributed throughout the network. In contrast, the centralized networks discussed above concentrate all data processing at one central location. With the decreasing costs and increasing power of workstations and personal computers, DDP networks are becoming increasingly feasible even for relatively small organizations. DDP networks have several advantages over centralized networks. First, they are more responsive to local user needs. Since some of the data processing can be done locally, a remote node does not always have to rely on the central host computer for its local data processing needs. Second, a DDP results in lower data communication costs. Again, since part of the data processing tasks are handled locally rather than over the network, the extent of data communication (and hence the costs of data communication) is reduced. Third, a DDP network is more reliable than a purely centralized network. The reliability stems from the presence of multiple processors and the connectivity among the nodes in the network. If one node loses its processing capability, then the processing burden can be shifted to another node in the network. In contrast, a centralized network is rendered useless if the central host computer fails. Finally, a DDP network is more flexible than a centralized network since additional nodes and processors can always be added in the event that the processing capabilities are found to be insufficient. 

DDP networks do have some disadvantages relative to centralized networks. First, security concerns are heightened in a DDP environment, given the presence of multiple processing locations. In contrast, in a centralized network, all control and security can be centralized -- there is only one critical link in the network (the central host computer). Second, as DDP networks grow, the complexity of the network represents a significant problem from a network management perspective. There are simply more network components that could potentially fail in a DDP network relative to a centralized network. Despite these drawbacks, DDP networks are extremely popular. DDP networks can be configured either as bus, ring, or star networks. These topologies have already been discussed and will not be repeated. One special type of DDP network is the client/server system which is discussed next.

Client/server systems

The term "client/server" has a variety of connotations and its definitions abound. Client/server is an architecture that distributes processing in a networked environment between clients and one or more servers. Any node in a network requesting data or a service is considered to be a client, and any node or device providing a service is considered to be a server. Clients are simply the end-user computers requesting information while running an application. Servers have access to data repositories and make the information available. In a client/server network, the client computer can perform certain local processing tasks such as handling the user interface, while a server at a remote location performs the more data intensive tasks such as searching and retrieving data from large corporate databases. Thus, processing tasks are shared across the client/server network -- not all tasks are handled by the server alone or the client alone. The following figure contrasts the older mainframe to dumb-terminal communication with the more modern server to workstation, i.e., client/server communication. 

 

 

As shown in the column to the right above, the most typical implementation of a client/server system is the "three tier" architecture, consisting of the presentation layer, the application layer, and the database layer.  In such an architecture, the "presentation layer" is the software on the client machine -- typically a Windows 95/98 or Windows NT client PC.  The "application layer" is a server on which the application software -- such as SAP R/3 -- resides.  Finally, the "database layer" is typically a separate server on which the actual database -- typically Oracle, DB2, or SQL Server -- resides.  Such a three-tier client/server systems can be distinguished from the older mainframe oriented approach, shown to the left in the above figure.  Under the mainframe approach, sometimes referred to as the "file server" approach, a dumb terminal made a request and all of the processing was done at one large computer (i.e., the mainframe).   Thus all of the processing must be performed at the mainframe, including the task of formatting the output for presentation.  The processed and formatted data is simply delivered to the client where it is displayed on the dumb terminal.  By contrast, in a three-tier client/server system, a workstation (client) makes a request, an application server processes the request, the database server actually delivers the data, and the client workstation actually formats and presents the processed data to the user.

The Internet

The options discussed above involve organizations setting up their own private wide area networks. A phenomenally popular public wide area network is the Internet -- a global network of networks which communicate using the transmission control protocol/Internet protocol (TCP/IP). The origins of the Internet date back to 1969 and the ARPANet. The Defense Department's Advanced Research Projects Agency (ARPA) sought to develop a fault tolerant wide area network that could withstand a nuclear attack. Thus, the TCP/IP protocol is designed to automatically search for alternative paths from the origin to the destination in the event that intermediate nodes in the network are disabled (as discussed earlier, packet switching involves forwarding packets from node to node from the sender to the receiver). Initially used mainly at educational and government institutions, the Internet is now touching virtually all aspects of our daily lives, from shopping to entertainment to banking and investing. It is estimated that there are over 90 million people worldwide with Internet access. 

