Chapter 4 A Graphical Introduction to the Structural Elements of

Chapter 4

A Graphical Introduction to the Structural Elements of Cyberspace

Elihu Zimet and Edward Skoudis

THE MAJORITY of Internet users do not have (or need) much understanding

of what is

going on “under the hood” of cyberspace, beyond use of their keyboard, display, and favorite

applications. To further the understanding of other chapters of this book by readers without a

technical background, this chapter explains various elements of cyberspace, particularly the

Internet, with a graphical presentation.

The purpose here is to shed light on elements of cyberspace by example, so there is little

attempt to be comprehensive. The focus is on Internet technology,

particularly networks using the

Internet protocol (IP). However, because legacy

telephone networks, cellular networks, and

cable networks, as well as private

enterprise, government, and military networks, increasingly

use Internet technology, we discuss them briefly as well.

The structure of cyberspace will change as new technology is developed

and employed

in a variety of new applications and social structures. Thus, the

representation here reflects a

snapshot of the present only. Figure 4–1 presents the domains of cyberspace addressed in this

chapter.

Figure 4-1 Domains of Cyberspace

The systems domain comprises the technical foundation, infrastructure, and architecture

of cyberspace. It includes hardware and software, as well as the

infrastructure items

supporting them, such as the electrical power grid.

The content and application domain contains both the information base that

resides in

cyberspace and the mechanisms for accessing and processing this

information.

Communications among people and interactions between people and information occur

in the people and social domain. Businesses, consumers, advocacy

groups, political campaigns,

and social movements are in this domain.

The governance domain overlays all of the aspects of cyberspace, including the

technological specifications for the systems domain, the conventions for data formatting and

exchange in the content and application domain, and the legal frameworks of various countries

associated with the people and social domain. This chapter focuses on the first three domains

defined above (the governance domain is described in chapter 21 of this book, “Internet

Governance”).

The Systems Domain

The systems domain of cyberspace is the infrastructure that carries, stores,

and

manipulates information. Hundreds of millions of computer and network systems interact in

cyberspace, carrying information generated by over a billion people. A major portion of the

modern economy is associated with manufacturing

the components and systems of cyberspace,

including computer chips, desktop

computers, routers, servers, and operating systems. Another

major component of the economy is associated with operating this infrastructure, including

Internet

service providers (ISPs), telecommunications firms, electrical power companies, and

other organizations.

The global open communications backbone—including telecommunications, cable,

cellular, Internet, and other public networks—is the principal infrastructure for both civil and

military communications today. In addition, many militaries, governments, and commercial

enterprises also employ closed private

networks. Many of these private networks have

interconnection points with

public communications networks, either integrated as part of their

design or created on an accidental, ad hoc basis.

Network Building Blocks

Most networks are made up of smaller subnetworks. For example, among the

subnetworks of the telephone network are smaller telephone networks operated

either by

telephone companies or independent enterprises, all of which are interconnected at various

points. The original telephone network employed

switchboards on which telephone operators

established a stable circuit between

two or more parties. The circuit remained dedicated to

those particular users

as long as the call was maintained. In such circuit-switched networks, the

network allocates resources—originally a physical connection using switched wires, later

timeslot in a protocol—for every communicating session, established in real time, to move

information across the network. With circuit switching, then,

the network is “aware” of

individual communicating sessions, tracking them

constantly and allocating resources for each

of them.

Instead of the circuit switching of the traditional telephony network, the

Internet relies

on the alternative technology of packet switching. Instead of the

network dedicating resources

between the end systems for each communications

stream, the data that make up that

communication are broken up into separate packets or chunks of data and each is delivered

independently by the network to

the desired destination. “Header” information is attached to each

packet to help the network move the packets to the right place. The network itself is generally

ignorant of the relationships between the packets but instead focuses on getting

individual packets

to their destination as quickly as possible.

The fundamental packet-switching network building block of the Internet

is a local area

network (LAN), an extremely focused form of subnet in a relatively localized geographical

area. Most enterprise networks for corporations, government agencies, and military

organizations are groups of LANs.

Increasingly, LAN technology is being deployed in

homes so that disparate devices such as personal computers, voice over IP (VoIP) telephones, and

servers

can communicate with each other. Components on a single LAN are typically

connected using a switch or wireless access point, perhaps using wired Ethernet

or WiFi wireless

technology; these simple devices send packets directly between

interconnected systems over

short distances ranging from one to a few hundred meters. Figure 4–2 shows a common LAN

setup.

