Submitted by Alex Scott on Fri, 10/11/2006 - 13:39.
Hi there - just popping my head in to see how you two are getting along...
Firstly - does this chatbot have some kind of 'intended' function? If we gave it a purpouse, and then gave it an internal logic that would frustrate it's ability to fulfill it's purpouse, we could have quite an interesting character... one with a sense of internal conflict...
The obvious example would be that the 'bot' is meant to serve as a user-friendly guide to the polyopticon website, but for some reason she becomes disgruntled easily, or confused, and then the 'game' for the user becomes: 'how to get the information out of her?' with each failed attempt more amusing than the last...
This would (of course) rely on the bot having some kind of a 'memory' of their conversation with a particular user. Perhaps once certain internal 'gateways' have been cleared by the user, they automatically move on to a different module or library of her personality. I am interested in concepts of internal archetecture as far as the character is concerned.
I very much like the idea of an AI with personality issues (and it is a classic subject of Sci-Fi after all) - the question is: how to make these issues 'human' enough to be entertaining... instead of simply a machine producing a torrent of random abuse.
Talking about the visual aspect to the bot, it would be great if the face altered and changed to reflect mood or subject matter. Perhaps the face is a composite of other faces - and depending on her mood, the bot may actually manifest herself as different people, or combinations of people... just a thought...
Is any of this useful? Or am I way off the mark with this stuff?
First Impressions...
Hi there - just popping my head in to see how you two are getting along...
Firstly - does this chatbot have some kind of 'intended' function? If we gave it a purpouse, and then gave it an internal logic that would frustrate it's ability to fulfill it's purpouse, we could have quite an interesting character... one with a sense of internal conflict...
The obvious example would be that the 'bot' is meant to serve as a user-friendly guide to the polyopticon website, but for some reason she becomes disgruntled easily, or confused, and then the 'game' for the user becomes: 'how to get the information out of her?' with each failed attempt more amusing than the last...
This would (of course) rely on the bot having some kind of a 'memory' of their conversation with a particular user. Perhaps once certain internal 'gateways' have been cleared by the user, they automatically move on to a different module or library of her personality. I am interested in concepts of internal archetecture as far as the character is concerned.
I very much like the idea of an AI with personality issues (and it is a classic subject of Sci-Fi after all) - the question is: how to make these issues 'human' enough to be entertaining... instead of simply a machine producing a torrent of random abuse.
Talking about the visual aspect to the bot, it would be great if the face altered and changed to reflect mood or subject matter. Perhaps the face is a composite of other faces - and depending on her mood, the bot may actually manifest herself as different people, or combinations of people... just a thought...
Is any of this useful? Or am I way off the mark with this stuff?
Alex Scott
www.growingbrain.com.au