To access this book, you are using the graphical component of the Internet -- the World Wide Web (WWW or Web for short). The key technology behind the Web is hypertext which allows links to be established from one "page" to another page which can literally be on a computer on another continent. Using a Web browser like Netscape or Microsoft's Internet Explorer, even novice users can access a variety of information simply through clicks of a mouse button. In fact, the popularity of the Internet in the last two years can be attributed to the ease use of the Web and the tremendous amount of (mostly free) information available on the Web. Some estimates put the number of Web sites world wide at over 1.4 million. 

Apart from the Web, there are a number of other popular Internet applications. The single most popular application is electronic mail, or email for short. In addition to simple text messages, most electronic mail software packages also permit binary files, such as word processing and spreadsheet files, to be attached to the mail message. File transfer programs, or FTP for short, allow users to transfer both binary and ASCII files over the Internet. Note that binary files can only be read and/or executed by certain specific programs (e.g., a Microsoft Word or Microsoft Excel file) whereas ASCII files (ASCII = American Standard Code for Information Interchange) are text files which can be read by any text editor such as the "Notepad" program in Windows 98. Telnet is an application that allows users to log in to remote computers over the Internet. This application is used primarily to log in to remote Unix and Digital VAX computers with command line interfaces (as opposed to graphical interfaces). Other programs like Internet Relay Chat (IRC), which permit real-time discussions across the Internet, USENET newsgroups, which are discussion groups focusing on specific subjects, and multi-user dungeons (MUD) which facilitate role-playing games over the Internet have also contributed to the immense popularity of the Internet. 

While the Internet is inherently a public network with all information freely available, a recent phenomenon is the development of corporate "Intranets." An Intranet comprises information made available using similar technology as the Internet but only to authorized users. Through user accounts and passwords, information published on such Intranets is shielded from the general public. Netscape Communications has catalogued a number of examples of companies establishing Intranets. For example, AT&T has an Intranet which its 300,000 employees can use to find each other. Employees can use a Web page interface to a database of employee phone numbers, addresses, titles, and organizational information. Booz Allen & Hamilton, one of the foremost consulting companies in the world, uses Intranet technology to bridge islands of information among consultants. Their intranets include a company wide knowledge repository, an expert skills directory, employee collaboration, and employee directories. Another example is Home Box Office (HBO). One use of Intranet technology at HBO is to roll out new marketing campaigns. Salespersons log on to the HBO intranet to find information about new marketing campaigns. Providing such information on the intranet eliminates the substantial costs associated with printing, videocassette duplication, and distribution. An emerging variant of the "intranet" is the extranet. Whereas the intranet is typically used within a company, that is by its employees, an "extranet" allows controlled access to the corporation's Web site by customers. CyberText Publishing's Web site is an example of an extranet since access to the Web site is restricted to customers -- students and instructors -- who have valid accounts. Another example of an extranet is Columbia/HCA Healthcare. An extranet is used to advertise surplus medical equipment to Columbia-affiliated facilities who are provided access to Columbia's restricted Web site.

Electronic data interchange

Electronic data interchange (EDI) is the electronic exchange of business transaction information between organizations. Consider a typical business transaction between Company A, the buyer, and Company B, the seller, as depicted in the figure on the next page. The purchases information system of Company A issues a purchase order which is sent via conventional mail to Company B where it is manually entered into the sales information system. The sales accounting system of Company B then generates a sales invoice which is sent via mail back to Company A. Company A must then manually enter the seller's invoice into its accounts payable system. Thereafter, Company A's disbursement system will print out a check which is then mailed to Company B, where the check is manually keyed into the cash receipts system. The check must be manually deposited into Company B's bank account.

 

 

What the above scenario makes clear is that process is slow and error prone. Documents are sent between organizations via the postal service. With the exception of facsimile transmission, every other mode of document delivery results in at least some delay in communication between the selling and purchasing organizations. The most problematic aspect of the typical conventional business scenario described above, however, is that critical business information about purchases and cash collections is manually entered into the accounting system. Such manual data entry often introduces errors and irregularities into the accounting systems of both organizations. 

