To start them chatting type in some text in the text field at the top of the interface and click the 'Start Talking' button.
They will keep going until you click the 'Stop Me' button. As they go they get more and more confused - this is one of the issues to be resolved. In particular, the issue of sentence splitting needs to be resolved. If one bot answers with a reply that contains multiple sentence (e.g. I love Star Trek. My favorite series is 'Voyager'.) the other bot will try to supply an answer for both sentences. This very quickly gets out of hand - more and more sentences being passed between the two. Just hit the 'Stop Me' button.
The bots are hosted on the Pandorabots site ( Pandorabots )
You can clear all the fields and all the text by clicking the 'Clear' button.
Most of the dialog is based on the standard Alice set. However, a few AIML elements have been changed - those relating to 'mother', 'science fiction' and 'Star Trek'.
Have a play and let me know what you think.
This is a link will open a new browser window displaying the Flash .swf interface
4 comments:
Hi Michael
Tried it, and yes, they chat effectively... It's progressively more obscure as it goes as you say. (But I have had many conversations like that with my fellow LOST fans, ie the TV series)
In fact - it would be interesting to see two bots discuss the meaning of "LOST". (A show so ambiguous in its audience readings that it's probably hard for either party to ever be "wrong", hehe)
I don't know how hard this is, but presumably, one option is:
Bot A: Setup,
Bot B: Feed, and
Bot A: Payoff.
For eg:
Bot 1: My mother-in-law is SO fat.
Bot 2: How fat is she?
Bot 1: She's so fat, when she sits around the house, she really sits AROUND the house.
(I stole that from a Garfield comic, but you get the idea.)
Anyway, given the way they interact, this seems a very effective EPOC to me...
Also, in MAD magazine years ago I enjoyed the old "switcheroos", about "That's bad", "No that's good!"
An example:
"Did ya hear I got married?"
"Oh, that's good."
"No, that's bad! She's ugly!"
"Oh, that's bad."
"No, that's good! She's rich."
"Oh, that's good!"
"No, that's bad! She won't give me a cent."
"Oh, that's bad."
"No, that's good! She bought me servants - and a big house"
"Oh, that's good."
"No, that's bad! The house burnt down."
"Oh, that's bad."
"No, that's good! She was in it."
Anyway... some food for thought, maybe?
I think the whole "funny man/ straight man" idea is great. - It means the Audience can sit back and "be entertained", rather than have to do the work of asking the questions...
Joe. That's pretty much the plan. I hadn't thought of the 'that's good - that's bad' switcheroo. That's really something to play with.
In order to help limit the conversation "blowing up" - i.e., more and more responses, I would include in my AIML some random tags that include blank responses. And I wouldn't worry if then you got a true blank response (i.e., where one bot says nothing) because I *think* you can have a specific response just for that - and have something randomly restart the conversation.
It also seems like your javascript is not working right... there was javascript embedded in some of the responses.
(Note... I started the conversation off by saying "why did the chicken cross the road?")
-Adeena
Hey Adeena.
Thanks for the comment and the 'heads-up' about the javascript issue. In the future I'll be developing a new AIML set for each character.
Thanks again.
Post a Comment