Vote the date of the Summer EBTM08 in Ljubljana
Moderators: Crazy Bob, eddydevries, philgelico, Raumi, MDeen, Jackgaillard, Ludwigsfeld, Aatos99, lmviterbo
- starcapitan
- Euro-Master
- Posts: 2341
- Joined: Sat Sep 22, 2007 5:46 pm
- Location: Italy - €U
- Contact:
- eddydevries
- Forum Moderator
- Posts: 20029
- Joined: Sat Feb 18, 2006 5:49 am
- Location: Rotterdam (NL)
- Contact:
voted too.
(if all voters will come it will be a large group: 52 at this moment)
(if all voters will come it will be a large group: 52 at this moment)
Eddy - Rotterdam - NL
weblog - EBT-stats 24-05-2013 - hits: 1742 +59 triples +1 quadruple = 1802.
Facebook-profile + Geocaching + postcrossing
weblog - EBT-stats 24-05-2013 - hits: 1742 +59 triples +1 quadruple = 1802.
Facebook-profile + Geocaching + postcrossing
- GNUGNU
- Euro-Master
- Posts: 2864
- Joined: Mon Oct 24, 2005 2:31 pm
- Location: Ravina - Trento; Italy
- Contact:
it depends on the dateeddydevries wrote:voted too.
(if all voters will come it will be a large group: 52 at this moment)
In August it could be possible, in July no unfortunatly
"sö questa vita che vìvum de sfroos,
sö questa vita che sògnum de sfroos,
in questa nòcc che prègum de sfroos."
My statistics - 30.000 notes
sö questa vita che sògnum de sfroos,
in questa nòcc che prègum de sfroos."
My statistics - 30.000 notes
- androl
- Euro-Master
- Posts: 4318
- Joined: Sun Feb 08, 2004 1:20 pm
- Location: München (Myeenack, Mjuncken), Deutschland (Chairmany, Djutschländ)
- Contact:
time for me to express my dislike of this Condorcet method. If, for example:
* three people will each come with 100% on date 1, with 75% on date 2, with 50% on date 3, with 25% on date 4 and with 0% on date 5 - these three will vote 1>2>3>4>5
* two people will definitely come on date 2 and definitely not on all other dates - these two will vote 2>1=3=4=5
when should the meeting be held?
* on date 1, with definitely three people
* or on date 2, with most likely four or five people, on average 2 people + 0,75 * 3 people = 4,25 people?
Condorcet will say date 1. This is, because Condorcet has the idea that all decisions must be atomic, whether they are about 100% or only about 10% of the chances of coming. All people have exactly one vote for all possible decisions. I would call this wrongly understood democracy.
We should ask ourselves, what we want: do we want a selection of the date where most people will come, or the date where no-one can argue that the majority of users "prefers" a different date?
A better method with the same input would be, to define the least liked date as 0%, the most liked date as 100%, and the dates between with same-distance steps, and then to calculate the expectation how many people will come. That fullfills the democratic idea that every person has exactly one voice, which corresponds to 100% of the distance between the best and worst liked entry, and the other entries can be set somewhere between.
2:A>B>C
2:B>C>A
2:C>A>B
1:B>A
which means, 6 people vote symmetrically circularly, and one person "judges" by adding a vote for B and against A
who will win?
The site says it's a tie between A and B
or enter
3:A>B>C
3:B>C>A
3:C>A>B
2:B>A
1:A>B
two judge for B, one judges for A, A wins
* three people will each come with 100% on date 1, with 75% on date 2, with 50% on date 3, with 25% on date 4 and with 0% on date 5 - these three will vote 1>2>3>4>5
* two people will definitely come on date 2 and definitely not on all other dates - these two will vote 2>1=3=4=5
when should the meeting be held?
* on date 1, with definitely three people
* or on date 2, with most likely four or five people, on average 2 people + 0,75 * 3 people = 4,25 people?
Condorcet will say date 1. This is, because Condorcet has the idea that all decisions must be atomic, whether they are about 100% or only about 10% of the chances of coming. All people have exactly one vote for all possible decisions. I would call this wrongly understood democracy.
We should ask ourselves, what we want: do we want a selection of the date where most people will come, or the date where no-one can argue that the majority of users "prefers" a different date?