Figure 4-2 A Local Area Network

LANs are connected by routers, devices that can direct packets to their

destination

subnets based on the addresses in the packets themselves. Routers

can also connect LANs to

point-to-point links to create even larger networks.

Point-to-point link technologies include T1,

Digital Signal 3 (DS3, also known

as T3), cable modem, fiber optic service (FIOS), and

digital subscriber line (DSL), a relatively higher speed network connection over telephone

network wires. The result is a network of networks, or internetwork, of a multitude of

interconnected LANs, routers, and point-to-point links.

Figure 4–3 depicts an internetwork of various LANs, routers, and point-to-point links,

such as might be found inside a small or medium-sized enterprise.

Figure 4-3 Internetworking Local Area Networks via Routers and Point-to-Point Links

The Internet is a publicly accessible internetwork with global reach tied

together by

means of common use of IP. Physically, the Internet is made up of a group of “backbone” routers

operated by top-tier ISPs (see figure 4–4).

Figure 4-4 Internet Links to Backbone Routers Operated by Internet Service Providers

A few hundred high-end routers distributed around the world constitute this backbone,

making up the major branches of the network architecture, moving packets

across ISP

networks and also between ISPs. The top-tier ISPs offer smaller ISPs

access to the Internet

backbone. Both the top-tier and smaller ISPs also give

enterprise networks, individuals, and

content providers access to the backbone infrastructure. In the United States and some other

countries, numerous top-tier

ISPs compete for this business. In other countries, a single

dominant provider,

often the legacy telephone company, acts as the only ISP.

From the introduction of the Internet in the early 1970s until today, the legacy

communications network has provided connectivity for networked computers in addition to

telephones. Other types of enterprises, including cable companies,

wireless operators, and

dedicated data communications carriers, have also begun

to offer carriage of Internet data as part

of their business. Bandwidth has been

increased significantly through the use of better signal

processing, more rapid transmitters, and fiber optic communications.

The communications networks require a significant support infrastructure, including the

electrical power grid, required for operating all components of the networks. Supervisory

control and data acquisition (SCADA) devices,

which consist of simple computers directly

connected to industrial equipment,

are used to monitor and manage many types of utility

networks and industrial operations, including electrical power grids, gas pipelines, and

manufacturing

lines. Increasingly, the SCADA devices themselves are managed remotely across

a private network or even the public Internet. This approach introduces a significant

dependency loop: the communications infrastructure relies on the electrical power grid, which

is controlled with SCADA systems that rely on the

communications infrastructure. There is

thus is a risk of cascading failures.

Another vital aspect of the support infrastructure is the

manufacturing base,

which creates the physical equipment that makes up the various systems

of cyberspace. These, too, may rely on SCADA systems and their communications

networks,

again raising the risks of interdependency.

Another View: Protocols and Packets

So far, this chapter has analyzed how internetworks and the Internet itself are built up

of smaller building blocks such as LANs and point-to-point links. To further analyze the

elements of the systems domain of cyberspace, a different

view of networking can be helpful:

the protocol and packet view. This view helps

explain how different machines connected to the

same Internet communicate

with each other by sending packets according to standard protocols.

Protocol layering is a critical concept underlying this view of network

communications. In 1980, the International Organization for Standardization released the Open

Systems Interconnection (OSI) Reference Model, a generic

description of how computer-to-

computer communications could operate using

protocol layering. In this model, a series of

small software modules on each system perform a set of tasks that allow two computers to

communicate with

each other. For example, one module might focus on making sure that data

are formatted appropriately, another takes care of retransmitting lost packets, and yet another

transmits the packets from LAN to LAN across the network. Each of these modules, referred to

as a layer, has a defined small job in communication (see figure 4–5).

Figure 4-5 Protocol Layering

The software of a given layer on the sending machine communicates

with the same

layer on the receiving machine. A layer is a collection of related

functions that provides

services to the layer above it and receives service from the layer below it. For example, one

lower layer might send packets on behalf of the higher layer that is focused on retransmitting lost

packets. This higher layer,

in turn, serves an even higher layer that generates the data in the first

place. In the example of figure 4–5, a layer of software inside a Web browser generates data to

send to a Web server. This Web browser application passes the data to

the transmission control

protocol (TCP) layer software on the sending machine,

which provides several services,

including retransmitting lost packets. The TCP

layer passes the software down to the IP layer,

which provides the service of carrying the packet end to end through all the routers on the

network. Although one layer relies on another to get things done, the layers are designed so

the

software of one layer can be replaced with other software, while all other layers

remain the

same. This modularity has proven especially useful in deploying new

types of networks—for

example, as IP version 4 (IPv4) networks are transitioned

to IP version 6 (IPv6), the successor

protocol for the Internet.

