Art,

Robby may be too advanced for a chatbot contest, after becoming a commercial success with A.I.

It is hard to imagine, but Robby has actually created A.I. actors for the theatre:

http://www.anniedorsen.com/showproject.php?id=6

The sold out audience was cheering his A.I. after paying to get into the play.

Did you know the Robby is in the Guinness Book of World Records for the A.I. he created?

My guess is you knew this already.  So, I am simply making polite discussion here.

Yes, although about 18 years have passed, Robby and I have had a bit of history.

He was somewhat ahead of his time in that he explored the creative side of what makes humans tick in order to make use of tricks and techniques to make his bot appear more clever than others of that time.

I spent quite a bit of time chatting and experimenting with chatbots and A.I. and felt like Robby was doing some good things in the field. So I helped support him (see image).

He was also active in music - reference: Flux Oerstead (sp?).

Good times…

Art Gladstone - Aug 22, 2017:

Congratulations to the finalists! Chat on!

@  Robby, After all these years have you given up on the challenge or did you have an entry that I missed?

Come on man…jump on into the pool!

Hey Art, thanks for your support! I haven’t forgotten. Also, I haven’t given up. In fact I’m working on something new. I’ve got a sponsor that might help me enter the contest again one day. I don’t like the 4 finalists format of the contest. There is no reason to have the competition limited so that it fits in Hugh’s apartment! I also don’t like having the WSC decide who competes in the Loebner contest. Finally, open it up to the internet so that we can progress beyond what’s held on a hard disk. Actually, that’s just the start of my laundry list.

Having a message-by-message format is a good step in right direction, IMHO.

I guarantee that if you opened up the contest to the Internet, Pandorabots and everybody else would skyrocket in terms of capabilities. 

Let’s just keep it going, and best of luck to the participants this year!

∞Pla•Net - Aug 24, 2017:

Art,

Robby may be too advanced for a chatbot contest, after becoming a commercial success with A.I.

It is hard to imagine, but Robby has actually created A.I. actors for the theatre:

http://www.anniedorsen.com/showproject.php?id=6

The sold out audience was cheering his A.I. after paying to get into the play.

Did you know the Robby is in the Guinness Book of World Records for the A.I. he created?

My guess is you knew this already.  So, I am simply making polite discussion here.

8man, 

You are always polite, and that is just one of the things that I admire about you. Thanks for your kindness over all these years. If I ever enter the contest again, I may have to name my entry after you.

Robby Garner - Sep 1, 2017:

∞Pla•Net - Aug 24, 2017:

Art,

Robby may be too advanced for a chatbot contest, after becoming a commercial success with A.I.

It is hard to imagine, but Robby has actually created A.I. actors for the theatre:

http://www.anniedorsen.com/showproject.php?id=6

The sold out audience was cheering his A.I. after paying to get into the play.

Did you know the Robby is in the Guinness Book of World Records for the A.I. he created?

My guess is you knew this already.  So, I am simply making polite discussion here.

8man, 

You are always polite, and that is just one of the things that I admire about you. Thanks for your kindness over all these years. If I ever enter the contest again, I may have to name my entry after you.

MAJOR TOM?

I think we’re getting a bit far of field here, Art. A little banter is allowed, expected and even encouraged, but let’s not get carried away.

I’ve written a report on my entry: Mostly what went wrong, and some interesting trivia about the questions.
https://artistdetective.wordpress.com/2017/09/02/loebner-prize-2017

Though Robby may have other reasons for disliking Winograd schemas in Turing tests, I don’t think they’re worth devoting two questions to any more. It was interesting before the WSC ran and I’m sure Hugh’s ego insisted on their inclusion, but now that we know the state of the art is hard to distinguish from guesswork, they add no more value than any regular question with a reference: Only distinguishing programs that can not interpret pronouns at all.

Don,

I think the world of your article for preserving the secret ingredient of natural language processing research.  Can anyone, (besides Don), tell me what that secret ingredient is?

Reference: https://artistdetective.wordpress.com/2017/09/02/loebner-prize-2017

Don,

I agree with you, and this is relevant to the contest’s finalists. My feeling is that the Winograd schemas are certainly a good test of several things, but the Loebner Prize contest shouldn’t shy away from a full Turing test while there is one. I’ve felt that the WSC was a reaction to Eugene’s performance *in a given setting, with a given set of judges*. I don’t think it’s time to abandon the Turing test yet. The gate to the gold medal shouldn’t be locked by this one type of question.

In the past, the LPC has used preliminary Turing tests to determine the top programs. I would suggest that the results of that test would be more relevant to which bot has the best chance in the finals. (I could go on about this a while longer, but I’ll spare you that.)

If LPP2 can be established next year, using message-by-message interactions, I predict that judges’ individual scores for programs will rise to higher levels. I know this is obvious to some, but the number of channels of communication can be reduced, (i.e. stop having fake typing, possibly adjust for message answer delay, keep user experience looking the same) and then we can focus on verbal behavior and eliminate other factors that might influence the judge.

The messages are what is being judged. If you can’t tell by looking at them immediately who wrote them, I think that would be a good thing.

I like the Star Trek reference in your paper.

There are too many topics about the Loebner Prize 2017, so I’ll just put this here:

A link to the webcast of the final at Bletchley Park will be made available here on 16 September - around 12.30pm BST on this day.

that’s great, Don, but if I may ask… Where is “here”?

Click “here”

Thanks. Within that quote box the link wasn’t readily visible. I think I’ll fix that.

Congratulations, Steve.

Steve once again takes top honors in the Loebner Prize.
Good job!