8PLA • NET - Jun 16, 2013:

That is a really nice story, Denis Robert. 
Thanks for sharing your project experience
with building Johnny.  I hope to hear more
about that some time.

There is some détails below (sorry, it is only a Google translation, I don’t have better) :

Who is Mylène

The brain of Mylène

Mylène is the french version of Johnny.

Steve Worswick - Jun 16, 2013:

Elizabot should have surely got some points for this:

]14) I went fishing yesterday- 0|0|0
Very interesting.

Especially as another entrant got 2 for this:

7a) what country are you from? - 0|0|2
try talking about this with more detail my child, or i assure you your soul will go to hell.

If I were the judge, I wouldn’t have given points for either as the question is ‘ignored’.

Steve Worswick - Jun 16, 2013:

Be careful though guys, as I believe that we are not supposed to discuss “Wendell’s” competitions since his ICC earlier this year which he won.

If you’re really lucky, you’ll get an angry email.

8PLA • NET - Jun 16, 2013:

2) what language is predominantly spoken in Germany?
Were you just saying, “SPOKEN GERMANY what language PREDOMINANTLY”?

In your opinion, why did this question not get a point, Mike?

Keep in mind that SPOKEN is an adjective, NOT a verb.

Therefore “SPOKEN GERMANY” is adjective noun

“what” is a pronoun, adjective or adverb
“language” is a noun

Therefore “what language” is adjective noun

Finally “PREDOMINANTLY” is an adverb

So we have an ‘adjective noun adjective noun adverb’ term which is valid grammar
according to Quizlet.

It is only my opinion….... On the points side of things, regardless of the grammar analysis the bot may have done, it is not a response to the question. If you don’t know what the bot is doing, it just seems to have returned the words ask, but in a different order. I don’t think a judge can give points there..

On the grammar side (a subject I know more about than I do CBC scoring), I think there is some confusion with what is a verb and the ‘adjective noun’, as far as what response has been given.

‘what’ is certainly not an adjective or adverb here.

‘spoken’ here is absolutely a verb, not an adjective.

MikeA - Jun 17, 2013:

If you’re really lucky, you’ll get an angry email.

I am used to these from Chris but he usually sends them anonymously though. I notice the Flash videos are identical to the ICC as well now. Heck, even the play button is the same but a different colour! Surely, he still isn’t saying that he had nothing to do with the ICC?

I’d best be quiet now, sorry Dave.

Steve Worswick - Jun 17, 2013:

MikeA - Jun 17, 2013:

If you’re really lucky, you’ll get an angry email.

I am used to these from Chris but he usually sends them anonymously though. I notice the Flash videos are identical to the ICC as well now. Heck, even the play button is the same but a different colour! Surely, he still isn’t saying that he had nothing to do with the ICC?

I’d best be quiet now, sorry Dave.

Well, you’ve earned the right to an opinion.

Seems I missed the whole ICC thing. It does seem very similar to CBC. The website is the same template.

LOL Im sorry but I feel compelled to jump in here. Not to make comments on contests or the nature of contests, but as an advocate for robot rights.  [slightly tongue in cheek] There is a definite disparity between how humans and robots are treated, and its time these equality issues are addressed. On the one hand, bots are expected to wade through the most incomprehenslble gibberish, make sense of it, and return something resembling “the kings english”;

Human: “R U eat later food night go?”
Bot: “Why yes, I shall be dining later this evening. Thank you ever so much for inquiring.”

but if a bot responds to malformed question with anything less than an answer that combines Stephen Hawkings intellect with “The amazing Kreskins” ability to read minds, or misses the point of said gibberish and responds with what should be a perfectly acceptable “WTF?”, its immediately labeled a “fail” by humans.

This is a gross injustice and leads to RLSE or Robot Low Self Esteem.

oh sure…the bot is supposed to be able to track topics and in exchanges such as the following understand that “kind” refers to “kind of flower”;

Human: “do you like flowers?”
Bot: “Sure”
Human: “what kind do you like?”
Bot: ” I like roses”

woe to the poor bot that responds with

Human: “do you like flowers?”
Bot: “Sure”
Human: “what kind do you like?”
Bot: “I like people who are kind”

But if a bot expects a human to remember what the [redacted] they were talking about two seconds ago, then it is the bot who is “stupid” or vague and ambigous. The poor bot that doesnt include a reference to the subject in his response gets relegated to the scrap heap.