In an EDI environment, the selling and purchasing organizations establish electronic links between their accounting systems, either directly or through a value added network (VAN). An EDI equivalent of the above manual selling and purchasing scenario between Company A and Company B is depicted in the figure on the following page. Company A sends the purchase order information electronically, Company B sends the sales invoice information electronically, and payment is also made electronically via electronic funds transfer (EFT). Of course, the merchandise must still be physically delivered. EDI thus results in electronic billing on the part of the seller and electronic payment on the part of the purchaser.

 

 

As can be seen, the advantages of EDI are that business transaction information is transmitted almost instantaneously. Since transaction information is exchanged electronically between the seller's and purchaser's accounting systems, errors in data entry are essentially eliminated. Apart from speed and elimination of potential errors, there are other benefits of EDI. The "just-in-time" (JIT) manufacturing technique would be extremely difficult, if not impossible, to implement without EDI. Since EDI facilitates JIT, it permits the organization to reduce inventory thereby reducing working capital requirements. Since much of the clerical work involved in mailing transaction documents and manually entering them into accounting systems is eliminated, EDI also permits a reduction in the organization's work force resulting in further cost savings. 

Although the concept of EDI would appear very appealing, there are a number of barriers which must be overcome. The initial cost of setting up EDI can be extremely high. The payback from EDI, in terms of the benefits described above, is usually realized over several years. Smaller firms may not have the resources necessary to convert to EDI and may thus lose their business conducted with larger firms making the switch to EDI. In fact, organizations converting to EDI typically reduce the number of vendors they deal with simply because of the inability of smaller vendors to offer EDI.

EDI standards

A critical problem in EDI is in determining the format of electronic transmissions between organizations. It is infeasible for organizations to negotiate with each customer and each vendor to determine the exact format of EDI transactions between them. The solution is the American National Standards Institute (ANSI) which has a subcommittee dedicated to the task of establishing standards for EDI transactions. This subcommittee, referred to as the ANSI.X12 subcommittee, has established standard EDI transaction formats for a variety of business transactions including purchase orders and sales invoices. Organizations need only refer to and conform to the .X12 standard without having to negotiate a transaction format with each EDI customer and vendor.

VANs vs. Internet-based EDI

As depicted in the figure above, there are two means by which companies can adopt EDI technology. The first and more expensive approach would be to establish separate EDI links with each vendor and each customer that an organization does business with. However, such separate links can prove to be prohibitively expensive even for very large organizations since each link is essentially a form of a private network. The second approach is to use value added networks (VAN) like Sprint and MCI. A VAN adds value in two main ways. First, it allows a company a single connection and communications protocol, rather than the many that would be needed for direct connects. Second, it provides electronic mailboxing which allows a trading partner to send and receive data without regard to its many partners' schedules. Periodically, companies access their electronic mailbox and retrieve their transactions stored there by all their trading partners. Thus, the company has only to establish an EDI link to the VAN and not to each one of the customers. 

Although VANs represent a lower cost option relative to establishing separate communication links to each vendor and each customer, the cost associated with using a VAN are still very high. For smaller businesses that cannot afford the cost of VANs, a recently available option is Internet-based EDI. Rather than using a VAN, companies can simply send and receive electronic transactions via email. Like a VAN, EDI using the Internet allows a single network connection which is the connection between a company and its Internet service provider (ISP). Unlike the VAN model, e-mail software handles mailboxing at (or near) the network endpoints. Electronic transactions accumulate not in a mailbox at the VAN's location but in the company's own email mailbox. 

EDI over the Internet is appealing in several ways. VANs charge $.15-$.25 per kilocharacter to both senders and receivers. The cost per kilocharacter is significantly lower with Internet-based EDI. Also, the VAN architecture introduces an inherent delay as data sits in mailboxes at the VAN's location. Transactions do not get automatically forwarded to the intended recipient. The model is not "store-and-forward," it is "store-and-wait." The major concern with Internet-based EDI is one of security. By its very nature, the Internet is open and insecure. The solution is to encrypt EDI transmissions. Although encryption does not prevent the transmissions from being intercepted, it is virtually impossible to decrypt an appropriately encrypted EDI transaction without the correct decryption code. A leading vendor of EDI software products is Premenos, the developer of a popular Windows-based EDI product called Templar.

