The Conventional View of the Web
Creating a comfortable mental model of the Internet
As a client, when you are connected to the Internet you are linked by telephone
to a local host computer. This host computer will be permanently linked
to a world-wide network of other host computers which are continually exchanging
electronic data with each other. Through this local host you can tap into
and make use of the whole Internet network to send data and messages to,
or receive data and messages from, any other computer connected to the system.
Physically, this system is simple to understand: the Internet is no more
than a network of interconnected computers which can send electronic data
to each other. However, this simple model masks the vast complexity which
can exist in such a system.
With a computing facility at each node of this network, the system is capable
of acting in ways which can go far beyond any previous experience the human
race has ever before encountered with information transfer and processing.
Scientists are familiar with this kind of complex situation and deal with
the problem by creating an abstraction or metaphor which can model the system
in a simple but useful way. This simplified model is then used to facilitate
understanding, exploration, testing and prediction. Clearly, such an approach
is appropriate for discussions about the Internet.
Computer programmers, skilled in the art of object-oriented programming
have an edge in being able to conceptualize the Internet. They are familiar
with the bizarre practice of visualizing a three dimensional empty space
and populating that space with objects which can communicate with each other.
Object-oriented programmers are quite happy about the fact that these objects
have shape or form only in their own imaginations.
It is this trick of being able to create pictures and models in the mind
which is at the heart of being able to understand what the Internet is all
about and being in a position to fully realize and exploit its vast potential.
For a non programmer, a specific tangible metaphor is needed to give substance
to the Internet. This metaphor must be stripped of all surplus or irrelevant
technical detail, yet, still be able to accurately model all the essential
features: allowing thoughts to be concentrated upon the dynamics of the
complete system without getting bogged down with unnecessary side issues.
Such a suitable metaphor might take the form of the space we are all familiar
with: the area outside of the earth's surface populated with satellites
and space platforms.
Let us imagine the Internet to be happening in this space and that the linked
computers are housed in satellites which are continually beaming data to
each other as they suspend in space. Let us also imagine that we each can
interact with this network directly through our own personal computers.
If you can hold this vision, of millions of people around the world being
able to communicate with each other, via a host of inter-communicating satellites
you have a clear vision of the structure of the Internet in general and
the e-mail service in particular.
At every interface and node of this system there is a computer. This means
that data passing through or around the system is constantly exposed to
computers and has ample opportunity to interact with them. The significance
of this cannot be underestimated: message data can be used to alter or modify
the system, or, the message data can be acted upon and processed at any
time during routing, transmission or reception.
Connecting to computers to obtain archived information brings to mind technical
researchers searching through arcane university data bases; statisticians
seeking data from published government figures; computer enthusiasts downloading
shareware; teenagers looking for the latest computers games.
If you step back from the reality that these Web sites are simply computer
files on computers located in back offices somewhere and put them into the
context of a space/satellite metaphor they take on a more interesting perspective.
The conceptual trick now is to see the virtual space of the Internet as
a space containing only space platforms objects which can contain any product
of your imagination without there being a computer in sight. To visit this
magical world you simply beam up to one of the platforms via your computer.
The removal of layers of complexity is a perfectly valid way to simplify
a conceptual model because removing the physical infrastructure has no effect
on the outcome of Internet interactions.
It has the great advantage though, that you can then ignore logging on procedures,
message routing and all the techy stuff involved in Internet communication.
All you have to do is imagine that you can go directly to any Web site space
platform which is present in the cyberspace of the Internet. You and 40
million others.
With this simplified model, it is much easier to imagine visiting a Web
site: you just beam up to any platform which catches your attention. It
is also easier to imagine many people beaming up to the Web sites at the
same time as yourself, perhaps even visiting the same site as you. It is
now easy to imagine meeting people at these Web sites and moving around
from one site to another. This simple conceptual trick allows you to bring
the whole concept of Web sites comfortably into your mind - just don't think
about them as being computers.
As you imagine yourself, and 40 million others, staring up at these space
platforms, it becomes obvious that these space platform Web sites could
have a host of corporate uses, particularly for businesses who wish to send
information to a wide range of different audiences. These Web sites are
permanently on display, instantly accessible to anyone from anywhere in
the world. They are also simple to modify and update. They can be used for
such purposes as informing shareholders, employees, agents, clients or customers
about specific current events, such as prices, deliveries, new developments
or any other constantly changing information about a company or its products.
A space platform Web site enables a corporation to portray a permanent public
profile of itself to a world-wide audience. A Web site can become a store
front for products or services.
Users of multimedia authoring packages such as Director are going to be
intimately involved with this virtual world because they are going to be
the builders. It is with the use of highly sophisticated applications like
Director that the Internet is going to expand into new directions which
currently are only in the realms of science fiction. To see how such creations
might be constructed it is necessary, as builders, to take a look at some
of the foundations.
The foundations of the World Wide Web
The basic concept of the World Wide Web emerged from an academic project
initiated at CERN, the high-energy physics research center in Switzerland.
The World Wide Web (WWW) is not a physical structure. It is simply a set
of protocols which allow different makes of computer to communicate with
each other in a common language.
