The best way to equalize the voting system: Input.

Forum for discussing AniDB rules & standards. No small talk!

Moderator: AniDB

akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

The best way to equalize the voting system: Input.

Post by akaneon.com »

I believe that we should keep the complete average system, however:

Voting 4-8 should be allowed without any given reason. If you feel a series was mediocre or not 10-worthy, you can vote 4-8 with no given reason.

However, I believe that you should have to give criteria if you're voting above 8 or below 3 (to amount to no less than 300 written characters about why you feel the series deserves a 9 or 10, or why you think it deserves a 3 or worse).

For those of you that think 300 characters is a lot, this post will amount to over 1000 by the time I finish it. The average word has 5 characters, that's 60 words on how you feel about something you're either pledging allegiance to or hating on.

There is no need for moderators to adjust user votes, rather, we need to adjust the input we collect from the users themselves.

If you think a given group/series/movie/episode is spectacular or really sucks, you can assuredly write 60 words on it. If you can't, then the validity of your vote itself should be drawn into question. We don't need a bunch of inconsiderate fanboys and/or useless bashers who don't provide input voting on series' that they didn't even really watch.

In addition, your vote and reason for voting should be publicly displayed next to your username, to prevent puppeteering. This is just my recommendation... it would show us who the fuck is throwing out all those 1s and 10s for no good reason.
goldenb
Posts: 10
Joined: Fri Sep 01, 2006 5:08 am
Location: Aboard the Vella Ciela / PT

Post by goldenb »

That's a nice suggestion, however i'm not sure a forced comment would be the solution the anidb staff is looking for (as far as I know even the average system was changed) but it was one of the best suggestions i read so far
akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

Post by akaneon.com »

goldenb wrote:That's a nice suggestion, however i'm not sure a forced comment would be the solution the anidb staff is looking for (as far as I know even the average system was changed) but it was one of the best suggestions i read so far
If you count the spaces, this post you just made is over 300 characters

I wouldn't call it 'forced', no one forces them to vote in the first place, this simply requires them to validate their reasons for voting
PetriW
AniDB Staff
Posts: 1522
Joined: Sat May 24, 2003 2:34 pm

Re: The best way to equalize the voting system: Input.

Post by PetriW »

akaneon.com wrote:I believe that we should keep the complete average system, however:
Thanks but that will not happen.
akaneon.com wrote:Voting 4-8 should be allowed without any given reason. If you feel a series was mediocre or not 10-worthy, you can vote 4-8 with no given reason.
Cute, just a sec while I go vote 4 for all SHS releases and we'll see if you still think it's allowed.
akaneon.com wrote:However, I believe that you should have to give criteria if you're voting above 8 or below 3 (to amount to no less than 300 written characters about why you feel the series deserves a 9 or 10, or why you think it deserves a 3 or worse).

For those of you that think 300 characters is a lot, this post will amount to over 1000 by the time I finish it. The average word has 5 characters, that's 60 words on how you feel about something you're either pledging allegiance to or hating on.
Cute again, so you wish to make AniDB what AnimeNFO was in the past? Filled with spam "reviews" of everything? Please...
If you want comments you can read the group comments or reviews, both which draw enough spam as is, no need to force users to spam to be able to do simple votes.
akaneon.com wrote:There is no need for moderators to adjust user votes, rather, we need to adjust the input we collect from the users themselves.
Please go read up on statistics before you think this is true.
akaneon.com wrote:If you think a given group/series/movie/episode is spectacular or really sucks, you can assuredly write 60 words on it. If you can't, then the validity of your vote itself should be drawn into question. We don't need a bunch of inconsiderate fanboys and/or useless bashers who don't provide input voting on series' that they didn't even really watch.
I'm sorry but I think you fulfill the basic requirement of a inconsiderate fan boy really, we know why you're upset and we're not impressed.
akaneon.com wrote:In addition, your vote and reason for voting should be publicly displayed next to your username, to prevent puppeteering. This is just my recommendation... it would show us who the fuck is throwing out all those 1s and 10s for no good reason.
Puppeteering is relatively limited on AniDB, we nuke most of the puppets down pretty quick, you being able to see votes for individual entries would not change this and that you even imagine it's possible shows you are clueless. How can YOU prove that the users "fanboi" and "fanboy" are actually the same person? Leave the task to our scripts instead which continuously detect real puppet votes.