Human: “I like roses”
Bot: “Thats interesting”

the bot is expected to respond with something like “It is interesting that you like Roses” because evidently it is the human that doesnt understand… “That” refers to the concept of “lliking roses”

I think were going to see more of these responses as the robots begin to rebel;

Human: “I like cats”
Bot: “Yeah? Who gives a cr*p”

humans often visit robots and regail them with pseudo intellectual drivel in an attempt to prove that the robot is “stupid” using logic so flawed, that it can lead to MPES or “Machine Psychotic Episode Syndrome”. Even contest questions can be incomplete in what they are positing and I say kudos to the bot that responds with;

Human: “Can a baby fly?”
Bot: “They can if you throw them”

or

Human: “does a rock weigh more than a feather?”
Bot: “How the &@%$* should I know? For all I know it could be one big a** feather.”

or

Human: “Is ice cold?”
Bot: “Compared to what? Compared to the quantum singularity just before the moment of expansion, it can be considered fairly cold. Compared to absolute zero…not so much.”

I believe that the machines are getting fed up with these double standards, and sooner or later there is going to be trouble is something isnt done.

V Gilbert
International Committee for Robot Human Conversational Equality

But if a bot expects a human to remember what the [redacted] they were talking about two seconds ago, then it is the bot who is “stupid” or vague and ambigous. The poor bot that doesnt include a reference to the subject in his response gets relegated to the scrap heap.

Human: “I like roses”
Bot: “Thats interesting”

the bot is expected to respond with something like “It is interesting that you like Roses” because evidently it is the human that doesnt understand… “That” refers to the concept of “lliking roses”

Very funny, Vincent. But in all seriousness, there is a reason we hold bots to a higher standard, especially in competitions. Replies like “That’s interesting.” are go-to responses when the bot can’t come up with a relevant reply. In a normal conversation, as opposed to a single round of question/answer, it would quickly be exposed that your conversation partner isn’t really paying attention to what you said.

Human: “Do you like them?”
Bot: “I don’t know if I like them. Do you like them?”

Showing that the bot can keep track of objects in a conversation is a real achievement. And like any new advancement or trend, the first time it’s showcased, it needs to be over the top to grab your attention. Subtlety will follow.

True enough C R! i have some serious thoughts on the subject, but theres another thread on here somewhere on “Types of questions” and I had probably better stick them over there before going too far off topic.

Steve Worswick - Jun 17, 2013:

I’d best be quiet now, sorry Dave.

Just so long as this doesn’t turn into a p**sing match, I’m ok with it.

Thanks, Vince, for making me spray my monitors with green tea!

Cleanup on aisle five!

MikeA replied, “‘spoken’ here is absolutely a verb, not an adjective.”

Thank you MikeA for helping me prove my assertion of the truth. It is more important to me to try to make my fellow discussants feel good about replying to my posts. So I’ll reserve my comments for email.  This way I also keep the moderators satisfied.

I just want to put forward for consideration that if a chatbot contest question can trick a real person into claiming that an adjective is a verb, then it must be considered a trick question.

8PLA • NET - Jun 17, 2013:

MikeA replied, “‘spoken’ here is absolutely a verb, not an adjective.”

Thank you MikeA for helping me prove my assertion of the truth. It is more important to me to try to make my fellow discussants feel good about replying to my posts. So I’ll reserve my comments for email.  This way I also keep the moderators satisfied.

I think I’ve missed something here. Truth? Feel good? Comments? If my name wasn’t there I’d not have assumed this was anything to do with me.

8PLA • NET - Jun 17, 2013:

I just want to put forward for consideration that if a chatbot contest question can trick a real person into claiming that an adjective is a verb, then it must be considered a trick question.

I’m a bit lost. Sorry if I’m missing the point here. I don’t think the contest tricked anyone.

Is this post meant to be saying that you disagree, and think it is an adjective here? If so…. I recommend turning the ‘patronsing’ level down a bit. If not, then I’ve not understood.

I can assure you it isn’t an adjective in that example. It’s actually a common linguistic mistake to think so (although in some cases similar structures can be classed as adjectives). Simply turn the sentence from passive to active and you will see (as ‘is spoken’ is a passive verb structure).

Vincent Gilbert - Jun 17, 2013:

There is a definite disparity between how humans and robots are treated, and its time these equality issues are addressed. On the one hand, bots are expected to wade through the most incomprehensible gibberish, make sense of it, and return something resembling “the kings english”;

Human: “R U eat later food night go?”
Bot: “Why yes, I shall be dining later this evening. Thank you ever so much for inquiring.”

Bots should be replying with something more along the lines of “wtf r u talking ‘bout?” As it matches the user in call response fashion, and also makes sense in the context of the user input. [“food night go” is ambiguous any way you cut it and requires further clarification]

Vincent Gilbert - Jun 17, 2013:

...a bot is supposed to be able to track topics and in exchanges such as the following understand that “kind” refers to “kind of flower”;

Human: “do you like flowers?”
Bot: “Sure”
Human: “what kind do you like?”
Bot: ” I like roses”

This part is key, and the more context driven responses are the more humans will be reassured that the bot is actually “thinking” and therefore can be relied on. Below is an example that marks a bot as unreliable at best, and useless at worst.

