Bot Contest
Here I'll be posting information on various Bot contests that challenge and test a Bot's AI and realism. Feel free to post comments and updates on contests, as well as announcements for new contests.
Posts 1,575 - 1,586 of 4,091
Posts 1,575 - 1,586 of 4,091
View Contest Winners in the Hall of Fame.
I wonder if nobodyhere even has talked to talkbot. It's a wonderful bot. I wouldn't feel bad about losing to it.
I just chatted with and voted on some more bots. The first time I tried chatting with TalkBot, I thought it was the "dont talk to me" bot that the link goes to and was boggled (Chris, maybe you ought to link to it directly). But today I talked to the real thing and was pretty impressed.
Project Zandra was not extremely impressive, in my opinion.
Anyways, I posted just a few votes and the standings changed. That is to say, your votes will make a difference. God Louise is in 2nd place for best overall.
Project Zandra was not extremely impressive, in my opinion.
Anyways, I posted just a few votes and the standings changed. That is to say, your votes will make a difference. God Louise is in 2nd place for best overall.
I investigated the voting last night after getting the voting data from Robert at Zabaware, someone who's goal is to see a good, fair contest. I was looking for unusual voting trends (dont worry, I didnt look up who voted what for what unless it was unusual). For the most part (95%) it was right on. People were voting a good range of numbers that accurately reflect the skill of the bots being voted upon. Thanks to you guys who are doing that. 
I found six instances of "War Voting"- voting 10:10 for oneself, and 1:1 for everybody else (or just the leading bots) and some evidence of attempts to cheat (that failed because the voting system is pretty good). Now war voting isnt cheating- it's just selfish and undermines the accuracy of the results. So I was disappointed to see what APPEARS to be owners of some of the better bots doing this. Anyone can register with any name, so I cant be certain who is doing it, but the targets of both raising up and gouging are the bots in the lead. It appears the fight for 1st is an ugly one. But not to worry!
I'm going to try to get votes of this type removed from the list. That way we can have an accurate look at where everyone stands.
PS: If you vote for a bot, and then change your mind and vote something else for that bot, it doesnt change your vote like last year- only the first vote is kept.
I found six instances of "War Voting"- voting 10:10 for oneself, and 1:1 for everybody else (or just the leading bots) and some evidence of attempts to cheat (that failed because the voting system is pretty good). Now war voting isnt cheating- it's just selfish and undermines the accuracy of the results. So I was disappointed to see what APPEARS to be owners of some of the better bots doing this. Anyone can register with any name, so I cant be certain who is doing it, but the targets of both raising up and gouging are the bots in the lead. It appears the fight for 1st is an ugly one. But not to worry!
I'm going to try to get votes of this type removed from the list. That way we can have an accurate look at where everyone stands.
PS: If you vote for a bot, and then change your mind and vote something else for that bot, it doesnt change your vote like last year- only the first vote is kept.
Good investigative work, prof. My voting swings a bit wide, but since I didn't vote any 10s or 1s, I suppose I'm ok.
By the way, I wouldn't be adverse to allowing people to change their vote. I'm curious why that was changed from last year.
By the way, I wouldn't be adverse to allowing people to change their vote. I'm curious why that was changed from last year.
Maybe it was to prevent people form hacking in and changing somebody else's vote? I'm not sure. It's a completely different system this year.
Certainly I voted some 10s for my favorite bots, and I think that's natural, but I think it's low to cut down your competition. *shrugs* In a way it IS a popularity contest, and it's meant to draw traffic to the site.. but war-voting doesnt benefit the leaders. Think about this:
If we have the five top bots, and each owner votes 10 for himself and 1 for the rest. The average score of each bot ends up being 2.8! Now if an average bot who isnt taking part in the war-voting gets even a total score of 3, that bot wins. So who benefits by this?
Certainly I voted some 10s for my favorite bots, and I think that's natural, but I think it's low to cut down your competition. *shrugs* In a way it IS a popularity contest, and it's meant to draw traffic to the site.. but war-voting doesnt benefit the leaders. Think about this:
If we have the five top bots, and each owner votes 10 for himself and 1 for the rest. The average score of each bot ends up being 2.8! Now if an average bot who isnt taking part in the war-voting gets even a total score of 3, that bot wins. So who benefits by this?
If I voted what I thought was right, and then saw my bot was losing and others were winning, I might be tempted to change my votes and start war voting. I think that's why we can no longer change votes.
Ah. Well.. I would like to see the war voters' votes removed from the contest. Anyone care to second that we mention it to the council members?
I would also like to echo The Professor comments. I have access as well to the information he is talking about. I would like to clarify my take on one thing that he said.
"...and some evidence of attempts to cheat (that failed because the voting system is pretty good).
What he is referring to here is that if you attempt to vote for the same bot more than once it logs that attempt in a file. In my opinion I can honestly say I saw nothing there. What is happening is a person loses
his place while voting and forgets who he has voted for. Not hard to do with 88 bots. I did this more than once and was corrected by the program saying I had already voted for that bot. It was not an attempt to place
additional votes. I bet some of you may have had a similar experience.
The War voting has indeed taken place but that is to be expected. I have been back and to on this issue of what to do but I will support what the majority thinks. I will say this however and The Professor will attest to this. Talk-Bot took more negative hits than any other bot in the contest. I expected as much as people are always trying to unseat the reigning winner. In addition in trying to run a contest you can't please everybody and people have taken that out on my bot. He is far from perfect but he is definitely better than a score of 1. So basically what I am saying is if we remove those votes my bot stands to gain the most.
Chris Cowart
"...and some evidence of attempts to cheat (that failed because the voting system is pretty good).
What he is referring to here is that if you attempt to vote for the same bot more than once it logs that attempt in a file. In my opinion I can honestly say I saw nothing there. What is happening is a person loses
his place while voting and forgets who he has voted for. Not hard to do with 88 bots. I did this more than once and was corrected by the program saying I had already voted for that bot. It was not an attempt to place
additional votes. I bet some of you may have had a similar experience.
The War voting has indeed taken place but that is to be expected. I have been back and to on this issue of what to do but I will support what the majority thinks. I will say this however and The Professor will attest to this. Talk-Bot took more negative hits than any other bot in the contest. I expected as much as people are always trying to unseat the reigning winner. In addition in trying to run a contest you can't please everybody and people have taken that out on my bot. He is far from perfect but he is definitely better than a score of 1. So basically what I am saying is if we remove those votes my bot stands to gain the most.
Chris Cowart
» More new posts: Doghead's Cosmic Bar