Yes I'm blunt but I'm trying to get a point through here:

THINK THROUGH WHAT YOU SUGGEST!
nwa
AniDB Staff
Posts: 585
Joined: Sat Jun 07, 2003 10:51 am

Post by nwa »

akaneon.com wrote:There is no need for moderators to adjust user votes, rather, we need to adjust the input we collect from the users themselves.
So you really think mods adjust user votes? You're wrong, we got an army of monkeys in rabbit suits that do that for us.


Honestly, we, the mods, actually don't much care about the votes, even though we vote ourselves, but the reason the voting system was changed, was cause many complained and the 1/10 voting really did get out of hand. So the new system, which isn't yet quite finalized, takes care of lame votes, showing more fair ratings.
akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

Re: The best way to equalize the voting system: Input.

Post by akaneon.com »

PetriW wrote:
akaneon.com wrote:I believe that we should keep the complete average system, however:
Thanks but that will not happen.
akaneon.com wrote:Voting 4-8 should be allowed without any given reason. If you feel a series was mediocre or not 10-worthy, you can vote 4-8 with no given reason.
Cute, just a sec while I go vote 4 for all SHS releases and we'll see if you still think it's allowed.
akaneon.com wrote:However, I believe that you should have to give criteria if you're voting above 8 or below 3 (to amount to no less than 300 written characters about why you feel the series deserves a 9 or 10, or why you think it deserves a 3 or worse).

For those of you that think 300 characters is a lot, this post will amount to over 1000 by the time I finish it. The average word has 5 characters, that's 60 words on how you feel about something you're either pledging allegiance to or hating on.
Cute again, so you wish to make AniDB what AnimeNFO was in the past? Filled with spam "reviews" of everything? Please...
If you want comments you can read the group comments or reviews, both which draw enough spam as is, no need to force users to spam to be able to do simple votes.
akaneon.com wrote:There is no need for moderators to adjust user votes, rather, we need to adjust the input we collect from the users themselves.
Please go read up on statistics before you think this is true.
akaneon.com wrote:If you think a given group/series/movie/episode is spectacular or really sucks, you can assuredly write 60 words on it. If you can't, then the validity of your vote itself should be drawn into question. We don't need a bunch of inconsiderate fanboys and/or useless bashers who don't provide input voting on series' that they didn't even really watch.
I'm sorry but I think you fulfill the basic requirement of a inconsiderate fan boy really, we know why you're upset and we're not impressed.
akaneon.com wrote:In addition, your vote and reason for voting should be publicly displayed next to your username, to prevent puppeteering. This is just my recommendation... it would show us who the fuck is throwing out all those 1s and 10s for no good reason.
Puppeteering is relatively limited on AniDB, we nuke most of the puppets down pretty quick, you being able to see votes for individual entries would not change this and that you even imagine it's possible shows you are clueless. How can YOU prove that the users "fanboi" and "fanboy" are actually the same person? Leave the task to our scripts instead which continuously detect real puppet votes.

Yes I'm blunt but I'm trying to get a point through here:

THINK THROUGH WHAT YOU SUGGEST!
I wouldn't mind if you voted 4 for all SHS releases because a 4 doesn't slope the overall score curve. The average of a 4 and an 10 is a 7. The diehard fans who actually watch the series WILL write 300 characters supporting their viewpoints. You can give SHS 100 4's for all I care, it won't skew the averages in the way that 1's and 2's do. (It takes 5 8+ ratings to account for 1 rating of 1 or 2 ratings of 2, but only 1.5 8's to account for 1 4).

Also, the comments should be periodically checked by human beings to ensure legitimacy. You won't be able to spam. You'll have to actually write, or you shouldn't vote at all.

