Charles M. Kozierok The TCP-IP Guide

Подождите немного. Документ загружается.

Key Concept: Each command sent by the device acting as the client in an NNTP

connection results in the server returning a reply. NNTP replies consist of a three-

digit reply code and a string of descriptive text; they are modelled after those of

SMTP, and in turn, FTP.

441 posting failed POST command failed.

450

authorization required for this

command

Response sent when server requires authentication but

client has not yet authenticated.

452 authorization rejected Failed authentication.

480 transfer permission denied Response to CHECK if transfer is not allowed.

500 command not recognized Bad command.

501 command syntax error Bad syntax in command.

502

access restriction or permission

denied

Permission denied; sent if the client has not properly

authentication but the server requires it.

503

program fault - command not

performed

General fatal error message.

Table 273: NNTP Reply Codes (Page 3 of 3)

Code

Reply Text Description

TCP/IP World Wide Web (WWW, "The Web") and the Hypertext

Transfer Protocol (HTTP)

So finally, we come to the Big Kahuna. ☺ In my overview of file and message transfer

protocols, I said that the World Wide Web was “almost certainly” the most important TCP/IP

application. If anything, I was probably understating the case. The Web is not only quite

clearly the most important TCP/IP application today, it is arguably the single most important

application in the history of networking, and perhaps even computing as a whole.

Does this sound a little melodramatic? More than a little? Perhaps, but consider what the

Web has done in the decade or so that it has been around. It has transformed not only how

internetworks are used, but in many ways, has changed society itself. The Web put the

Internet “on the map”, so to speak, moving it from the realm of technicians and academics

and making it a big part of the mainstream world.

In this section I describe the World Wide Web in two subsections. The first discusses the

Web and the concepts behind hypertext and hypertext documents in general terms. The

second explains the operation of the all-important Hypertext Transfer Protocol (HTTP), the

TCP/IP application layer protocol that makes the Web work.

Note: As I've said in many other section headers, I simply cannot do complete

justice here to a topic as large as the World Wide Web. There have not only been

whole books written about the Web, there are shelves full of such books. Due to

the already large size of this Guide, and its overall focus on how protocols and technologies

work, I must contain my enthusiasm and limit my scope to providing an overview of what

the Web is all about, while focusing on the nuts and bolts of HTTP itself.

TCP/IP World Wide Web and Hypertext Overview and Concepts

The World Wide Web expands the concepts of messaging beyond the limits of simple text

file transfer of e-mail, FTP and Usenet. Its power is in its combination of hypertext, a system

that allows related documents to be linked together, its rich document format that supports

not just text but graphics and multimedia, and the special protocol that allows efficient

movement of those media. The result is a powerful system that once introduced, caught on

almost immediately amongst everyone from large companies to individuals, and in a few

short years came to dominate all other applications on the Internet.

In this section, I take a high-level, summarized look at the concepts behind the World Wide

Web. I begin with a short overview and history of the Web and hypertext, and a discussion

of the components that comprise the Web system. I briefly describe the documents and

media used on the Web and explain the importance of the Hypertext Markup Language

(HTML). I conclude with an overview of how documents are addressed on the Web using

Uniform Resource Locators (URLs).

World Wide Web and Hypertext Overview and History

At one point, I considered putting my section on the Web in a distinct location, separate

from the discussion of other file and message transfer protocols. It didn’t “seem” at first like

a message transfer protocol to me. But I realized that, boiled down, the Web really is just

another way of storing information and then transferring it from one computer to another. A

question immediately struck me: if this is the case, what was so special about the Web that

caused it to become popular in a way that no prior messaging applications ever had?

There is no simple one-word answer to this question. However, if I had to give one anyway,

it would be this: hypertext. Sure, applications like e-mail and Usenet allow users to send

and receive information, and FTP lets a user access a set of files on a server. But what

these methods lack is any way of easily representing the relationship between documents

or providing a way of moving from one to another. Highly simplified, hypertext does exactly

that: it allows the creator of a document to include links to related information either

elsewhere in that document or in others. With the appropriate software, a user can easily

move from one location to another.

Big deal? In fact, this is more important than it may initially seem. Without some way of

linking documents together, they remain in unconnected “islands”. In some ways, hypertext-

linked documents are to unlinked documents what networked computers are to those that

are not networked. Consider the difference between this Guide and a hard-copy book; if I

mention a topic in passing I can provide a link to it; a printed book cannot. Hopefully you’ve

already discovered the advantages that this offers. ☺

History of Hypertext

The ideas behind hypertext actually go back far beyond the Web and even electronic

computers. Vannevar Bush (1890-1974) is generally credited with introducing the idea in his

1945 description of a theoretical device called the Memex, which was intended to be used

to store and retrieve documents. He described the concept of a trail that would link together

related information to make it easier to organize and access the information in the device.