The protocols of the WWW are completely text based and allow certain symbols,
words and phrases to have the same meaning to whatever make of computer
might be reading them. In a stroke, this allows everybody, everywhere, to
be included in one single unifying network irrespective of the physical
hardware. This is why Web sites are so called - because they use the World-Wide
Web protocols.
The WWW protocol is typified by the hypertext Markup Language (HTML) which
sends out text as strings of ASCII characters which are accompanied by tag
symbols which tell computers how to display the text on screen: complete
with instructions for spacings, indents, ruled lines, type sizes, etc. This
markup language also extends to include graphics and forms. In this way,
it doesn't matter which type of computer is supplying text or graphics,
or which type of computer is reading the documents, everything is always
presented in a way reasonably similar to the way that the designer planned
for it to be seen.
As useful as this facility is, the CERN project was concerned with far more
than the relatively trivial task of making documents look the same on screen.
Its main objective was to provide a facility which would allow documents
to have hypertext links to one another.
A hypertext link is a means whereby certain words in a document can be clicked
upon and the user is transported from one document to another. For instance,
you may be reading a document about the Arctic and come across the hypertext
expression "polar bear" (identified by being in a different colored
type). You can click upon this hypertext phrase in the text and be immediately
transferred to an article about polar bears.
The WWW protocols are able to facilitate this switch to another document
because the real power of WWW lies in its truly versatile ability to allow
computers to send all kinds of programming instructions to each other. The
tag symbols used in HTML not only provide concealed text formatting instructions,
they also facilitate more general programming and navigational directives.
This hypertext facility is so powerful that not only can it take you to
another document, it can take you to another document even if that document
resides in the data base of another computer. In other words a hypertext
link can link documents across computer boundaries. If these computers are
also linked into the Internet, the hypertext links can transport you to
different documents residing in a vast assortment of different computers
located in countries all around the world.
The significance of this seemingly innocuous little trick takes a long time
to sink in - even to many experienced Web site designers. On the surface
it may appear to be no more than another useful convenience - that is until
you switch paradigms and think about this document switching in terms of
the web site platforms in the virtual space metaphor.
Imagine, you've beamed up to (dialed up) an information carrying space platform
Web site and you are reading a document on a Web page and you come across
a hypertext link. Clicking on the link takes you to a document in the files
of another computer.
From the point of view of the space platform metaphor, clicking on the hypertext
link moves you through space to another information space platform Web site.
From this new space platform Web site you may encounter another hypertext
link to a third space platform Web site. In this way you can imagine yourself
traveling around in space, hopping from one information space platform Web
site to another as you click away at the various hypertext links you encounter.
Physically you are transferring from one computer to another; mentally you
are moving between space platform Web sites. This is the kind of mental
transform you have to be able to make to understand the power of the Internet.
Moving around in the virtual space of the Internet
At this point, the virtual world of the Internet is a dull shell and not
very inspiring. Let us light the touch paper of inspiration by imagining
that we own one of these information space platform Web sites and are free
to do what we want with the files it contains:
1) Divide all the information files in this Web site into four groups and
add to the title or headers of these documents a group name: port; starboard;
forward; aft.
2) Divide each of these four groups of files into two and add to one half
the subtitle "Upper_deck" and to the other half the subtitle "lower_deck".
This simple expedient has given the Web site a virtual three dimensional
form where documents are held in virtually different parts of the Web site
space platform: upper or lower levels, at the front or back, to the left
or right.
3) Now, create a new document and entitle it "Starboard.lower_deck.games_room"
and from this document attach a hypertext link to a computer game.
4) Imagine now that we can contact the owners of some of the other Web site
platforms which are present in the Internet virtual space:
5) We arrange with each of these owners to install a hypertext link called
"The Good Ship Lollipop" in a document (a Web page) on their Web
site platform.
6) We then program these links from these other Web sites to link to a document
(Web page) on our Web site which holds a picture and a plan of a ship which
clearly marks the various areas our ship is divided into .
7) On the picture of the ship we place a hypertext word called "games-room"
which links to the document we created which links to the computer game.
Now, from the perspective of the virtual world we have created, visitors
to any of the other Web site platforms in cyberspace might come across this
hypertext reference to "The Good Ship Lollipop" (maybe in an index
entitled "Cool places to visit").
Clicking on this text takes them to a ship in which they can travel around
to find a games room where they can play a computer game.
In this way the cold dull world of computers can be transformed into a virtual
world of exciting and interesting places to visit. This is what the Internet
is all about. This is also the world of OOPS where objects are created firstly
in the imagination and then given substance with computer code.
Interactive Web sites
Through clever and creative text and graphics, HTML documents can give the
illusion that a Web site is a virtual representation of any physical structure
in real life.
Clicking upon pictures rather than text, a Net surfer can be beguiled into
believing that he or she is moving around a model of the real world - a
world of information and entertainment. Links between Web sites then become
bridges and roads, the interlinked virtual structures become a three dimensional
city in cyberspace.