For the record I am not a fanboy... but look through any series and tell me that any of these groups deserves a 1 or a 2 on their rating?

Show me ONE episode or ONE group or even ONE SERIES that has no redeemable qualities whatsoever and I'll gladly give up my case.

Your argument insults the level of intelligence of all anidb users in saying that we're:

1. Too lazy to write 300 characters, I've already written over 3000 in this topic alone
2. Too stupid to actually comment based on legitimacy
3. Willing to randomly spam 300 characters with no check in order to give out 10s
4. Too stupid to use the 'report' function to inform a moderator about spam
5. Too unobservant to notice that your scripts have obviously failed and that haters are still giving out 1s and 2s.



You guys are a community site, right? Why not give the community some pull in this decision? Test it out, see what happens. I can guarantee it will make the voting system more legitimate, hands down. This is VERY thought out.

I think that perhaps YOU are the one with the bias.... People of anidb, listen closely... If the moderators truly care about us...they will allow us to hold an open forum and vote, and then base their decisions on the contributed opinions of the actual contributing members of this forum who are not afraid to openly disclose their opinions (albeit that this is overly redundant, it has to be written this way to make sure everyone understands).
akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

Post by akaneon.com »

nwa wrote:
akaneon.com wrote:There is no need for moderators to adjust user votes, rather, we need to adjust the input we collect from the users themselves.
So you really think mods adjust user votes? You're wrong, we got an army of monkeys in rabbit suits that do that for us.


Honestly, we, the mods, actually don't much care about the votes, even though we vote ourselves, but the reason the voting system was changed, was cause many complained and the 1/10 voting really did get out of hand. So the new system, which isn't yet quite finalized, takes care of lame votes, showing more fair ratings.
What I meant by that is that the system auto-balances for 1's and 2's... However, the 1's and 2's should not be there randomly... If you give out a 1 or a 2, your name should appear on a list so that you can be a target of scrutiny; if you can dish out shit, you can certainly take some.
PetriW
AniDB Staff
Posts: 1522
Joined: Sat May 24, 2003 2:34 pm

Post by PetriW »

akaneon.com wrote:I wouldn't mind if you voted 4 for all SHS releases because a 4 doesn't slope the overall score curve. The average of a 4 and an 10 is a 7. The diehard fans who actually watch the series WILL write 300 characters supporting their viewpoints. You can give SHS 100 4's for all I care, it won't skew the averages in the way that 1's and 2's do. (It takes 5 8+ ratings to account for 1 rating of 1 or 2 ratings of 2, but only 1.5 8's to account for 1 4).
Hint, if enough people vote 4, it'll be a 4.
akaneon.com wrote:Also, the comments should be periodically checked by human beings to ensure legitimacy. You won't be able to spam. You'll have to actually write, or you shouldn't vote at all.

For the record I am not a fanboy... but look through any series and tell me that any of these groups deserves a 1 or a 2 on their rating?

Show me ONE episode or ONE group or even ONE SERIES that has no redeemable qualities whatsoever and I'll gladly give up my case.
So we FORBID people to vote 1 and 2, they'll just vote 3, people are people, they'll vote as they want.
Anyway, I personally would feel these qualify:
http://anidb.info/perl-bin/animedb.pl?s ... me&aid=459
http://anidb.info/perl-bin/animedb.pl?s ... me&aid=504
akaneon.com wrote:Your argument insults the level of intelligence of all anidb users in saying that we're:

