A network protocol is a standard way of regulating data transmission between computers. Just as diplomats adhere to protocols — rules of behavior — when in foreign lands, network communications do the same. They have to obey agreed rules if they are to communicate and 'get on with each other'. After many years of both public and private research and development, two network protocols are now dominant: TCP (Transaction Control Protocol) and IP (Internet Protocol), together known as TCP/IP. (These were actually unlikely protocols to be so widely accepted, as faster, standardized protocols had been agreed upon, but none had the same robustness and extensibility as TCP/IP.)

Very often protocols were implemented without any formal acceptance and, because they worked most of the time, they became standards by default. Although TCP/IP is an accepted, de facto standard, work on Internet protocols continue in order to improve communication quality and support the continued growth of the Internet. There is no dictating authority for the Internet. Without a controlling authority, interim proposals about protocol changes are made by groups of interested individuals and then opened up for discussion. Documents containing the various proposed standards are published as Requests For Comment documents (RFCs). You may see references to a specific RFC as the best description of a protocol!

An URL is needed to locate any resources on the Web. It is an address format that specifies how and where to find a document. The general format is as follows, where the various items in italics must be substituted with part of a real URL, or omitted altogether.

     http://machine_name:port/path/file_name.file_extension
   

machine_name is either an IP address, for example 137.234.33.89, or a Fully Qualified Domain Name (also known as a DNS name, because Domain Name Servers map between Domain Names and IP addresses), for example, www.apple.com

port is the TCP port to connect to; this is an entry point to software on the server; an optional part of a URL

path is a relative file path from the server's document root; the server will start looking for a file in a specific directory and paths are relative to this

file_name is the name of the file to be browsed, e.g. welcome

file_extension is one of a number of suffixes which, by convention and operating system setup, indicate the type of data contained within the file, e.g. htm,html, txt.

This language provides the format for specifying simple logical structure and links in a hypertext document. As a markup language, special formatting commands are placed in the text describing how the final version should appear. These formatted documents are interpreted by a Web browser which uses the HTML code to format the page being displayed. Several units in this module deal with HTML extensively. Although most professionals use special authoring tools to write HTML documents and to manage sites, developers of e-commerce sites and applications need to know the nitty-gritty detail of HTML, and this is what you will study.

HTTP is a network protocol used to retrieve documents from a variety of machines in a minimum of time. It was invented by Tim Berners-Lee to support a project in developing a distributed hypertext system. Distributed hypertext requires the retrieval of documents from many different machines. File Transfer Protocol (FTP), which predates the Web, would be too slow for this purpose as it is a connection-oriented protocol that requires a permanent connection to a server, thus requiring a connection-maintenance overhead when accessing different machines.

Therefore, to support browsing, HTTP has the following characteristics:

HTTP is not mandatory for distributed hypertext; there are other techniques and protocols that can be used to access or transfer information. However, like TCP/IP and HTML it is ubiquitous, and so enables investment to develop e-commerce.

The Web began as a tool to share knowledge and has successfully evolved into a general communications mechanism. With the support of transactions and synchronous communications, the Web has application in many different fields.

A primary use is the dissemination of knowledge, which takes many forms. For example, chat rooms and bulletin boards are integral to interactive discussion of all kinds of subjects. Frequently Asked Questions (FAQs), published on Web sites, offer answers to users' questions on how to do certain kinds of tasks. The variety of information that can be pulled out of the Web is wide-ranging.

Education includes a variation of the dissemination of knowledge. Open- or distance-learning programmes spearhead this aspect of the Web. Basically, any kind of demonstration on how to carry out certain tasks can be considered education. For example, a user can learn how to create a Web page from the numerous websites publishing such instructions.

With the possibilities of online trading, business transactions are carried out on the Web. The user supplies their order and credit card details so as to buy products advertised on the Web. The Selling module would cover this subject area in depth.

The Web as a vast digital library is becoming what is known as a 'Global Information Structure'. It will have a profound effect on how we live, work and play. We shall now look into a few of the social implications of the Web as a digital library and a marketplace.

The hypermedia concept includes not only text and illustrations, but also music, animation, digital movies, video games and computer software. This diversity changes the form of literacy required when using the Web. The literacy needed when listening to music and watching a movie is different from that used when reading a book. Less literacy may be required with innovative ways of using this digital library. For example, software that reads text aloud can assist people with visual handicaps.