Bush's ideas were used as the basis of the work of several researchers who followed. One

of these was Ted Nelson, who coined the term “hypertext” and in 1960, first described a

system called Xanadu, which is considered one of the original hypertext software models.

The history of the World Wide Web itself goes back to 1989 at CERN, the European Organi-

zation for Nuclear Research, in Geneva. (The acronym stands for Conseil Européene pour

la Recherche Nucléaire, the French name of the organization.) Many of the projects under-

taken at CERN were large, complex, and took many years to complete. They also involved

many scientists who had to work with and share related documents.

Development and Growth of the Web

A researcher at CERN, Tim Berners-Lee, proposed the idea of creating a “web” of electron-

ically-linked documents. The rapidly-growing Internet was the obvious conduit for this

project. He designed the first (very crude and simple) version of the Hypertext Transfer

Protocol (HTTP) for TCP/IP in 1990. He was also responsible for developing or co-devel-

oping several of the other key concepts and components behind the Web, such as Uniform

Resource Identifiers (URIs) and the Hypertext Markup Language (HTML).

The ability to link documents and files together had tremendous appeal, and it took little

time before creative individuals found many different uses for this new technology. The

early 1990s saw a flurry of development activity. Web server and client software was

developed and refined, and the first graphical Web browser, Mosaic, was created by the

National Center for Supercomputer Applications (NCSA) in 1993. (The developer of this

program, Marc Andreesen, eventually formed Netscape Communications.)

Once the Web started to form, it grew very quickly indeed. In fact, to call the growth of the

Web anything but “explosive” would not do it justice. In early 1993, there were only 50

active HTTP Web servers. By late 1993, there were over 1,000. By late 1995, thousands of

new Web sites were coming online every day, and HTTP requests and responses had

overtaken all other TCP/IP application traffic. By the end of the decade there were millions

of Web sites and over a billion documents available on the Web.

The World Wide Web Today

While the rapid growth in the size of the Web is amazing, what is even more fascinating is

its growth in scope. Since you are reading a Guide about networking, you are most likely a

Web user, and familiar with the incredible array of different types of information you can find

on the WWW today. Early hypertext systems were based on the use of only text

documents; today the Web is a world of many media including pictures, sounds and

movies. The term hypertext has in many contexts been replaced with the more generic

hypermedia—functionally, if not officially.

The Web has also moved beyond simple document retrieval to providing a myriad of

services. A Web site can serve much more than just documents, allowing users to run

thousands of kinds of programs. This enables everything from online shopping to enter-

tainment. Web sites are also blurring the lines between different types of applications,

offering Web-based e-mail, Web-based Usenet access, bulletin boards and other inter-

active forums for discussion, and much more.

The Web has had an impact on both networking and society as a whole that even its most

enthusiastic early fans could never have anticipated. In fact, the Web was the ultimate “killer

application” for the Internet as a whole. In the early 1990s, big corporations viewed the Web

as an amusing curiosity; by the end of the decade it was for many a business necessity.

Millions of individuals and families discovered the wealth of information that was at their

fingertips, and Internet access became for many another “utility”, like telephone service. In

fact, the huge increase in Web traffic volume spawned the spending of billions of dollars on

Internet infrastructure.

The “dot-com collapse” of the early 21st century took some of the wind out of the Web's

sails. The incredible growth of the Web could not continue at its original pace, and has

slowed somewhat. But the Web as a whole continues to expand and mature, and will likely

be the most important information and service resource on the Internet for some time to

come.

Key Concept: The World Wide Web (WWW) began in 1989 as a project designed to

facilitate the representation of relationships between documents and the sharing of

information between researchers. The main feature of the Web that makes it so

powerful is hypertext, which allows links to be made from one document to another. The

many benefits of the Web caused it to grow in only a few short years from a small appli-

cation to the largest and arguably most important application in the world of networking; it is

largely responsible for bringing the Internet into the mainstream of society.

World Wide Web System Concepts and Components

Hypertext is the main concept that makes the World Wide Web more than just another

message transfer system. However, the idea behind hypertext had been around for

decades before the Web was born, as had certain software products based on that idea.

Obviously, more than just a concept is needed for an idea to be developed into a successful

system.

The Web became a phenomenon because it combined the basic idea of hypertext with

several other concepts and technologies, to create a rich, comprehensive mechanism for

interactive communication. This system today encompasses so many different concepts

and software elements, and is so integrated with other technologies, that it's hard to find

any two people who agree on what exactly comprises the Web, and which are most critical

parts.

For example, one of the keys to the success of the World Wide Web is undeniably the

combination of the TCP/IP internetworking protocol suite and the Internet infrastructure that

connects together the computers of the world. Is the Internet then an essential component

of the World Wide Web? In many ways it is; and in fact, due to how popular the Web is

today, it is common to hear people call the Web “the Internet”.