1. Too lazy to write 300 characters, I've already written over 3000 in this topic alone
AniDB has 200000 users, 8000 have registered on the forum.
akaneon.com wrote:2. Too stupid to actually comment based on legitimacy
You expect people who vote 10 for naruto and 1 for Mushishi to provide good comments?
akaneon.com wrote:3. Willing to randomly spam 300 characters with no check in order to give out 10s
They did this on AnimeNFO already, I have no doubt whatsoever this is what'll happen because it always happens, requiring a huge amount of moderation work. We already nuke idiotic comments when they're brought to our attention.
akaneon.com wrote:4. Too stupid to use the 'report' function to inform a moderator about spam
They do already, never said this.
akaneon.com wrote:5. Too unobservant to notice that your scripts have obviously failed and that haters are still giving out 1s and 2s.
Our scripts don't target 1 and 2 voters for deletion unless it's excessive. You could say the exact same thing about 10 votes but then we'd have to ignore the votes of a lot of people (like 70% of all voters).
akaneon.com wrote:You guys are a community site, right? Why not give the community some pull in this decision? Test it out, see what happens. I can guarantee it will make the voting system more legitimate, hands down. This is VERY thought out.
We see no compelling reason to test this out, you've 100% failed to convince us in any way that it would be useful and >reduce< work for anidb moderators.
akaneon.com wrote:I think that perhaps YOU are the one with the bias.... People of anidb, listen closely... If the moderators truly care about us...they will allow us to hold an open forum and vote, and then base their decisions on the contributed opinions of the actual contributing members of this forum who are not afraid to openly disclose their opinions (albeit that this is overly redundant, it has to be written this way to make sure everyone understands).
PS: don't confuse the anidb forum with the anidb website. We're talking about 200k users and over 1M votes here.

The forum has 8000 registered users and like 10-20 posts per day, like 10 users sign up on anidb EACH HOUR. (no a statistic but from watching #anidb-log today)

The underwhelming people storm about this change only proves most find the change reasonable. I don't know if you noticed but so far very few have complained. Even in the SHS thread which you came from half the people don't seem to think it a bad change, and those people represent the most dedicated users. The average user does not visit the SHS forum, ever, they click the torrent somewhere and is happy.

Oh, and your ridiculous claim that people would be AFRAID to state their opinion... Are you fucking stupid? I hardly know what to say, I'll just sum it up with:
AniDB has never banned anyone for being critical of how we do stuff, we only ban people for abusing the servers / breaking the public rules. And everyone banned is welcomed back if they come talk with us about their ban and promise to stop breaking the rules.

akaneon.com wrote:What I meant by that is that the system auto-balances for 1's and 2's... However, the 1's and 2's should not be there randomly... If you give out a 1 or a 2, your name should appear on a list so that you can be a target of scrutiny; if you can dish out shit, you can certainly take some.
1 and 2 votes are not an issue compared to 10 votes.
Last edited by PetriW on Sat Jan 13, 2007 4:37 pm, edited 1 time in total.
akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

Post by akaneon.com »

Bahahaha.

Obviously you don't understand how the system of averages works.

1 1 or 2 can throw off 7 or 8 10s because of the curve modification.

A ton of 4's won't throw off the curve, and 8's won't unbalance it on the high end. Do you really need for me to draw you a linear example of this?

Also, those series you mentioned are from 97 and are definitely at least worthy of 4's, considering the time frame they were created in.

Forbid people to vote 1, 2, and 3 UNLESS they write a valid 300 character comment explaining their decision to downvote it. Forbid people from voting 9 or 10 unless they explain what's so great about it. You will be able to see their reviews, and delete the ones that are pure spam.

For the record, I voted 9 for Naruto and I voted 5 for Mushishi because Mushishi is boring as hell, in my opinion. Like, oh wow, it's an anime about plants! Secret plants! zzzz.... Lack of plot and character development. The animation quality is great, so is the voice acting, too bad the series itself puts me to sleep quicker than a Sigur Ros album and 10 tabs of valium.

GTO should be #1 without question. The fact that GTO isn't #1 draws this whole database into question, but that's a totally different story.

The thing about 10 votes is this: Most of us who vote 10 for a series are more than happy to write 300 characters about it. The people who vote 1...chances are, they didn't even watch the series.

You guys have basically set this up....

Also, you're lying through your teeth, man. I hate to say that, but it's the truth. The SHS forum that you're bashing has over 70 times the amount of posts and replies that this forum does.

You've blatantly admitted that you're biased and you support the new system because it gives points to the groups and animes that you personally like.

