I was looking at some past Loebner Prize contest logs to see how well the robots were answering judges’ questions and how the chat was going at the contest. To my disappointment, I couldn’t really read the scripts because they are not displayed in a dialog format. Here is my question. Why the test format has to be that way - four windows? Can it be just two windows, one is human chatting and the other bot chatting and in each window an ABAB dialog would be displayed? I think that will also be easier for the judge to review the conversations at the end of the contest to make the more accurate final judgement as to which bot is better than which bot. Is there a reason you can’t display the chat contents that way in the contest and after the contest? I think many people would be interested in readable conversations between a human and a language robot, especially the human-machine conversations produced at such a high profile contest, which represent the current language robots’ capabilities. Thanks.

I have converted some transcripts, they are available here: Click here

Regards

This. Is. Awesome. Thanks a lot!

Hi Denis,

Why didn’t I know this earlier? Thank you for the work. At least Erwin Van Lun should make a link to it in chatbor.org so that more people will know it. Did you let him know?

Seen!!! I’m work on this on Friday!! Many thanks Denis!

created a link on this page:
http://www.chatbots.org/awards/loebner_prize/year_2012/

We’d be happy to host the transscripts in the future btw, in case you don’t want to maintain these scripts forever..