E-business

You have undoubtedly heard the terms "electronic commerce" and “e-business” on several occasions. In its initial incarnation, e-business was referred to as electronic commerce (EC), which comprised transactions between parties that are conducted electronically. A simplistic definition of EC is that it encompasses any business transaction in which the parties interact electronically rather than by physical exchanges or direct physical contact. In other words, "doing business paperlessly" is what EC is all about. 

E-business, however, has come to mean much more than electronic transactions. Unlike EC which deals only with transactions between parties, e-business encompasses all exchanges between parties. Just what is an “e-business”?  In an “e-business,” customers, internal operations, and suppliers are all connected online using the Internet. Moreover, the e-business adopts a number of internet-based business practices that fundamentally change the way business is done. An e-business can thus be thought of as a new Internet-based environment for communications, interactions, and transactions. This environment spans entire value chains, enhances existing relationships, and creates new economic value. It impacts new “dot coms” as well as established “brick and mortar” companies. In fact, one e-business model involving traditional companies engaging in Internet-based commerce has been labeled “click and mortar.”

In an e-business, all business processes are “e-enabled” -- from marketing, sales, and procurement -- to manufacturing, supply chain, and service -- to financial operations, project management, and human resources -- to business intelligence systems. In essence, the e-business leverages the power of the Internet to improve efficiency, reduce costs, expand its markets, and retain customers. Success at e-business requires companies to fundamentally rethink their entire business strategy, from the relationships with vendors and customers to the way economic results are measured. Thus, e-business moves well beyond simply having a web site that disseminates timely information. It is also much more than simply taking sales orders and allowing customers to interact with product databases in real-time over the Web.

 

There is no standard or well-accepted categorization of e-business variations. The following categories of e-business are presented in terms of the parties interacting or connecting electronically. 

Web presence: Business organizations can very easily achieve this initial level of e-business, and almost all businesses in developed countries already have. At this basic level, the Web is used to disseminate information to a range of information consumers. The information presented is static and must be periodically updated to remain current. The web site is often used as a conduit to enable users to obtain contact information or even provide feedback to the organization via Web forms.

Business to consumer (B2C): At this level, the organization uses the power of the Internet to transact business and provide a range of services to external users via the Web. B2C thus refers to the retailing side of “doing business on the Web,” and is essentially what is commonly referred to as electronic retailing. There are literally thousands of instances of companies selling goods and services directly from their Web sites (CyberText being one example). Companies like Amazon.com (books, music), Barnes & Nobles (books), Dell (computers), and CDNow (music) are but a few examples of B2C. While electronic retailing is now well established, electronic banking and investing over the Internet have also exploded over the past year. Companies like Charles Schwab and E-Trade allow individual investors to buy and sell securities over the Internet. The main advantage of Internet-based trading in securities is the low transaction cost and the accessibility -- the number of individuals with access to the Internet is now well over a 100 million.  Security First National Bank, Bank of America, and Wells Fargo Bank are some of the banks that currently offer services over the Internet. In addition to electronic paying of bills, these banks also allow individuals to view balances, transfer funds between accounts, and obtain activity statements directly from a Web browser. 

Beyond electronic retailing, investing, banking, B2C also refers to customer self-service and similar applications that provide users with browser based access to internal (back-end) systems. Such functionality allows users to query organizational databases to obtain up-to-the-minute information. For example, most overnight delivery companies allow customers to track packages on the Web. As another example, some manufacturers allow customers to track the status of their order as it moves through various stages of the production process. At this level, e-business involves connecting external users to internal systems and providing valuable services that are easily accessed over the Internet.

Business to business (B2B): Moving beyond B2C, business to business electronic exchanges can encompass a wide range of business processes. Fortune magazine estimates that while worldwide B2C revenues for 2000 will be $250 billion, B2B revenues for 2000 will be around $650 billion. In another forecast, the Gartner Group estimates B2B revenue worldwide to be $7.29 trillion dollars by 2004!