READ INTO WHAT THIS GUY IS WRITING, PEOPLE. HE DOESN'T CARE ABOUT YOU OR WHAT YOU THINK, OR THE ANIDB COMMUNITY, HE JUST WANTS TO SIT AROUND AND ACT HIGH AND MIGHTY. HE'S TOO LAZY TO DO ANY ACTUAL MODERATION, WHICH IS WHY HE'S TAKING SHOTS AT US. Scroll his posts, you can see 3 things:

1. He doesn't really care about making anidb better, he just likes to stick his fucking nose in the air and take shots at me
2. He can't do basic math, i.e. averaging, and is actually stupid enough to say some retarded shit like '1's and 2's don't matter!'. What a fucking idiot.
3. He can write over 10,000 characters across these threads but is too lazy to develop a voting system that works.


Fuck you buddy. Anidb sucks. Thanks for proving conclusively that you rigged the rating system, you fucking faggot.
exp
Site Admin
Posts: 2438
Joined: Tue Oct 01, 2002 9:42 pm
Location: Nowhere

Re: The best way to equalize the voting system: Input.

Post by exp »

Another thing to keep in mind is that we've already implemented your suggestion ages ago. Check the anime review score for anime entries. That's exactly what you're suggesting.
It works well for animes, but it's too much effort for other entry types.

*EDIT*
Now now now, watch where you're going there.
Do you even realize how ridiculus your claims are? I won't stop you, but you're making a complete idiot out of yourself here.

PS: and once again, please take care to distinguish between the anidb _website_ and the anidb _forum_.
*/EDIT*

BYe!
EXP
Last edited by exp on Sat Jan 13, 2007 4:37 pm, edited 1 time in total.
akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

Post by akaneon.com »

Lets say a series has 10 7 votes.

That gives you 70/100, or 7.0.

A 10 vote will give the series a 80/110, or a 7.2.

However, a 1 vote would give the series a 71 out of 110, or a 6.4.

That's not only a difference, that's a colossal difference. But you're too stupid to account for that.
Last edited by akaneon.com on Sat Jan 13, 2007 4:37 pm, edited 1 time in total.
akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

Post by akaneon.com »

Lets say a series has 10 7 votes.

That gives you 70/100, or 7.0.

A 10 vote will give the series a 80/110, or a 7.2.

However, a 1 vote would give the series a 71 out of 110, or a 6.4.

That's not only a difference, that's a colossal difference. But you're too stupid to account for that.

Notice how a 10 only gives you a .2 boost but a 1 gives you a .6 decrease? Stupid fuck.
PetriW
AniDB Staff
Posts: 1522
Joined: Sat May 24, 2003 2:34 pm

Post by PetriW »

akaneon.com wrote:Lets say a series has 10 7 votes.

That gives you 70/100, or 7.0.

A 10 vote will give the series a 80/110, or a 7.2.

However, a 1 vote would give the series a 71 out of 110, or a 6.4.

That's not only a difference, that's a colossal difference. But you're too stupid to account for that.

Notice how a 10 only gives you a .2 boost but a 1 gives you a .6 decrease? Stupid fuck.
You're just grasping for straws, this is very far from how it works.
akaneon.com
Posts: 11
Joined: Sat Jan 13, 2007 8:26 am
Location: Colorado Springs, CO
Contact:

Post by akaneon.com »

PetriW wrote:
akaneon.com wrote:Lets say a series has 10 7 votes.

That gives you 70/100, or 7.0.

A 10 vote will give the series a 80/110, or a 7.2.

However, a 1 vote would give the series a 71 out of 110, or a 6.4.

That's not only a difference, that's a colossal difference. But you're too stupid to account for that.

Notice how a 10 only gives you a .2 boost but a 1 gives you a .6 decrease? Stupid fuck.
You're just grasping for straws, this is very far from how it works.
Pick up a calculator, you'll see exactly how it works, dipshit.
PetriW
AniDB Staff
Posts: 1522
Joined: Sat May 24, 2003 2:34 pm

Post by PetriW »

akaneon.com wrote:Pick up a calculator, you'll see exactly how it works, dipshit.
Then compare said value on calculator to the vote shown in the list, omg they differ!!!
Locked