The communications modules taken together are called a protocol stack

because they

consist of several of these layers, one on top of the other. The OSI conceptual model defined by

the International Organization for Standardization in 1980 includes seven such layers, each with a

defined role in moving data across a network. At the “top,” layer seven, the application layer, acts

as a window to the

communications channel for the applications themselves by interpreting data

and turning it into meaningful information for applications. The application might

be, for

example, a Web browser or server, an email reader or server, a peer-to-peer file copy program,

or an enterprise financial system.

Layer six, the presentation layer, deals with how data elements will be represented for

transmission, such as the order of bits and bytes in numbers, the

specific method for encoding

textual information, and so on.

Layer five, the session layer, coordinates sessions between two communicating

machines, helping to initiate and maintain them as well as to manage

them if several different

communications streams are going between them at the

same time.

Layer four, the transport layer, supports the reliability of the communications

stream

between two systems by offering functions such as retransmitting lost

packets, putting packets

in the proper order, and providing error checking.

Layer three, the network layer, is responsible for moving data across the network from

one system, possibly across a series of routers, to the destination

machine. This layer is

absolutely critical to making the network function end to

end.

Layer two, the data link layer, moves data across one “hop” of the network, getting it

from one system, perhaps to its destination on the same LAN, or to the

nearest router, so it can be

sent between LANs or to a point-to-point link.

At the “bottom” of the stack, layer one, the physical layer, actually transmits

the bits

across the physical link, which could be copper, fiber, wireless radio transmitters and receivers

or another physical medium.

Today’s Internet is loosely based on the OSI model, but it does not break out

each layer

exactly as the OSI model specifies. Most commonly, IP is paired with a transport protocol called

the transmission control protocol (TCP)—hence the

term TCP/IP to refer to the duo of

protocols in most common use on the

Internet today. TCP/IP is roughly equivalent to layer four

(transport) and layer

three (network) of the OSI Reference Model, plus a little interaction with

layer

two (data link). Everything above TCP/IP is left to the application with the

application,

presentation, and session layers (seven, six, and five) of the OSI Reference Model all folded

into the application program itself. TCP/IP is mainly responsible for transmitting data for that

application. It is important to note

that the application layer is not TCP/IP itself: the

application comprises the

particular program trying to communicate across the network using

TCP/IP.

The application might be, for example, a Web browser and a Web server, or two mail

servers, or a video player communicating with a streaming video server, or a file transfer

protocol (FTP) client and server. Based on the OSI model, the application layer is often

referred to as layer seven, even in TCP/IP networks.

The transport layer could be the transmission control protocol or its cousin,

the user

datagram protocol (UDP), a simpler but less reliable protocol. The

transport layer ensures

that packets are delivered to the proper place on the

destination machine. For those

applications requiring such functionality, TCP

also delivers packets in the proper sequence or

retransmits packets.

The network layer is based on the IP. Its purpose is to deliver packets end to

end across the

network, from a source computer to a given destination machine.

Using terminology from the

OSI Reference Model, the IP layer is sometimes

referred to as layer three.

The data link layer transmits each packet across each hop of the network. For example, this

layer moves data from a home computer to a router that connects

the LAN to the Internet. Then,

the router uses its data link layer software to

move the data to another router. In the OSI

Reference Model vernacular, the

data link layer is often referred to as layer two.

The physical layer, called layer one, is the physical media, such as the wire,

fiber optic

cable, or radio frequencies, across which the information is actually transmitted.

To illustrate how these layers on IP networks typically communicate, figure 4–6 shows

an example in which two computers, the sender machine and the receiver machine,

communicate.

Figure 4-6 Protocol Layering to Transmit Packets on an Internet Protocol Network

Suppose a user on the sender machine wants

to surf the Internet with a Web browser

application such as Internet Explorer or Firefox to access a Web site. The browser on the sender

needs to communicate with the Web server on the receiver, so it generates a packet and passes it

to the TCP/IP stack software running on the sender machine. The data, which consists of a Web

request, travel down the communications layers on the sender

to the physical layer and get

transmitted across the network (which usually

consists of a series of routers). The packet is

sent through one or more routers

this way, until it reaches the receiver machine. It then travels

up the receiver’s

communications stack.