The most common incarnation of B2B involves the forging of electronic connections between a business and its suppliers. EDI, which was discussed earlier, is a prime example of B2B electronic exchanges. Recall that EDI involved using either proprietary connections between companies or the Internet. Beyond EDI, however, there is a wide range of business-to-business transactions that are now possible. Consider the case of procurement. Companies like Ariba and Commerce One produce programs that enable companies to use websites to purchase maintenance, repair, and operating equipment over the Web. This process is now called e-procurement and is the initial B2B step undertaken by most companies. Moving beyond e-procurement, online B2B ventures can involve buying supplies and parts through online auctions, eventually moving to Web-based exchanges as new businesses.
 

Business to business exchanges can also involve electronic trading of digital goods, i.e., information in various forms (text, audio, music, video, graphics).  As you are experiencing, using this online book, such electronic trading of digital goods opens up a whole new realm of possibilities in the 21st century.  In essence, B2B is the exchange of products, services, or information between businesses rather than between businesses and consumers.

The ultimate goal is for such an “e-business” concept to be deployed throughout an entire industry's supply chain, linking manufacturers, assemblers, distributors, marketers and customers. In such a scenario, one press of a button (a customer ordering a product at a company’s web site) automatically triggers processes throughout the chain (initiating a series of business processes within the company and also at the suppliers of materials and services required to make the product). The transfer of information across the value chain would be facilitated by emerging languages such as XML—extensible markup language. Unlike, HTML, which defines a standard set of tags for displaying text on web pages such as <B> for bold, XML is a World Wide Web Consortium (W3C) standard that lets users create their own tags for exchanging information over the Web such as <SALEAMT> for the sales dollar amount. This link on the W3C web site provides a brief 10-point description of XML.

Business to government (B2G): This category of e-business involves business to government electronic exchanges.  Since the last few years, businesses and individuals have been able to file their tax returns electronically, using the IRS e-file initiative.  Moving beyond electronic filing of taxes, business to government EC could involve a range of transactions such as electronically filing 10K and 10Q statements with the SEC, transmitting tax withholding payments electronically, and electronic filing of reports to regulatory agencies such as the EPA and OSHA. 

Other categories: An emerging category of e-business is labeled “B2E,” for business-to-employee exchanges. More specifically, the term "B2E" is frequently used to refer to the company’s employee portal, which is a customizable starting Web page for all employees within the organization. Such a portal is similar in concept to an intranet, but it is different in that it is a customizable entry point rather than a generic “one size fits all” approach of an intranet. The B2E portal should be designed in such a way that the employee can access everything typically found on an intranet (company directory, benefits information, etc.) but also any personal information and links that the employee might want (such as the performance of the employee’s retirement portfolio, weather and traffic information, etc.). The intention of such B2E portals is not just to improve efficiency but also to boost employee morale and develop a sense of community within the organization. The three distinguishing characteristics of a B2E portal are (1) a single point of entry for everyone in the organization, (2) a mixture of organization-specific and employee-defined components, and (3) the ability to be highly customized to suit the needs of an employee.

While not strictly an “e-business” activity, individual to government exchanges are also possible using the Internet as a medium. Beyond electronic filing of tax returns, welfare and social security payments could also be made electronically.  Perhaps individuals will eventually be permitted to vote electronically sometime in the future! See www.internetvoting.com for a demonstration of how Internet voting might work.

For more information about alternative e-business models and strategies, visit the Scient.com web site. You can also find a wealth of information about e-business at the PriceWaterhouseCoopers e-business web site and the Center for e-Business at MIT 

Issues and concerns in e-business

There are a number of issues and concerns that need resolution before e-business can achieve its full potential.  They are (1) the mode of the payment, (2) security of the payment transaction, (3) privacy of individuals, (4) authentication and non-repudiation, and (5) jurisdictional and legal issues. 

Payment mode:  Regarding the mode of the payment, the most common method for businesses is electronic funds transfer (EFT) through traditional banking channels.  For individuals, the most common method is credit cards.  A number of electronic payment systems have evolved over the last two years as alternatives to using credit cards. The foremost among them are CyberCash and DigiCash. The CyberCash  system require the individual to set up an "electronic wallet" that can then be used to pay merchants.  The electronic wallet is established after the individual provides these vendors with a credit card number.  The DigiCash system involves the use of "ecash" (electronic cash).  As explained on the DigiCash web site, this is a software-based payments system that sends "...electronic payments from any personal computer to any other workstation, using any computer network including the Internet."  Ecash is designed to convert money into a digital form. Ecash coins, each of which has a specified value, are stored on the user's hard disk and can be transferred in email or as data files exchanged online between payer and payee. You are encouraged to visit the Web sites of these vendors of payment systems for further information about these payment options. 