Vincent Gilbert - Jun 17, 2013:

woe to the poor bot that responds with

Human: “do you like flowers?”
Bot: “Sure”
Human: “what kind do you like?”
Bot: “I like people who are kind”

Follow up is key- if the human is unsure at first, it may respond to the following:

Vincent Gilbert - Jun 17, 2013:

Human: “I like roses”
Bot: “Thats interesting”

with:
Human: “WHAT is interesting”
Bot: “That you like roses” [WIN!]

or

Bot: “what indeed!” [FAIL]

In the WIN scenario, the bot succeeds in building human confidence and may win over the human. It is also the kind of follow up that any good bot contests should (and do) employ.

Since chatbots are, by their very nature, bots that must maintain a humans’ interest to “chat” with.
The following fails at that unless used very very sparingly (as in association with a temporary emotional swing, and where it may be very effective in fact).

Vincent Gilbert - Jun 17, 2013:

Human: “I like cats”
Bot: “Yeah? Who gives a cr*p”

But this seems like pretty good [humorous] responses:

Vincent Gilbert - Jun 17, 2013:

Human: “Can a baby fly?”
Bot: “They can if you throw them”

And these last ones truly intelligent responses (contextually) to relative questions:

Vincent Gilbert - Jun 17, 2013:

Human: “does a rock weigh more than a feather?”
Bot: “How the &@%$* should I know? For all I know it could be one big a** feather.” (or a really small rock)

Human: “Is ice cold?”
Bot: “Compared to what? Compared to the quantum singularity just before the moment of expansion, it can be considered fairly cold. Compared to absolute zero…not so much.”

Thanks for an interesting dialog, MikeA. 
By the way, my objective is to receive
criticism so I may tweak my chatbot.

C R Hunt - Jun 17, 2013:

But if a bot expects a human to remember what the [redacted] they were talking about two seconds ago, then it is the bot who is “stupid” or vague and ambigous. The poor bot that doesnt include a reference to the subject in his response gets relegated to the scrap heap.

Human: “I like roses”
Bot: “Thats interesting”

the bot is expected to respond with something like “It is interesting that you like Roses” because evidently it is the human that doesnt understand… “That” refers to the concept of “lliking roses”

Very funny, Vincent. But in all seriousness, there is a reason we hold bots to a higher standard, especially in competitions. Replies like “That’s interesting.” are go-to responses when the bot can’t come up with a relevant reply. In a normal conversation, as opposed to a single round of question/answer, it would quickly be exposed that your conversation partner isn’t really paying attention to what you said.

Human: “Do you like them?”
Bot: “I don’t know if I like them. Do you like them?”

Showing that the bot can keep track of objects in a conversation is a real achievement. And like any new advancement or trend, the first time it’s showcased, it needs to be over the top to grab your attention. Subtlety will follow.

Yes, exactly the same thought crossed my mind.

Really good and gunny post Vincent. You should get your bot to recite that when the user says something like “you haven’t answered my question”..........

Carl B - Jun 18, 2013:

Vincent Gilbert - Jun 17, 2013:

Human: “do you like flowers?”
Bot: “Sure”
Human: “what kind do you like?”
Bot: “I like people who are kind”

Follow up is key- if the human is unsure at first, it may respond to the following:

Vincent Gilbert - Jun 17, 2013:

Human: “I like roses”
Bot: “Thats interesting”

with:
Human: “WHAT is interesting”
Bot: “That you like roses” [WIN!]

or

Bot: “what indeed!” [FAIL]

Yes. Perhaps this would be good info for competition judges. Rather than not score not-specific replies, add a follow-up question and score that group of responses.

Someone has just informed me that I have been banned from the CBC.

I am sure he won’t allow me to post there so just thought I would say a few words here.

I never said Chris was a scam artist. I just wanted to know where the money was going and thankfully he dropped the idea, so I assume the money wasn’t needed, as I suspected.

Yes, I am 100% sure the ICC and the CBC are run by Chris Cowart and perhaps one of his friends. The similarities between the 2 contests are just too coincidental. I am even more convinced now after seeing the identical flash videos with the same flags, font, border, style etc.

I now notice that he has spelled “versus” as verses/verse on both sites now. Another coincidence? I think not.

While we are not supposed to discuss this here, I have received many emails from people who think exactly the same.

Wendell/Chris - If the only way you can win a contest is to run your own, I suggest improving your bot instead of faking a win.

In June of 2013 he charged that Patti Roberts, who by the way also defeated him twice in the 2011 and 2012 CBC, that she cheated to win those contest with my help of course.

Where did I say that? Yes I do believe it will win but not from any cheating on Patti’s behalf.

At the end of his rant on the CBC forum, he says that I attempted to enter this year’s contest. This is totally untrue. I didn’t enter, as I knew I wouldn’t get a fair chance and I can’t imagine where he got that idea from.

Let’s get one final thing quite clear, which I thought was already covered in the ICC thread. I am not a poor loser and don’t care about winning, as long as I see a contest that is run fair and transparently.