To start this process, the sender’s transport layer (that is, TCP software running on the

sender machine) takes the packet from the browser application and formats it so it can be sent

reliably to the transport layer on the receiver.

This TCP software also engages in a packet

exchange (called the TCP Three-Way

Handshake) to make sure all of the sender’s packets for this

connection arrive in sequence. (Other types of transport layer protocols, such as UDP, do not

care

about sequence, so they have no such packet exchange for ordering packets.)

Just as the two applications, here the Web browser and the Web server,

communicate

with each other, so do the two transport layers. On the sender,

the transport layer passes the

packet down to the network layer, which delivers it

across the network on behalf of the transport

layer. The network layer adds the

source and destination address in the packets, so they can be

transmitted across

the network to the receiver’s network layer. Finally, the data are passed to

the

sender’s data link and physical layers, where it is transmitted to the closest router on the way

to the destination. Routers move the packet across the network, from subnet to subnet. The

routers include the network, data link, and physical

layer functions required to move the packet

across the network. (Because these

routers are focused on moving packets, not receiving them,

they do not require

the transport or application layers to deliver the packet to the receiver.) On

the

receiver side of the communication, the message is received and passed up the

protocol

stack, going from the physical layer to the data link layer to the network layer to the transport

layer to the ultimate destination, the application.

This passing of data between the layers is illustrated in figure 4–7.

Figure 4-7 Protocol Layering Applying Various Headers and a Trailer

Each layer attaches some information in front of (and in some cases, behind) the data it

gets from the layer above it. The information added in front of the data is called

a header, and it

includes critical information for the layer to get its job done. As

figure 4–7 shows, the

application generates a packet. This packet might be part of a Web request, for example, or a

piece of an email message. The transport

layer adds a header to this data, which is likely to

include information about where on the destination machine the packet should go. When TCP is

used, the

resulting header and data element is called a TCP segment. The TCP segment gets

passed to the network layer, which adds another header with information about the source and

destination address in the IP header. This header is analogous to

an envelope with a postal address

for the data. The resulting packet is called an IP

datagram. This package is sent to the data link and

physical layers, where a header (and a trailer) are added to create a frame that makes it possible

for the data to be

transmitted across the link.

The packets sent between machines pass through different layers of this

stack and have

various headers in front of them. The one layer that all systems

and routers communicating with

each other on the network must conform to is

the network layer; for the Internet, the network

layer is the Internet Protocol.

Today’s Internet relies on IPv4 from one end to the other. Certain

subnets and routers also support the successor protocol, IPv6. Every packet sent across the

Internet today using IPv4 has the structure shown in figure 4–8. It includes a source IP address

(a 32-bit number indicating the network address of the system

that sent the packet) and a

destination IP address, which routers use to determine

where to send the packet.

Other fields of

the IPv4 header are associated with controlling the number of hops a packet can take as it

traverses the network (the

time-to-live field), fragmentation (which breaks larger packets into

smaller ones), and other network functions.

Figure 4-8 An Internet Protocol Version 4 Packet

Although the packet in figure 4–8 looks two-dimensional, that formulation is done

purely for diagram purposes. Each field is transmitted on the wire (or wireless radio) one byte

after another, linearly. The version number (such as IPv4)

goes out first, followed by the header

length (the size of the overall header in front of the data), field by field as shown, through total

length, the identification field,

which is associated with fragmentation, and so on. Bit after bit, the

packet leaves

the sender source. When it reaches a router, the router absorbs the packet on one

network interface, inspects its various fields, and moves it to the appropriate network interface for

transmission to its destination, or one router hop closer to

its destination. In this way, a packet is

sent and routed across the Internet.

The Content and Applications Domain

The systems domain of cyberspace provides the technical underpinnings of the

network, but it is merely an infrastructure on which to store, transmit, and manipulate content

or information using various software applications. The

content and applications domain rides

on top of the systems domain and provides

the usable applications and the information they

handle. This section describes

some of the common methods for content storage in cyberspace,

provides an overview of various application architectures, and outlines some of the most

common application types used today.

Content Storage

Although content is stored on computer systems in cyberspace in many ways,

two

methods of information storage dominate: hierarchical file systems and relational databases.