The functional components of the World Wide Web are illustrated in Figure 314. Let’s take a

look at them in a bit more detail now.

Major Functional Components of the Web

We know that equating the Web to the entire Internet is not a precise use of terms, but it

shows how important the Web is and how closely it is tied to the Internet. Still, while the

Internet and TCP/IP are obviously important parts of the Web's success, they are generic in

nature. When it comes to defining the World Wide Web system itself more specifically, three

particular components are usually considered most essential:

☯ HyperText Markup Language (HTML): A text language used to define hypertext

documents. The idea behind HTML was to add simple constructs, called tags, to

regular text documents, to enable the linking of one document to another, as well as to

allow special data formatting and the combining of different types of media. HTML has

become the standard language for implementing information in hypertext, and has

spawned the creation of numerous other related languages.

Figure 314: Major Functional Components of the World Wide Web

The World Wide Web (WWW)

Web Hardware

and Software

Web Server Hardware

and Software

Hypertext

Markup

Language

(HTML)

Web Client Hardware

and Browser Software

Hypertext

Transfer

Protocol

(HTTP)

Uniform

Resource

Identifiers

(URIs)

☯ Hypertext Transfer Protocol (HTTP): The TCP/IP application-layer protocol that

implements the World Wide Web, by enabling the transfer of hypertext documents and

other files between a client and server. HTTP began as a very crude protocol for trans-

ferring HTML documents between computers, and has evolved to a full-featured and

sophisticated messaging protocol. It supports transfers of many different kinds of

documents, streaming of multiple files on a connection, and various advanced features

including caching, proxying and authentication.

☯ Uniform Resource Identifiers (URIs): A method of defining labels that identify

resources on an internet so that they can be easily found and referenced. URIs were

originally developed to provide a means by which the users of the Web could locate

hypertext documents so they could be retrieved. URIs are actually not specific to the

Web, though they are most often associated with the Web and HTTP.

Note: Uniform Resource Locators (URLs) are a subset of Uniform Resource

Identifiers (URIs). The terms are often used interchangeably in World Wide Web

discussions.

All three of these were created and developed at around the same time, and taken together

represent the key technologies that define the World Wide Web. The next two topics in this

section describe HTML and the use of URIs in the context of the World Wide Web. HTTP is

really the heart of the Web and is covered extensively in its own section later in this Guide.

Web Hardware and Software

These three main components are supplemented by a number of other elements that play

“supporting roles” in rounding out the system as a whole. Chief among these are the

hardware and software used to implement client/server communication that makes the Web

work: Web servers and Web browsers.

Web servers are computers that run special server software to allow them to provide

hypertext documents and other files to clients who request them. Millions of such machines

around the world now serve as a virtual, distributed repository of the enormous wealth of

information that the Web represents.

Web browsers are HTTP client software programs that run on TCP/IP client computers to

access Web documents on Web servers. These browser programs retrieve hypertext

documents and display them, and also implement many of the Web's advanced features,

such as caching. Today's browsers support a wide variety of media, allowing the Web to

implement many different functions aside from simply hypertext document transfer.

Examples include displaying images, playing sounds and implementing interactive

programs.

Last but certainly not least, it is the users of the Web that are perhaps its most important

component. This is sort of “cheating”, as this is in some ways like defining the Internet as

being part of the Web. At the same time, I feel that user involvement has had more of a role

in shaping the development of Web technology than any other networking application. The

Web began as a simple means of exchanging documents; today it has grown to encompass

thousands of different applications and services, largely as a result of the creativity of users.

Content providers have pushed the boundaries of what the Web can do by creating new

ideas for information and services, to satisfy the insatiable demands of the end user

community.

Key Concept: The World Wide Web is a complete system that is comprised of a

number of related components, of which three are most essential. The first is

Hypertext Markup Language (HTML), which describes how hypertext documents are

constructed; HTML is what allows links between documents to be represented. The second

is the Hypertext Transfer Protocol (HTTP), the application layer protocol that moves

hypertext and other documents over the Web. The third is the Uniform Resource Identifier

(URI) mechanism, which provides a consistent means of identifying resources, both on the

Web and more generally on the Internet as a whole.

World Wide Web Media and the Hypertext Markup Language (HTML)

We've seen in the last two topics that the World Wide Web is based around the central

concept of hypertext. The prefix “hyper” usually means “above” or “beyond”, and thus

hypertext is like text, but goes beyond it in terms of functionality. Documents written in

hypertext are similar to regular text files, but include information that implements hypertext

functions. These are, of course, usually called hypertext documents or hypertext files.

