The illusions of Bots and COISes

Bots (short for robots) are known by various other names, such as crawlers and wanderers. They consist of software programs which can insert and extract information from Web pages.

Normally, bots are associated with server side computers where the continuous on-line presence of the computer appears to give the bots an indefinite continuous life. Server side bots are mostly concerned with cataloguing and indexing and because they are travelling across mostly unknown or unprepared Web pages have very limited roles and capabilities.

Far ranging server side bots are also used for information retrieval (IR). These can involve highly complex AI programming but in essence operate simply by recognising combinations of key phrases or words. They are very limited in their successful application because ideally they require standard formats and specified structures of Web pages (i.e. an environment which is virtually an Intranet).

Client side bots are best suited to visiting prepared pages within an Intranet system. In such a system, they can be designed (together with the Web page contents) to be far more usefyul and versatile.

The simplest type of bot is the info bot. These can be as simply as an email which carries an attachment. Their power is in the way they can be organised to carry selective information under the control of an intelligent sytem.

Combined with an object oriented data base and an intelligent system that can build information on the fly, info bots can create unbelivable systems of communication.

This Web site object makes use of info bot systems and the conceptual device known as an auto responder. Auto responders record the address of an incoming message and return data back to that address according to the content of the message. This technology is only in its infancy, but, has the potential to revolutionize the way the Web works.

To visualize the way in which info bots and auto responders could be used, you only have to consider the way in which systems of Lingo objects can be used to construct highly complex self organising structures. It make the head reel just to think about it.

Bot Movements

Although these bots do not actually leave the computer which generates them, they can be visualized as roving around the Web because information is taken to or brought from Web pages in sequence and thus it can seem as if the bots are moving between the pages involved.

This paradigm is so strong that many people actually believe bots and wanderers have a reality which allows them to exist on the Internet independent of the computers which create them.

Such a vision illustrates the power of client oriented Intranet browsers to create fantasy and illusion to enhance the client's experience of Web delivered information. Take for example, the following scenario:

Bot Party

Participants are given their own "intelligent" bot which is equipped with "the ability to feel emotions". The bot will "ask" the owner a series of questions which will allow the bot to adopt the character and personality of its owner (effectively becoming a clone of the owner).

The owner opens a connection to the Internet and allows his or her bot to go off into the World Wide Web to look for cyberSpace bars where bots hang out. The owner then disconnects from the Internet while his or her bot goes around making friends with other bots (many other bot owners are doing the same thing).

After about half an hour or so, the owner connects up to the Internet again for the bot to come back in. It comes back in with a crowd of bot friends, which it has met at the bot meeting places and invited over to its owner's "house' for a bot party. The connection to the Internet is closed and the party begins.

At this party, the owner's bot will talk with all the other bots and secretly tell the owner how much it likes or dislikes each of them. Now you have to remember that the owner's bot is a cloned personality of its owner and each of the other bots are clones of their owners, so, if everybody has told the truth to their bots, the bots ought to respond in a similar way to each other as their owners might if they were meeting at a party in the real world.

All the bots, which the owner's bot likes best, are invited to stay on. Another connection to the Internet is opened and the owner's bot takes away all the bots it doesn't like, leaving behind the ones it does like. The connection to the Internet is closed again while the owner's bot goes off to look for more bots to bring back home.

A little later the owner opens up the door to the Internet and lets the bot back in again with another crowd of bots. The Internet door is closed and another party is started up with the new bots partying together with the nice bots that have stayed on from the earlier party. This is repeated many times with bots continually coming and going with each party consisting of bots which are becoming increasingly more to the owner bot's liking.

At the end of several parties, the bots which the owner's bot has liked best are given messages to give to their owners which asks them to get in touch with the host bot's owner. When these "nice" bots have been given their messages to take home to their owners, the Internet door is opened once again and all the bots go home.

While the owner's bot is out on the Internet, it will be invited into the "houses" of other bots for parties at their houses. If the owner's bot is "liked" by the host bot at one of these parties, it may well come back with an email address of a compatible owner itself.

This scenario can be played out quite realistically in an Internet environment. Graphics can portray a bot image which goes out through the door of a house drawn onto a client's screen when a connection is made to the Internet. This is easily arranged in a handler, which will organize the correct sequence of animations immediately an Internet connection is opened - a door opening and a sprite figuring disappearing into an open doorway.

Similarly, another connection to the Internet could trigger another sequence of animation which portrays sprite figures (bots) coming in through a door and then moving around to appear to be socializing with each other at a party.

Cloning the owner's personality onto a bot is easy. The bot asks the owner some questions by means of a questionnaire on screen with radio buttons to facilitate answering. The answers which the user gives to these questions can then be stored as a string in a field. The bot personality then exists as a string of logical true or false answers to the questions the user has been asked. Such a record can easily be stored and transferred as a string of text in a text field.

Of course, nothing goes in or out of the "house" or the client's computer. The only thing that happens is that messages go backwards and forwards between a prepared Intranet site when connections are opened. This will result in the return of text pages full of text strings describing the characters of various bots whose details have been taken from other bots who have previously connected to the Web site.

Returned files of bot data from the Intranet server are then inserted into the properties of objects created on the client side. (It is like creating people for a party and sending out to get the information to program their personalities).

With personalities stored as strings, it is a trivial matter to operate on sets of strings to compare various combinations of items and through suitable weightings draw conclusions. Thus it can be realistically arranged for bots to assess each other in a simulation which is not too unlike what happens in the real world.

The question as to whether or not any conclusions drawn would have any real significance is not an issue here. The point is that this simple exercise can give a realistic illusion of bots going out onto the Internet and bringing back friends for a party. It has an eerie reality - especially when emails are sent which brings participators (bot owners) together in the real world.

It is this ability of client oriented Intranet systems to produce fantasy and illusion which is the great promise of the mix of multimedia with the Internet.

Imagine bots, which go out each morning to get the day's news for you. Bots which go out once a week to see if any of your applications need updating. Bots which will go for initial job interviews on your behalf. Bots which go shopping for you and having learned of your tastes bringing back just the things you want together with a few surprise items which the bot thinks you might like.

All these things are possible using nothing more than a bit of HTML and javascript coding - the DNA of the Internet.

Speculations on future directions

There are so many possible changes on the horizon, connected with multimedia and the Internet, that it would be ridiculous to take any specific view right now. However, the alternative is to take no view at all...

Fortunately, object oriented thinking allows us to proceed without having to firm up on specifics. We don't even have to work with a final master plan. We can begin with a few modules and evolve into the future, adapting and modifying as we go. However, every object oriented design has to have a goal and a few underlying concepts.

Server or client oriented?

The most critical decision at this stage is whether to be server or client oriented. As explained in another section of this Web site, my personal preference is to be client oriented in some kind of Intranet system (COIS). This is probably very much a minority way of approaching the Web.

However, if you look at the Netscape Navigator you will realize that this is a client oriented Intranet system (COIS) - the users of Netscape Navigator browsers are confined to the Web sites which Navigator can read and interpret. Web sites which want to be included in Navigator's Intranet system have to design their site to optimally use Navigator's features (even though Navigator is using a universal standard).

As far as I can see, most of the large players in the computer and communication games recognize that the Web is the ultimate challenge. Whoever controls the Web controls the world (Remember in Dune? "Whoever controls the spice..."). Strange that they are all surrendering control to Netscape.

[Index]
[Back - Object oriented view of the Internet]

Return to main website