Nearly all cyberspace applications rely on at least one of these concepts; most applications

use both. In a hierarchical file system, as shown in figure 4–9, one or more computers stores

information in individual files, which themselves are made up of small sections of the hard drive

or chunks of memory, depending on the computer system.

Figure 4-9 A Hierarchical Files Structure

Files on a typical computer

system include the software of the operating system itself

(such as Windows,

Linux, or Apple’s Mac OSX), executable programs that make up the

applications of the computer (such as Adobe Acrobat, a Web browser, or a word processing

program), configuration files that hold settings for the programs on the machine, and the data

stored on the system by the application and its users, such as

document files. Files are located

inside of directories (also called folders on some operating systems). The directories themselves

may contain subdirectories. This results in a hierarchical structure.

Today, many computer users take this structure for granted, and may assume

that this

organization of content is the way that computers necessarily work. But

early computers did not

have such a structure. Development of the hierarchical file structure revolutionized human

interactions with machines. A hierarchical file system is useful because it provides an

unambiguous way to refer to files, indication that there is a relationship between files—for

example, files in the same directory are likely to have some common purpose—and a method

for navigation between files. This paradigm is so useful for information storage that it is found in

all of the major general-purpose operating systems available today—

including Windows, Linux,

UNIX, Mac OSX, and others—across various kinds

of computing equipment, including laptop

computers, desktops, and servers as

well as many cell phones, music players, and video storage

devices. Even systems

without hard drives are increasingly likely to use a file system

implemented in memory.

Another common method for storing content involves a relational database. As illustrated

in figure 4–10, a relational database stores information in a series of tables. The tables specify a

number of fields under which specific types of data, called values, are stored. In the value

section of each field, all of the elements in a given row of the table relate to each other as a

single record in the database.

Two or more tables may have the exact same field in them to forge

a relationship between them. Such a relationship is illustrated by field C in the figure.

Figure 4-10 A Simple Relational Database

For example, suppose the database in the figure is associated with electronic commerce.

Table 1 might hold transaction information, perhaps indicating items

various people have bought

from an online store. Table 2 might hold customer

credit card information. Within table 1, fields

A, B, and C could be, respectively, the date, the item purchased, and the account number of the

purchaser. Table 2

could also hold field C, the account number, and, in addition, field D, the

credit card number for that customer. Data are presented in these tables discretely, for flexible

search and update. A program or user could write a query and pull out or update information

directly from table 1, analyzing transactions or adding new ones as purchases are made.

Because of the relationship between the tables,

which is embodied in field C (the account

number), database query software can research, for example, the number of times a given credit

card number (table

2 data) was used on a given date (table 1 data). This “join” operation

offers

significant flexibility in handling and manipulating content. Of course, figure

4–10 is a

highly simplified diagram to illustrate the fundamental constructs of a relational database.

Most real-world databases have dozens or hundreds of tables, each with 2 to 20 or more fields

and thousands to millions of rows.

Relational databases are the most popular form of database storage today

in market

share. However, other kinds of database storage options are available,

including hierarchical

databases, object-oriented databases, and flat files, each offering different performance and

functions from relational databases.

Although relational databases and hierarchical file systems are fundamentally different

paradigms for organizing content, the two have some connections.

Relational databases are, for

example, almost always implemented as software running on top of a hierarchical file system.

That is, the database programs themselves consist of a group of hierarchically organized files.

The database tables and their contents exist inside of files as well. Just as large-scale packet-

switched networks are often built up of point-to-point links from circuit-switched

networks,

relational databases are an overlay on top of a file system, offering a more flexible way of

organizing information for query and update. In addition, most applications in cyberspace today

are a mixture of files interacting with a back-end database using a variety of application

architectures that manage the

interaction of the files and the database. Such systems are referred

to as back-end infrastructure to differentiate them from the front-end systems such as browsers

that users run to initiate transactions.

Application Architectures

From the 1950s through the 1970s, computer applications typically resided on

mainframe computers. Users logged into the system from “dumb” terminals.

These terminals

displayed information stored and processed on the mainframe but did no processing themselves.

They were connected to the mainframe using

fairly limited network protocols. All aspects of the

application were stored on the mainframe itself.

Starting in the 1980s, personal computers (PCs) with limited processing

power began

replacing dumb terminals. In the 1990s, many mainframes were

replaced with lower cost and

smaller server machines. Both enterprise and public

networks connecting PCs and servers began

to make significant use of IPv4. A variety of new, distributed application architectures arose.