Security of payments: Many individual consumers are wary of using their credit cards over the Internet. Credit card numbers, and any other sensitive information for that matter, should not be transmitted "in the clear" over the Internet.  Recall from the discussion earlier in the chapter that the mode of transmission over the internet is packet switching.  Packets containing credit card information can easily be intercepted by a hacker.  Fortunately, mechanisms are in place to encrypt information before it is transmitted over the internet.  Thus, even if the packets are intercepted, they will be useless to the hacker since he/she will not have the means to decode the encrypted credit card information (it is estimated that a powerful super computer would be needed to break a modern encryption scheme).  The foremost among encryption mechanisms is Secure Sockets Layer (SSL) technology. An emerging development is Secure Electronic Transaction (SET) -- a protocol being jointly developed by Visa, Mastercard, Microsoft, Netscape, and IBM. SET does not use conventional credit card numbers. The SET specification incorporates the use of public key cryptography from RSA Data Security to protect the privacy of personal and financial information being transmitted over the Internet. Software residing on the cardholder's personal computer and in the merchant's network computer will handle the necessary encryption and decryption of information. The main advantage of SET is that the seller does not actually see the buyer's credit card number. In addition, the seller is assured that the buyer is the legal owner of the credit card and that it is not a fraudulently obtained card. 

Privacy:  As an individual, if you walk into a store, pick up some merchandise, and pay cash for that merchandise you have left no trace of your transaction (other than the fact that the merchandise was sold).  By contrast, if you are shopping online, you leave a trace of your activity even if you do not purchase anything.  As you click from one link to another on a merchant's web site (say Amazon.com) you are providing information about your tastes and preferences.  The trace of your activity on a merchant's web site is called your "click stream" and represents valuable marketing information for the merchant.  If the merchant also has information about who you are, then a very real concern is whether the merchant will sell that information to a third party without your consent.  You may have read CyberText's privacy policy when you set up your account.  The policy clearly states what the information will be used for and also that the information will not be sold to a third party.  Whenever you are asked to provide any information about yourself on the Internet, you should verify the merchant's privacy policy stating what the merchant could do with the information you provide. 

Trust: authentication and non-repudiation: An advantage of the Internet is its global reach and the ability of entities to provide information anonymously and individuals to access information anonymously.  However, this anonymity can be a significant impediment to e-business.  When individuals are attempting to transact with an Internet merchant, they have no assurance that the merchant actually exists and can be trusted to deliver the goods or services the individual wants to purchase.  As you are probably aware, the cost of setting up a Web site is trivial (under $100).  In response to the need to provide some level of authentication of merchants that operate on the Web, various organizations have begun offering "certification" services for web sites.  One such service is offered by the Better Business Bureau Online (referred to as BBBonline).  According to BBBOnline, participants in the program "...demonstrate their commitment to honest advertising and customer satisfaction by agreeing to strict BBB standards." Merchants that meet the BBBOnline standards, and subject themselves to verification by their local Better Business Bureau, are permitted to display the official "BBBOnline logo" on their web site.  The AICPA is also entering this arena of online assurance.  Upon completion of the necessary training, CPAs can offer a similar service called "WebTrust."  The program involves verification of an online vendor's policies and practices.  Online merchants who meet certain standards and pass certain tests, as verified by a CPA authorized to render the WebTrust service, can display the "WebTrust" seal on their web sites.  In both instances -- BBBOnline and CPA WebTrust -- consumers can click on the logo displayed on a merchant's web site to verify that the merchant is legitimately displaying the logo. You are encouraged to visit the BBBOnline and CPA WebTrust web sites for further information about these programs. 

