The
ChatterBox challenge is an annual
chatterbot contest offering money and awards to contestants in several different categories.
Unlike the Loebner prize, this competition is designed to be easy to enter.
It takes place on the web, and the bots themselves may be on the web, downloadable or available via ICQ, MSN or some other messenger service.
Over 100 bot per year take part.
It has been running since 2001.
Rules
The rules, as featured on the contest's site, are as follows:
The Chatterbox Challenge is open to all kinds of chatterBots - Website
Bots, Downloadable Bots, and IRC Bots - written in any programming language.
Your
bot doesn't have to be set up in any particular way to enter.
It only needs to be
available so people can talk with it.
All botmasters are invited to enter their bots.
Rules for entering are:
Only English speaking chatterbots are allowed to enter. Only one chatterbot per person.
This also means you can't enter different versions of the same bot that has been specifically geared towards winning a certain category.
Although the Chatterbox Challenge welcomes all bots originality is a key factor in how successful a bot will be.
Blatant clones will be disqualified from the contest. AIML Bots - you are free to enter a bot using AIML (Artificial Intelligence Markup Language).
However, you can not enter a bot that uses Alice's data base.
Clones such as these will not be allowed to enter the contest.
Clones however maybe hard to detect.
If during the 10 question part of the contest a bot has identical or virtually identical responses that bot will be disqualifed from that category.
These bot however will still be allowed to compete in other categories.
Judging <br />
9 bots will be selected for the finals based on conversations held with the
bots by the judges in the contest.
The 10th bot will be selected by a vote of
the botmasters.Because of human errors and/or technical problems judging is a
very imperfect process.
With that said the finals may not be limited to 10 bots only.
If deserving additional bots maybe added.Prizes and Awards
James Livingood, the contest sponsor, has donated the following awards
for Best Overall Bot:
$1,000 for 1st place and a Gold Medal $500 for 2nd place and a Silver Medal $300 for 3rd place and a Bronze Medal In addition the Gold, Silver, and Bronze Medals will be awarded in the following
categories: All bots may enter multiple categories and win multiple awards.
Most Popular Most Knowledgeable Most uniquely interesting character/personality Best Interface Functionality or "most capable" Best Teachable/Learning bot Public Voting
The public will be invited to vote on who they feel are the best 3 bots in the contest.
Please chat with as many bots as possible and then ask yourself these questions.
How did the
overall conversion go?
Was the bot responsive or simply evasive?
Did the bot for the
most part answer and respond back appropriately or was the conversation full of off
the wall remarks?
The better bots should give you the impression that you are really
talking to a human being.
Compare the qualities and shortcomings of each bot and then
place your vote in a fair and unbiased way.
The contest depends on the goodwill of all
those taking part.
2006 Fiasco
The members running this contest seem to have no problem alienating their fellow bot masters.
When foul play was suspected by some participants in the 2006 contest, the contest organizer Wendell Cowart not only denied the claims but attempted to ostracize those that suspected it.
This lead those contestants to be all the more suspicious since Mr. Cowart ended up winning his own contest.
The debate comes in over a line in the rules that states "If deserving additional bots maybe added."
It has been confirmed by the
Internet Archive that this rule has been in place since at least December 15, 2005.
After it appeared all had blown over and people had either forgotten the problems or simply held them at bay, the following was posted on the contest site:
<blockquote>
January 13, 2007 - It's unfortunate that one of our botmasters from last years contest Bob Norris has attempted to sabotage the contest.
This was posted on Wikipedia as far back as May of 2006:
<br /><br />
Controversy There is much controversy over the legitimacy of this contest.
Many have pointed out that the people running the contest should not also be participating in the contest.
In the 2005 contest, six of the nine finalists were contest organizers.
In the 2006 contest, the rules were changed after judging was finished so that Wendell Cowart, the contest organizer, could get his bot in the finals.
Although Cowart's chatterbot had scored extremely low during the questioning, the added line in the rules stating, ,” allowed his bot to slip in.
Less than a week later, the final results were released and Cowart’s chatterbot had won first place and the $1,000 that accompanies it.
<br /><br />
This of course is untrue.
The line "If deserving additional bots maybe added" has been in place for several years now which many botmasters will attest to.
Oddly enough Bob Norris had the gumption to enter this year's contest.
When asked why would he want to enter a contest he thought was rigged he replied:
<br /><br />
"Because I have a good chance to win"
<br /><br />
To which I replied:
<br /><br />
"How can you possibly think you have a good chance to win when you believe the contest is rigged?"
Sorry Bob but you can't have it both ways.
<br /><br />
The CBC has had its distracters from the very beginning from various botmasters who are sore losers who didn't like the end results.
Instead of looking at the shortcomings of their own bot they attack the contest itself.
However, the vast majority realize this and despite their negative efforts the contest continues today.
In short, it's really very simple.
If you believe the contest is rigged then simply don't enter.
No need to go posting lies about the contest when it spoils the fun for the rest of us.
</blockquote>
This personal attack posted on a supposedly professionally run contest site only leads to more questions about it.
The matter should have been dropped by the contest staff if there was no merit to it.
In addition, neither the quote above nor this edit were made by Bob Norris.
See also
Chatterbot Turing test Loebner prize Chatbot competitionsExternal links
The ChatterBox Challenge Chatterbox Challenge 2003 Bronze Winner
Wikis
Search