As a passive network, where you just move around and click upon things,
this vision of the cyberspace city has very little substance and is of limited
interest. What brings it to life is the involvement of the Web site computers
with the files and documents they are displaying.
They need not be passive data delivery systems, they can be made to play
an interactive part in a Web surfer's experience.
Interaction, between Web servers (the name given to the Web site computers)
and the users is arranged through another set of protocols called the Common
Gateway Interface (CGI). Like the HTML convention, CGI allows one computer
to communicate with another, but, whereas HTML allows the server to control
the display at the user end, the CGI protocol allows actions at the users
end to control programs at the server end.
Practically all the interesting Web sites rely on CGI controlled scripts
to provide their most interesting features. These CGI scripts can allow
users to place orders and pay money for goods and services online. They
can facilitate communication links between visitors to a site to enable
online discussion groups. They can facilitate online "telerobotics"
where users can maneuver cameras, machinery by remote control.
CGI scripts allow sites to record the number of visitors, track their progress
as they explore the site, have visitors sign visiting books, provide identity
references or passwords. The scripts provide facilities for voting or survey
schemes, allow visitors to fill in forms.
Most importantly of all, these CGI scripts can alter, modify, change or
evolve the Web images presented to the user in direct response to user (client)
feed back. In terms of information the user can be shown only the information
which is applicable or useful. Information can be streamed to a user according
to the user's responses.
Marketers are familiar with the practice of sorting mailing lists into demographical,
interest, location and income groupings and sending them suitable different
sales material. With CGI scripts on a Web site, this can be done on the
fly, with forms and information composed in real time so that a customer
or client is given just the correct sales pitch.
In a similar way companies can set up information services and trouble shooting
booths. They can answer questions from shareholders or deal with customer
and employee queries. Feedback control using CGI scripts can be used in
all manner of imaginative ways for both internal and external applications.
Feed back forms have all manner of different uses, allowing Web site visitors
to give opinions, identify problems or just ask questions. Another variation
is the feedback form which can be linked to order processing documentation,
order verification and even the preparation of delivery and billing procedures.
Most interestingly of all, information can be arranged to evolve. It can
be programmed to evolve according to a single person's responses or according
to an accumulation of different people's responses.
CGI scripts also facilitate clickable image maps. Not only can these provide
navigation prompts they can also provide image or textless selection of
information. These are being used with increased frequency as designers
try to get away from the stereotyped document image of their Web pages.
Search engines are another popular facility enabled by CGI scripts. These
allow Web site visitors to search data bases without having to scroll through
pages of indexes.
Search engines can also be linked so that one search engine of one database
can find a profitable area of search in another data base and transfer the
searching to the search engine of the new data base.
CGI scripts are used extensively for games, entertainment and social activities
on the Net. They can set up quizzes, treasure hunts and lotteries. They
can be used in interactive adventures games where player's moves are recorded
and the status or scenery is changed according to the player's progress.
Through the CGI scripts the players can interact or communicate with each
other, even throw virtual parties and invite people from all over the world
to attend.
Business conferences, meetings can be arranged. Forums and discussion groups
of all kinds can be set up. Virtual parties can be held in exotic locations
all over the world. This is why the Internet has become so popular because
it can create virtual communities where people can exchange ideas, give
each other help and information and simply be friends and feel a sense of
belonging.
User interactions need not be limited to discussion groups. Users can interact
through community databases where people submit information to a large data
base. Through combined feed back, the database grows and evolves to become
more useful and informative. An example is the Cardiff movie data base.
Ratings assigned to movies by visitors to this Web site are used to compute
an overall rating for each film.
An even more ambitious interactive database, called Ringo, is set up on
a Web site by the MIT Media Lab. This site asks you to list your music preferences
which the program adds to a main database. Your preferences are then compared
to preference patterns of all previous users of the data base and an attempt
made to suggest other pieces of music which you might enjoy. Such a technique
may have many applications in the world of commerce.
CGI scripts can make the World Wide Web come alive. Without them, the web
is nothing more than files and graphics.
This is the conventional view of the World Wide Web. It is the view from
the server side and uses the same paradigms as traditional media. It is
the world of Netscape and the giant all purpose browsers with their plug
ins and supportive programming languages.
Another Way of Looking at the Web
As attractive as this conventional view of the Web seems to be, it is looking
at the Internet and the World Wide Web from a single limited perspective:
from the server side of the Web.
The world of OOPS is capable of reorganizing this comfortable view of the
Web into quite different communication systems, the like of which has never
before been visualized or created - simply by bringing into play the vast
potential of the client side computers and looking at the structure from
different paradigms.
This is the world of OOPS, of COISes and Intranets. It is the world of interplay
between intelligent agents, CD-ROMs, computers and the Internet. It is the
world of info bots, Web site objects and object oriented landscapes. It
is a world where the imagination can be given full reign.
[Index]
[Next - Intranet - An OO definition]
[Back - Object-oriented programming]
Peter Small August 1996
Email: peter@genps.demon.co.uk
Version 1.00
© Copyright 1996 Peter Small
No reproduction in whole or part without prior permission