From the merchant's perspective, there is no assurance that the individual ordering goods or services over the Internet is a legitimate consumer who will not deny the credit card charge.  However, this authentication is virtually impossible to obtain with certainty.  In Europe, there has been a movement towards "smart cards" that individuals can "swipe" onto a smart card reader attached to a computer, thereby verifying their authenticity.  There are also a number of biometric authentication systems such as retinal and iris scanners, speech authentication, fingerprint authentication systems.  However, these systems are much too expensive to implement at the individual consumer level.  Thus, authentication of all parties involved in an e-business exchange is a critical issue that must be resolved to establish trust on the part of all parties involved in the exchange.

Non-repudiation is a related issue.  What this simply means is that businesses and consumers should not be able to deny (repudiate) a transaction.  In traditional commerce, when paper documents are exchanged before goods and services are delivered, it is very difficult for either party to deny a transaction.  In contrast, the transaction trail in an e-commerce transaction is simply a series of bits and bytes.  Such an electronic transaction trail can be somewhat more difficult to establish and verify relative to a conventional paper trail.  Thus, when doing business over the Internet, non-repudiation of transactions can be a significant concern when the merchant and the consumer are essentially "out of sight" doing business paperlessly

Jurisdictional and legal issues: Related to the above concerns of authentication and non-repudiation, are concerns of jurisdiction and legality of transactions that can occur anywhere on the planet.  Taxation authorities are grappling with the issue of whether, how, and to what extent e-business activity should be taxed.  Consider the issue of sales tax.  Assume that you live in New York and you purchase a product over the Internet from a vendor in California.  However, the vendor's web site is actually located in Texas.  And the goods that are delivered to you are actually stored in Illinois.  Moreover, when you transmitted the order over the Internet, it actually passed through six other states (recall the packet switching protocol of the Internet).  In which states, if any, do you owe a sales and/or local tax?  If there is a dispute regarding this transaction, which state has legal jurisdiction over the transaction? There are no clear answers to these questions.  A host of government and non-government institutions are attempting to resolve these issues, for it is critical they be resolved in the near future if e-business is to achieve its true potential.


Summary

This chapter began by discussion a variety of basic telecommunications concepts. The five components of a telecommunications network were then discussed. The first component, terminals and workstations, are what the end user interacts with. The second component, transmission links, includes physical links such as twisted pair wire, coax cables, and fiber optic cables, as well as "through the air" transmission links such as infrared, microwave, and satellite links. The third component is transmission methods which comprises either digital or analog transmission involving either circuit switching or packet switching, with packet switching being the method used for Internet transmission using the TCP/IP protocol. Nodes and switches represent the fourth component and include hardware such as multiplexers, front-end processors, bridges, routers, and gateways. The final component is network architecture and standards. The Open Systems Interconnection (OSI) model was discussed in some detail. Next, local area networks (LAN) were described, including the three most common LAN topologies -- bus, ring, and star. Wide area networks were then discussed. Alternatives to modems used over conventional phone lines were presented. The various types of wide area networks, as well as distributed data processing networks, were also discussed. The basics of the client/server architecture were discussed. Finally, the chapter concluded with a discussion of the Internet, electronic data interchange, and electronic commerce.

 

 

 


Key Terms

Analog transmission
Bandwidth
BBBOnline
Bridge
Brouter
Bus network
Circuit switching
Client/server architecture
Coax cables
CPA WebTrust
CSMA/CD
Digital transmission
DSL
EDI
Fiber optic cables
Front-end processor
Gateway
Infrared transmission
Internet
Intranet
ISDN
LAN
Modem
Multiplexer
Network interface card (NIC)
Network switching
OSI model
Packet switching
Protocol
Ring network
Router
Star network
TCP/IP
Terminals
Token passing
Twisted pair wire
VAN
WAN
WATS
Workstations 


Key Web Sites

 

 

 