Figure 4–11 depicts

some of the common consumer and business application architectures in

use

today on enterprise networks and on the Internet.

The common applications architecture depicted on the left of figure 4–11 is a traditional

client-server application. The client consists of a custom program dedicated to that application

running on a desktop or laptop computer. The

client handles the user interface, some data

processing, and communications

with a server, accessed across the network. The server may

include database

software, although some client-server applications instead place the database

on a separate machine, as shown in the second architecture of the figure. This separate database

can then support multiple application servers.

Figure 4-11 Common Modern Application Architectures

One of the concerns of the architectures on the left half of figure 4–11 is the

distribution and update of specialized client-side software. Each application has a custom

program that controls access to the server. Thus, synchronization of updates to several different

client applications is rather complex. In the mid-

1990s, various applications began to move to the

third architecture of the figure, which involves using a Web browser as a generic client that

formats, displays, and processes information for the user, relying on a single program—the

browser—

for multiple applications, thus simplifying client-side updates. This generic

browser

accesses a server across the network, which in turn accesses a back-end

database server. This so-

called three-tier architecture is the dominant enterprise and Internet application architecture

today. Some applications introduce an additional element, an application server, shown in the

fourth example of the figure. The application server handles some of the critical information

processing

capabilities that were previously handled by the Web server.

While the three-tier architectures shown on the right side of figure 4–11 will continue to

be dominant for several years, a newer architecture paradigm started

to emerge in the mid-2000s:

the so-called service oriented architecture (SOA)

model. Figure 4–12 illustrates this approach. In

this model, client-browser software still accesses Web servers; often, this interaction relies on the

Extensible Markup Language (XML), a flexible format for defining data structures to be

exchanged on the network. Web servers may, similarly, interact with other Web servers

using

XML to pull together information for the client, processing information in a distributed,

coordinated fashion on the network. This is sometimes called

computing “in the cloud.” Using

this SOA model, different functions, such as

geographical mapping, product search, or price

calculation and comparison, can be distributed across different servers. Applications can be

stitched together by

information flows between different Web-enabled services. An example

might offer a user a way to create a consumer grocery shopping list that minimizes

cost and

travel time. A user enters a desired grocery shopping list and starting

location into a browser.

The browser sends XML requests to a grocery pricing Web service that responds with prices at

various local grocery stores. This Web

service might pass on a request to a separate street maps

Web service, asking for directions between the consumer and different stores. The service might

then

forward a request to a traffic service that provides information about clogged highways.

All of the information is reassembled at the user’s browser, indicating

the most efficient options

for buying the desired groceries.

Common Application Types

The architectures shown in figures 4–11 and 4–12 are used to create many

application types on internal networks and the Internet. New applications are constantly being

devised and becoming popular. Some of the most common applications on today’s Internet

include email, instant messaging, search engines,

and others described below.

Email is one of the oldest and most widely used applications of the

Internet. Many

companies are heavily dependent on their email infrastructure

for communication between

employees and also as the technical underpinnings of vital business processes. Email servers

exchange email messages, which are accessed by users running either specialized email reading

software or a general-purpose Web browser. In addition to user-to-user email, hundreds of

thousands

of mailing lists are used by communities with shared interests to exchange

information to all subscribers. Email is also used to spread malicious code and online scams.

Figure 4-12 A Service-Oriented Architecture

Email is store-and-forward technology: messages wait for users to access

them. By

contrast, instant messaging focuses on real-time exchange of messages between users. Many

such chats occur one-on-one, as two users send

information to each other. Chat rooms allow many

users to exchange information simultaneously. America OnLine, Yahoo! Messenger, and

Microsoft Messenger,

as well as independent, open standards for messaging such as Jabber and

Internet

Relay Chat, are widely used on the Internet. An alternative technology used for cell

phones, called Short Messaging Service, has much the same functionality but

different underlying

communication protocols.

Search engines are another important set of applications on the Internet.

Search engines

employ their own specialized browsing machines, known as

crawlers, to fetch billions of

pages from the Internet by following links from one page to the next, discovering new pages

as they are posted and linked to

other pages. The search engine company’s software then

assembles a searchable index of the pages fetched by the crawlers, offering users a Web-based

front-

end service by which users can search the massive index, which is stored in a

distributed database maintained by the search engine company.