The extra information in a hypertext document is used to tell the computer program that

displays the file to a user how to format it. This information takes the form of special instruc-

tions that are interspersed with the actual text of the document itself, which are written

according to the syntax of a defining language. This addition of extra elements to the

content of a document is commonly called marking up the document.

Overview of HTML

The standard language used by the World Wide Web is thus called the Hypertext Markup

Language (HTML). HTML is one of the three primary system components of the World

Wide Web, and was invented in 1990 by the creator of the Web, Tim Berners-Lee. It was

not created in a vacuum; rather, it is a specific application of the general concept of a

markup language that is described in ISO standard 8879:1986: Standard Generalized

Markup Language (SGML).

The idea behind a markup language is to define special items that provide information to

the software displaying the document about how it should be presented. For the purposes

of hypertext, the most basic type of information in a document is a special instruction that

specifies how one document can be linked to another—after all, this linking process is the

defining attribute of hypertext.

However, HTML goes far beyond just this; it defines a full set of text codes for describing

nearly every aspect of how a document is shown to a user. This includes instructions for

formatting text (such as defining its color, size and alignment), methods for displaying

tabular data, specifications for how to present images and other media along with the

document, interactive forms and much more. In theory, the language is only supposed to

define the document and leave up to the browser how it should be displayed, but in

practice, modern HTML documents also usually contain rather specific instructions for how

their information should be presented.

To do justice to HTML, I would have to devote several dozen pages to the subject. I have

decided not to do this, because even though HTML is an important part of the Web, it is

actually not that important in understanding how the Web works. Knowing HTML is

essential if you are writing Web content, and is also critical if you want to understand how to

write Web software. Perhaps ironically though, to the actual mechanisms that make the

Web work, such as HTTP, “a document is a document”. HTTP is not designed under the

assumption that it will transfer HTML, and servers in most cases do not even look at the

contents of an HTML file—they just transfer it.

That said, a basic understanding of HTML is important, and it just wouldn't seem right not to

provide at least an overview of the language, so I will do that here. I encourage you to seek

out one of the many good HTML resources if you want to learn more: there are dozens of

them on the World Wide Web (where else? ☺)

HTML Elements and Tags

In simplest terms, an HTML document is a plain ASCII text file like an e-mail message or

other text document. The biggest difference between HTML and regular text, however, is

that HTML documents are structured. The document is logically organized into a series of

elements that are arranged according to the rules of the language. Each element defines

one part of the document as a whole. For example, the title of a document, a paragraph, a

table and a hyperlink to another document are all examples of elements.

Each element is described using special text tags that follow a particular syntax. Each tag

begins with the “<” symbol, which is then followed by the (case-insensitive) element name,

and optionally, additional parameters that describe the element. The tag ends with the “>”

symbol. Here's how a tag looks generally:

Some elements are entirely described by the presence of a tag, and in such cases that tag

is the entire element. More often, tags occur in pairs surrounding the actual content of the

element; the start tag begins with the name of the element and the end tag begins with the

name of the element preceded by a slash symbol. For example, the title of a document is an

element defined as follows:

<title>This Is A Great Story</title>

The content of each element can contain other elements, which causes tags to be nested

within each other. For example, if we wanted to highlight the word “Great” in our title by

displaying it in bold letters, we can add the “” tag as follows:

<title>This Is A Great Story</title>

Each whole HTML document is in fact defined as a single “element” called “html”; the whole

document is enclosed in “<html>” and “</html>” tags. Within this element, the document is

divided into two standard sub-elements that must be present in each document: the head

and the body. The head of the document contains information that describes the document

and how it is to be processed; it most commonly contains the title of the document as

described above. The body contains the actual content of the document. These three

elements define the basic HTML document structure:

<html>

<head>

(head elements go here…)

</head>

<body>

(body elements go here…)

</body>

</html>

The bulk of the document consists of the body elements that are placed between the

“<body>” and “</body>” tags. HTML documents can range from very simple bodies

containing only elements such as text paragraphs and perhaps a few links, to very sophisti-

cated documents that are computer-generated and contain hundreds or even thousands of

nested tags of various sorts.

Common HTML Elements

Table 274 provides a brief description of some of the more common elements used in the

body of an HTML message and the tags that define them, to give you a feel for how the

language works:

Table 274: Common Hypertext Markup Language (HTML) Elements (Page 1 of 3)

Element Example Element and Tags Description

Paragraph

Jack and Jill went

up the hill to fetch a

pail of water…

Delineates a paragraph of text. Note that everything

between the start and end tags will be considered

one paragraph, even if split onto multiple lines as I

have done here; line breaks are not significant in

HTML formatting, only tags.

Line Break

George W. Bush

The White House

1600 Pennsylvania Ave., NW

Washington, DC 20500

Forces a line break. Used instead of the paragraph

tag to present lines close together, such as

addresses.