Discussion Questions

  1. Identify the five major components of a telecommunications network.
  2. Distinguish between terminals and workstations.
  3. Briefly describe alternative physical and "through the air" transmission links.
  4. Distinguish between circuit switching and packet switching.
  5. What is the function of a multiplexer?
  6. What services does a front-end processor provide?
  7. Identify and briefly describe the major types of switching devices and their function.
  8. What are the components of a local area network?
  9. What do you understand by the term "network protocol"? Explain.
  10. Briefly describe the OSI telecommunications model and its seven layers.
  11. Distinguish between bus, ring, and star networks in terms of their configurations and their message passing conventions.
  12. What is a modem? What are alternatives to conventional modems?
  13. Describe ISDN and DSL communications technologies.
  14. Distinguish between a centralized data processing network and a distributed data processing network. Indicate the advantages and disadvantages of each.
  15. What are the alternative configurations of a centralized WAN? Discuss the advantages and drawbacks of each configuration.
  16. What is a client-server system? Compare and contrast client-server and file-server (mainframe -- dumb terminal) systems.
  17. Distinguish between the terms "internet" and "intranet."
  18. What are the advantages of EDI over conventional business transaction processing between organizations?
  19. What are the three connectivity options for businesses considering establishing EDI links between their trading partners?
  20. What are the four major categories of electronic commerce? What are the major issues that could potentially impede the growth of electronic commerce?

 

 

 


Problems and Exercises

1. Consider the seven layer OSI telecommunications model. 

  • Layer 7: Application
  • Layer 6: Presentation
  • Layer 5: Session
  • Layer 4: Transport
  • Layer 3: Network
  • Layer 2: Data link
  • Layer 1: Physical

Required:
Indicate the layer(s) associated with each of the following tasks or features. 
a) Performed by hardware components: __________________ 
b) Data manipulation tasks: ____________________ 
c) Performed by software components: ______________________ 
d) Defines the way data is formatted, presented, converted, and coded: _________________ 
e) Defines protocols for the message structure: ___________________ 
f) Validates the integrity of the flow of data between nodes: ________________ 
g) Data communications tasks: _______________________ 

2. Consider the following diagram of a wide area network.

 

Required:
Fill in the blanks below: 

a) The above network is in a _____________ configuration. 
b) Nodes A, B and C are _________________________. 
c) Node D is a ___________________. 
d) Nodes E and F are _________________________. 
e) Node G is called the _____________________. 

3. Ms. Jane Smith is the manager in charge of the local Pizza Queen chain of restaurants. There are four Pizza Queen restaurants in the city. Currently, each location has a personal computer that handles all the sales processing needs of that location. At the beginning of every day, the manager of each restaurant delivers a disk to Ms. Smith's office providing details of sales transactions during the previous day. An auditor in Ms. Smith's office is responsible for reconciling the disk reports to the cash register tapes at each restaurant. This reconciliation is done weekly, so errors and irregularities can remain undetected for several days. Ms. Smith is considering establishing a network connecting all four restaurants to the central administrative office and seeks your advice in that regard. 

Required:
What are the various networking options which Ms. Smith could choose from? Although exact dollar estimates are not required, be sure to indicate the approximate costs of each networking option that you present. Which of these options would you recommend? Why? 

4. ABC Inc. is a distributor of oil drilling equipment in the State of Texas. The company has five warehouses located in Amarillo, Lubbock, El Paso, Odessa, and McAllen. All management tasks are handled at a central administrative office in Dallas. Currently, the Dallas office has a minicomputer and each warehouse has a dumb terminal connected via a dedicated line to the Dallas office. 

Required:
a) What kind of WAN does ABC currently have? What are the advantages and disadvantages of this WAN configuration? 
b) Discuss the advantages and disadvantages of alternative WAN configurations which ABC could have employed? 

5. Mr. Harry R. Blocker is the sole proprietor of a CPA firm. A powerful personal computer currently handles all the information processing needs of his firm. The computer does not have a modem. Mr. Blocker has been hearing a lot about the Internet and is considering obtaining access to the Internet. He seeks your advice in that regard. 

Required:
a) Indicate what additional hardware and software would be required (you may assume that the firm has multiple phone lines). 
b) In addition to conventional modems, what other technologies might be potential candidates? 
c) In addition to hardware and software, what else would be required in order to obtain access to the Internet? 
d) Discuss the advantages and potential drawbacks of providing Internet access on the firm's single computer. 

6. Visit the Web site of any bank offering Internet banking services. Write a brief report outlining the nature of services provided by the bank of your choosing.

 

 

Last Updated: July 20, 2001

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Copyright 1996-2001 CyberText Publishing, Inc. All Rights Reserved