E-commerce applications have burgeoned since the late 1990s, as retailers

have set up

Web sites to sell products and services to users. Consumer e-commerce companies such as

Amazon.com and Apple’s iTunes Store are among the most

familiar. Numerous business-to-

business e-commerce applications are also in use.

Business management systems have moved

online to automate business

purchasing, payroll, supply-chain management, and other vital

aspects of enterprise operations. Many organizations have deployed enterprise resource planning

and enterprise resource management systems such as those offered by SAP, PeopleSoft,

Oracle, and Microsoft.

Wikis are Internet-based applications that use specialized Web server software to inventory,

categorize, and store information about a given topic or group of topics and allow it to be created

and updated quickly. The most widespread and widely used wiki is Wikipedia, a comprehensive

encyclopedia of information that can be updated by anyone who wishes to contribute to the

information.

Blogs (a shorted form of “web logs”) are another important kind of Internet

application,

made up of online diaries that are frequently updated. Most blogs are devoted to a single topic,

such as politics, news, sports, a business or industry, or hobbies. Hundreds of thousands of blogs

are available today, and some acquire

readership numbering in the hundreds of thousands of

users per week.

Social networking sites, another fast-growing form of Internet application, allow users to

store information online in a personal profile and then link that profile to those of friends or

business associates. By following these links, people with related interests can reach out to each

other, keeping personal relationships up to date via socially oriented services such as MySpace or

Friendster, or making business connections using professional services such as LinkedIn and

Orkut.

In other applications, users are increasingly turning to the Internet for audio and

video news, information, and entertainment. Such services are pack-

aged as podcasts, audio or

video shows that are automatically downloaded to

user machines periodically when new content

is published by the podcast author.

Radio shows, television programs, and content from other

mass media are published in tens of thousands of free and commercial-subscription podcasts

on a daily basis. Downloadable and streaming video services with short-form

videos are also

becoming quite popular; offerings include those of YouTube and Google Video. Increasingly,

entire television programs and movies are available on the Internet through related services,

such as Amazon’s Unbox and the

television studios themselves.

Peer-to-peer file-sharing services allow large number of users to share files

on the

Internet, copying music, movie, and other files automatically between user systems. User

machines that join the peer-to-peer network are used to store files

that are redistributed to other

users without any direct user interaction. Some of the most popular peer-to-peer networks are

BitTorrent, Gnutella, and eDonkey.

Many of these networks are used to exchange song and

movie files in violation of copyright, but they are also increasingly used by legitimate

publishers to

distribute movies and television shows on a pay-per-view basis.

Internet telephony applications such as Gizmo and Skype offer free or very

low-cost

telephone calls from a user’s computer across the Internet. Gizmo and Skype carry calls from

end-user machine to end-user machine without traversing

any intermediary servers. Vonage, on

the other hand, carries calls via VoIP

servers, which provide value-added services, such as

voice mail. Because they

often undercut the cost of long distance and international calls, both

kinds of Internet telephony services are starting to supplant more expensive traditional

telephone calls.

Another important kind of Internet application bridges cyberspace and the

real world:

mapping applications. These tools, often offered for free on the Inter-

net, provide comprehensive

street maps and driving directions on the Internet;

MapQuest and Google Maps are dominant.

With virtual reality sites, users are presented with a well-developed “virtual

world” that

they can explore using an avatar, a digital icon representing their

online persona. People meet

in virtual reality sites for a variety of reasons,

including social interactions, business

relationships, and online gaming. Second

Life is one of the most popular virtual reality sites, in

which users interact in social and business settings. Many online games are also a form of virtual

reality

simulation, including the popular World of Warcraft game. Within some virtual

reality

simulations, different “cities” are founded with different focuses, such as

a given hobby or

business pursuit.

While these are among the most popular applications on the Internet today,

the content

and applications domain of the Internet is constantly changing and growing. New applications

rise to prominence frequently, as they empower people

to interact with their information and with

each other on a more flexible basis.

The People and Social Domain

Cyberspace is a human-made artifact, created to gain access to information and share it

between people and machines. With the increasing presence of cyberspace

in modern life, the

underlying systems domain and content and applications domain have given rise to a new

people and social domain as individuals build communities in cyberspace. Numerous types of

online communities exist in cyberspace, visited by users from around the world who share

personal, business,

political, or other interests. These communities are sometimes built using a

single

application: a community may be entirely based on an email list of hobbyists. More

commonly, however, online communities consist of people who use multiple

applications for

different aspects of their community; the hub might consist of a particular blog, augmented with

mailing lists, sections in video-sharing sites, chat rooms, or cities within a virtual reality

simulation, as shown in figure 4–13.

