@keeton 52 & Patrick2011
Not to take a side in this disagreement over whether Cleverbot is just a site where you talk with others, constantly switching conversation partners, but I tried this with Cleverbot:
User: Hello. Look, I'm just testing a theory, according to which there is actually no cleverbot and 'http://www.cleverbot.com' is a site where people actually talk with each other for one post or two, and then have their conversation partners switched. How do you know I'm not Cleverbot? I can do a speling Mistake as well as a capitalisation and a grammar one ('speling', 'Mistake', 'do' instead of 'make', which a computer programme would not do, right? For this theory to be correct, also, this should look completely irrelevant to you, as it most probably is me taking part in an already begun 'conversation'. So, please, tell me, are you human, as I am?
Cleverbot: It probably should.
User: Hello. Look, I'm just testing a theory, according to which there is actually no cleverbot and 'http://www.cleverbot.com' is a site where people actually talk with each other for one post or two, and then have their conversation partners switched. How do you know I'm not Cleverbot? I can do a speling Mistake as well as a capitalisation and a grammar one ('speling', 'Mistake', 'do' instead of 'make', which a computer programme would not do, right? For this theory to be correct, also, this should look completely irrelevant to you, as it most probably is me taking part in an already begun 'conversation'. So, please, tell me, are you human, as I am?
Cleverbot: Why should I?
User: Hello. Look, I'm just testing a theory, according to which there is actually no cleverbot and 'http://www.cleverbot.com' is a site where people actually talk with each other for one post or two, and then have their conversation partners switched. How do you know I'm not Cleverbot? I can do a speling Mistake as well as a capitalisation and a grammar one ('speling', 'Mistake', 'do' instead of 'make', which a computer programme would not do, right? For this theory to be correct, also, this should look completely irrelevant to you, as it most probably is me taking part in an already begun 'conversation'. So, please, tell me, are you human, as I am?
Cleverbot: And that would be?
User: Hello. Look, I'm just testing a theory, according to which there is actually no cleverbot and 'http://www.cleverbot.com' is a site where people actually talk with each other for one post or two, and then have their conversation partners switched. How do you know I'm not Cleverbot? I can do a speling Mistake as well as a capitalisation and a grammar one ('speling', 'Mistake', 'do' instead of 'make', which a computer programme would not do, right? For this theory to be correct, also, this should look completely irrelevant to you, as it most probably is me taking part in an already begun 'conversation'. So, please, tell me, are you human, as I am?
Cleverbot: That also means nothing.
User: Hello. Look, I'm just testing a theory, according to which there is actually no cleverbot and 'http://www.cleverbot.com' is a site where people actually talk with each other for one post or two, and then have their conversation partners switched. How do you know I'm not Cleverbot? I can do a speling Mistake as well as a capitalisation and a grammar one ('speling', 'Mistake', 'do' instead of 'make', which a computer programme would not do, right? For this theory to be correct, also, this should look completely irrelevant to you, as it most probably is me taking part in an already begun 'conversation'. So, please, tell me, are you human, as I am?
Cleverbot: And that would be?
Even though it could be argued that "It probably should.", the first answer, could have been given by a human, it was 'typed' and posted in less than a second. Accordingly, the other answers did not take any more than a second and a half or two to be posted. I know, I know that, much more than often, the answers and the turns and twists in 'conversations' look a lot like what Cleverbot would be according to keeton's theory, but I have to say that, even in that case, there is no consistency; some 'changes' appear to take place in only one post, while other 'conversations' last for over five or six posts. Not to mention, also, that human - like answers support the idea of Cleverbot being a bot, too, considering it is suppposed to 'assimilate' humans' answers. In contrast to that, it could be asked "What's with the mistakes it occasionally makes?", which I can say I cannot answer.
Personally, I want to believe it actually is a "clever bot", and I do think so, although with the shadow of a doubt. Actually, If you look at a conversation with Cleverbot as a whole, most of it, or all of it, can easily be explained by keeton's theory, but single parts on their own show otherwise.
Anyway, although all that does not conclusively prove either theory, I would say it makes keeton's one look sufficiently implausible.
