From the official rules:

The rules and guidelines maybe changed at anytime to insure fairness. It is impossible
to forsee all contingency that may arise during the contest so the CBC reserves that right

And, from the email:

Let me say the CBC has tried each year to accommodate as many bots as possible without having a lot of rules to meet. However, based on collecting the responses to the bots additional rules are needed. We had approximately 40 bots with 4 judges so each judge had around 10 bots each. We started around 10:00 am Sunday morning and went basically non-stop until 11:00 pm that night. That’s 13 hours to do a mere 10 bots a piece. You can not expect a volunteer judge to put in that amount of time.

And:

Bottom line is we are not doing that again. It is a necessary rule change and one where you have an entire year to implement. In next years contest you must provide a way where the questions can be copied and pasted into your input box as well as your bot’s responses. We can not accept the idea of the botmaster submitting his logs after the testing due to the possibility of altering the logs. That shouldn’t be hard to implement.

Personally, I don’t have a single problem with either rule change. In fact, I think that the new rule about making sure that there’s an easily accessible means of copying the bot’s output should have been in the original rules to begin with, but “hindsight being 20/20” and all…

As a freshman at CBC, I felt 79b.org played well.

I’m very interested to see these questions.

I don’t expect to hear about the judging until the weekend, considering how long it took for the collection phase took, but I’m a patient person, so I’m content to wait.

Victor,

I just want to credit you for being a positive influence
in the design of 79b.org which is now all Perl/CGI.

79b.org originally starting out in PHP, my personal favorite.
However, Perl is awesome and so reliable for building a bot
for a contest.

Thank you.

-Tom

P.S. Besides this simple thank you note to Victor,
      credits are certainly due to others.

Hi guys, I know someone who entered and they just posted all of the questions up on a forum.  Should I warn them about that rule you mention or is it okay at this stage ?

I think that’s a good idea, Freddy. We were all warned in one of the emails each entrant received, but some may not have read the emails. And the consequences for revealing the questions and answers before official notice is given is a lifetime ban from the competition, so it wouldn’t be a good idea to “jump the gun”, quite yet.

Thanks Dave, I will get right on it.

Excellent.  My work here is done   Just kidding.  But I’m pleased that you’re happy with your choice!

The transcripts are up on the CBC 2011 web site.
http://chatterboxchallenge.com/

Whereabouts? I don’t see them.

Now I do. I guess this means the questions are now available to be discussed?

From first glance, it appears that a few bots were being prompted along by the judge. Especially the question about “who was the computer who appeared on Jeopardy”. Many of them had a follow up question “Who is Watson”.

Another example being “How many sides on an octagon”. A few bots who didn’t get this correct were asked a rephrased version - “what is an octagon”

I am assuming this isn’t being counted towards the marks? If it is, then ALL the bots should have been given the same chance.

What happened to the Slankola bot? I was looking forward to reading those logs!

I noticed a lot of “leeway” from the judges with some bots, but not others, as well. Perhaps not enough for me to cry foul, but enough to give me pause. Please note that I’m not complaining here. When you have volunteer judges that put in 13 hours straight to judge a contest of this sort, there are bound to be instances like this.

I haven’t read all of the transcripts yet, but so far, Mitsuku has done the best, IMHO. And given the other entrants whose logs I’ve read, I feel a lot better about Morti’s chances for a “good” showing.

Try as they might, there is always variation between judges. It is compounded because of the user interfaces and connection problems of some of the bots. It took the 4 judges 13 hours to complete the task. I know the judge only spent 9 minutes with my bot, so some must have been very hard to work with.

Hopefully it won’t come down to a point or 2 between those that make the finals and those that don’t. Since the points don’t carry over it is just a question if the bot is in or out for the next round.

English Tutor is also missing.