Figure 4-13 Online Communities Built of One or More Applications

Such cyberspace communities have been founded for numerous purposes.

New

communities form on a regular basis and older communities dry up.

Flourishing

communities today include those built around social interactions,

such as dating sites, teenage

hangouts, and corporate “virtual water coolers.”

Nearly every hobby imaginable - model

rocketry, chess, recipes, and more—has a Web site, and many have full-blown online

communities.

Online news

communities include the Web sites of major news services

such as CNN and the New York

Times, local newspapers, local television news,

professional bloggers who work as independent

journalists supported by online ads, and amateur bloggers who fact-check other news sources

and provide

commentary. Over 50 million Americans read news online, according to a

2006 survey by the Pew Internet and American Life Project, and the number is

trending upward

rapidly.

Numerous online communities are devoted to discussions of a given type

of technology,

helping users understand and relate to the technology more effectively, such as the Web sites

Slashdot, devoted to open-source software and related technologies, Ars Technica, focused on

detailed aspects of PC hardware and software, and Gizmodo, focused on technical gadgets and

consumer

electronics.

Many health care–oriented online communities help doctors and patients

information to better understand and cope with various medical conditions.

Online support

communities offer help for people suffering from particular

diseases.

Adherents of many religions, major and minor, have created online communities for

proselytizing, fundraising, and placing their messages and beliefs in front of a wider audience.

Political communities on the Internet are used for debate, analysis, and fundraising

activities. Several popular blogs from nearly every aspect of the

political spectrum, and their

associated mailing lists and online video distribution

sites, are increasingly helping to shape

political messages and candidacies.

Both business-to-consumer and business-to-business communities have

flourished in

cyberspace, helping to make business processes more efficient. Sites such as eBay and

Amazon.com offer storefronts through which numerous other organizations and individuals sell to

consumers. Some industries have established

their own online communities to improve

efficiencies in bidding and service delivery. For example, American and some international

automotive companies

created the Automotive Network Exchange.

Not all communities in cyberspace have beneficial impacts. Terrorists rely on

cyberspace to recruit, plan, and spread propaganda. International criminals likewise use

cyberspace to commit crime and track their business ventures, as

described in more detail in

chapter 18, “Cyber Crime.”

The walls between these types of communities are permeable: a single

individual may

participate in numerous communities. The communities themselves may also interconnect:

communities created for social interactions may

flow into hobbyist communities or start to

engage in political debate.

Conclusion

As cyberspace continues to expand, the diversity of elements in all of the

domains

discussed in this chapter is likely to increase. The underlying networks will become more

complex, pulling in other kinds of technologies and other protocols. The application mix will

be updated as people apply technology to

new uses. Online communities will continue to evolve

and merge as technology

grows more immersive and people grow more accustomed to living

various aspects of their lives in cyberspace.

This evolution of cyberspace also appears to be increasing in speed. Much of cyberspace

got started in the 1970s as computers were interconnected

and primitive email systems

were established. In the 1990s, Web sites for information distribution became popular. In

the early 2000s, search engines

and e-commerce flourished. Many of the most popular

applications and online communities are even newer, such as blogs, wikis, and virtual reality

applications. With this pace of evolution continuing to accelerate, cyberspace is

likely to have an

increasing impact on the economy and society in coming years.

For this reason, policy

decisionmakers need to understand the underpinnings of the technology and its rapidly changing

status to help inform their actions.

While IP-based networks are packet-switched technologies, individual links, especially

stable point-to-point links with constant

connectivity, are usually carried over circuit-

switched technology provided by legacy telecommunications, cable, or cellular net-

works.

IPv6 uses 128-bit addresses.

The hierarchical file system may or may not be visible to the user. Some operating systems and applications hide this organizational

structure from users, trying to simplify human interactions with the machine.

Queries for relational databases are usually formatted in the Structured Query Language

(SQL), a standardized language for interacting

with databases. Relational database software is available from a large number of commercial software companies, such as Oracle and

Microsoft. An open-source relational database called MySQL is also popular.

The Slashdot Web site is <http://slashdot.org>; Ars Technica is available at <www.

arstechnica.com>; and the Gizmodo Web

site is <http://gizmodo.com>.