Editors and publishers employing traditional methods of publishing have little to gain from this type of publishing. As digital works can be copied at low costs, stored in almost no space and transported instantly anywhere in the world, writers can be their own publishers. Therefore, the works published are of indeterminate quality and value. Web publishing may provide no evaluation of work published.

An article may perhaps interest a group of specialists in that field. With the Web, an average reader may browse through the article according to their degree of interest in the field. He or she may not want to be burdened with an additional flood of technicalities, or perhaps would navigate further to extract more in-depth information to supplement a deeper interest in the field.

The ease of copying digital works causes difficulties in protecting copyrights. It may be tempting to make illegal copies rather than finding the rightful owners and paying them a fee. On the other hand, the non-issue of distance and the 24-hour, 365-day activity on the Web means that much can be easily bought through on-line shops. Consumers may come from distant areas or different time zones. With the Web, this market place is open at all times and can serve a very large global region. New technology even allows computational agents to staff the market place rather than people. Therefore, businesses are not constrained by distance or time.

Despite the irrelevance of distance, an electronic marketplace may be attractive as it goes to the consumers instead of them physically moving to the business environment. Its sense of place is created as an illusion for the benefit of the consumers.

We shall proceed with an analysis of hypertext documents.

Links can be created to associate related subjects. Therefore the information given can be extensive and wide. The combination of two related subject areas is known as knowledge additivity.

Let's say you want to find out how to tailor a shirt using a sewing machine. You would probably look in a book on tailoring a shirt and another on using a sewing machine. The information read would then be linked together in your brain. However, with the hypertext concept, this knowledge additivity would be simpler with association links. You can just continue clicking to read on both subject areas within the perceived single document.

There are difficulties with working with hypertext, and hence also in providing or obtaining goods or services.

When you are surfing the Web, you are using a Web browser. When you go to a website for documents, the site delivers them using software called the Web server. The browser is considered to be a client in the relationship with the server as it is requesting information services from the server. This is just one particular example of the client-server model of computing.

Information is passed from the server to the browser. This information may be in the form of HTML documents, GIF files, Excel spreadsheets, movies — just about any digital content.

Information can also be passed from the browser to the server. When you click on a hyperlink you are sending information to the server, and when you fill in an online form, you are usually sending information to the server.

In addition to passing information backwards and forwards, some processing can also be done in the browser. For instance, you might have a simple Web page that calculates the overall cost of a loan once the initial value of the loan, the interest rate and the length of the loan have been entered.

But where does the processing take place? Does the server process the information and generate the result, or is it the client that processes the information? If the client does the processing, then this is a client-side application; if it is the server, it is a server-side application.

In the loan example above, the client has the information (the principle, rate and time). It could send this information to the server to process the information, generate the result and send it back to the client. Alternatively, the server could send a programme to the client that will carry out the processing. In the latter case, since the client has all the information and the programme is pretty small, it is probably better to run the application on the client side.

Of course, there is also a problem of who has the information. If the server has a database, and the client wants to query it, then there are two possibilities. The server could send the database and the querying programme to the client to process it or the server could process it and simply send the result. In this case, it would probably be better to do the processing on the server side.

To summarize, where the processing is undertaken largely depends on where the information is, but also depends on the processing loads of the machines as well as the size of the programme being run.

A browser receives binary data from the server which it has to cope with. How does it know if the binary data is an HTML document, a GIF picture file or something entirely different? Even if it does know what kind of document it is, how does it process it? The answer to this is MIME types.

MIME types — Multipurpose Internet Mail Extensions — were created to identify the differing types of possible email attachments. The MIME types have been extended to include new multimedia types as they have been introduced, and are now used with a variety of protocols including HTTP. When information is sent to a browser, a MIME header identifies the file type of the document. Attaching a MIME type to a file allows the browser to process the file's contents correctly without the browser having to guess at the data type from the file's extension. This is important, since while MS-DOS files require a three letter extension to identify a file type (and Windows XP uses a similar file extension system), not all operating systems do this.

This MIME header information has the following format:

Content-type: type/subtype

where

type is one of several general types such as: text, audio, image, video, application etc.

subtype is a more specific designation. This is a large and ever expanding category.

Some examples of MIME headers are

 text/HTML
 video/MPEG
 image/GIF  
 

Mime types are processed as follows.