Vrach said:
Wait, the application is actually making you doubt whether it's a human or software? Sorry, missed that part, it's kinda ridiculous to consider it talking to a person as it's very easy to guide Cleverbot into revealing it's software. For a start, it NEVER says anything until you do, it only replies (very obviously calculatingly and on a timer+search time)
Interrupting you here to state that I doubt it's calculating(it may be), but while I'm waiting for a response I can't write nothing, and if we were to consider that 2 humans were chatting then this procedure would assure that no one would interrupt the other.
and never says more than one line at a time, clearly showing it's not matchmaking software to begin with,
Err, actually I didn't notice that... although I doubt I used it enough times to observe if it makes more that one sentence at a time.
unless you actually believe there's someone out there 'hired' to work the site, which actually makes you a lot more similar to a religious person than an atheist.
No way do I believe that there's a hired goon to reply. I only consider that it's like a roulette chat site. Wait a minute! This guy actually speaks my mind:
Tydanubus said:
I've had intelligible conversations with "Cleverbot." I've also had "Celverbot" butcher grammar and spelling and insist that I'm a machine. I've had "Cleverbot" sing Still Alive and exchange Team Fortress 2 memes with me. Other times, "Cleverbot" doesn't seem to know what I'm talking about when I say "Mentlegen" or "This was a triumph." It's almost as if it's just a chat roulette program that pairs you with a different person after exchanging several lines of conversation.
Thank you Mr. Tydanubus for accidentally putting into words what my exact thoughts are. Couldn't have said it better myself.
The answers are also usually easily predicted, not in a way you'd predict a person's response, it's simply a logical way to implement it. If you asked a person the value of pi, they'd also either round it to the usual 2 decimals or tell you to go fuck yourself
The software for it is also not at all impossible to program, or even overly hard. The reason you're not seeing a more practical application is because of the flaws... any practical application of the algorithm(s) would likely require them to have a lot better success rate. But all it really requires is smart algorithm database building. You tell it something, it has an algorithm to validate how likely the information is true (and likely waiting until it's told that enough times to believe it's true, perhaps check the information with other people using it by asking them a question).
While I would find it easy to believe this answer, I find one thing rather hard to rap my mind around. I did a little programming back in my school days(which I've just finished), and I have to admit that it's not impossible or hard as I rushed away to say. It's just really time consuming which I don't find to profitable for such a program if it's results are somewhat minimal. Ok, let's say it is profitable(this is not my major problem).
The biggest problem is that I can't understand how this so called "program" matches the question(s) to a response.
Actually to make things clearer: it's not how it matches the response, but
how it understands the question.
Ok let's say that that the database has the question "What is the value of pi?" correlated to "3.14". Then how come it still manages to respond to my question if I change it?
I mean the program could have more rephrases of the question correlated to the answer "3.14", but I doubt it knows like every rephrasing of every question out there. How come it still manages to give a response when it dose not understand me/the question?
And how come if I ask it the same question over and over it gives me weird responses?
As in, I keep asking the value of pi, and it answers:"3.14", "3.14...", "3.14?", "Briefly 3.14... But I prefer steak and ale pie" and then "No it's 3.14..." and plain simple "No".
I mean yeah it can have variations of the answer, but sometimes it just fucks up at the same question it just answered a few minutes ago?
These are the only thing I have troubles understanding. And if you add paranoia to my disbeliefs, then you can easily understand why I have troubles believing that this is not a hoax.
And unfortunately, because of this paranoia I Doubt I won't find it hard to believe that it's true no matter how much evidence I get.
The fact you're actually doubting it's software is cool though, shows how far the potential can go
It could be cool, the only problem is that I kinda lost faith in the whole human race and stuff and I regard all technological progress as "get rich quick" schemes.
But it it actually is like the real deal, then humanity deserves a brownie.
P.S: Please excuse my lack of comas, I'm really a slacker when it comes to punctuation.