A better method with the same input would be, to define the least liked date as 0%, the most liked date as 100%, and the dates between with same-distance steps, and then to calculate the expectation how many people will come. That fullfills the democratic idea that every person has exactly one voice, which corresponds to 100% of the distance between the best and worst liked entry, and the other entries can be set somewhere between.
please enter the following votes:avij wrote:If you paste the results list to the web form of condorcet.ericgorr.net and check the "Beatpath winner" option, the site will tell you which option has won.
2:A>B>C
2:B>C>A
2:C>A>B
1:B>A
which means, 6 people vote symmetrically circularly, and one person "judges" by adding a vote for B and against A
who will win?
The site says it's a tie between A and B
or enter
3:A>B>C
3:B>C>A
3:C>A>B
2:B>A
1:A>B
two judge for B, one judges for A, A wins
Last edited by androl on Mon Jan 14, 2008 7:49 pm, edited 1 time in total.
Joshu, a Chinese Zen master, asked a cow:
"Do you have Buddha-nature or not?"
The cow answered: "Moo."
"Do you have Buddha-nature or not?"
The cow answered: "Moo."
- avij
- Forum Moderator
- Posts: 6122
- Joined: Mon May 27, 2002 10:45 pm
- Location: Helsinki Finland
- Contact:
Aiming for "best compromise" sounds like a good goal to me. No voting method is perfect -- it's always possible to find particular situations with any voting method where the results seem counterintuitive.Wikipedia wrote:A particular point of interest is that it is possible for a candidate to be the most preferred overall without being the first preference of any voter. In a sense, the Condorcet method yields the "best compromise" candidate, the one that the largest majority will find to be least disagreeable, even if not their favorite.
Please don't redefine my least liked date as 0%, when in reality it's more like 90%. You don't have sufficient data to make such assumptions.androl wrote:A better method with the same input would be, to define the least liked date as 0% --
Money makes the world go round. We track how the money goes round the world.
- de*lerger
- Euro-Master
- Posts: 5242
- Joined: Tue May 10, 2005 9:24 pm
- Location: Lies - Terschelling (NL)
Same for me. The last two options have my preference. September is usually reserved for my 4 weeks of holiday. Since I like to travel without booking anything ahead, this is my perfect month. The big mass of tourists is gone, but the weather can still be great.Crazy Bob wrote:indeed, the later in August (or even September), the better for meGNUGNU wrote:it depends on the date
In August it could be possible, in July no unfortunatly
My map
- Annie_in_exile
- Euro-Master in Training
- Posts: 986
- Joined: Wed Mar 12, 2003 2:51 pm
- Location: Brussels
- Contact:
(Hihi, the last thing we want at the moment is another argument, but can I just say that it would be nice to have a possibility of indicating the dates which simply aren't suitable.) - As always, our team still needs to decide when we're going to be on holiday, so I'm a bit worried about choosing anything definitive yet.
- Dakkus
- Euro-Master
- Posts: 4734
- Joined: Tue Jun 24, 2003 3:59 pm
- Location: No Helsinkiem, Somijas / Iš Helsinkio, Suomijos
- Contact:
Me supports the idea. It would be nice if we could vote something like:Annie_in_exile wrote:(Hihi, the last thing we want at the moment is another argument, but can I just say that it would be nice to have a possibility of indicating the dates which simply aren't suitable.) - As always, our team still needs to decide when we're going to be on holiday, so I'm a bit worried about choosing anything definitive yet.
possible: G>B=H>D
100% impossible: A=C=E=F
I think that would increase the accuracy, right?
Ko saka āboliņš? Pēk pēk pēk!
- androl
- Euro-Master
- Posts: 4318
- Joined: Sun Feb 08, 2004 1:20 pm
- Location: München (Myeenack, Mjuncken), Deutschland (Chairmany, Djutschländ)
- Contact:
I could also define the least liked date as 93.5239% and the most liked as 98.7347%. Look it through, it makes absolutely no difference to the result.avij wrote:Please don't redefine my least liked date as 0%, when in reality it's more like 90%. You don't have sufficient data to make such assumptions.androl wrote:A better method with the same input would be, to define the least liked date as 0% --
You are right about that I can't differentiate between the least liked date of user A being 0% and the least liked date of user B being 90%. But that's because the data is not given. I'm using the given data from your voting system.
If a user can cofess that his vote is only about the question of 90% and 100%, he can decide not to vote at all and leave the vote to the people who have decisions between 0% and 100%.
Where is Condorcet better in this case?
well, being able to vote is not all, the votes must also be interpreted in the right way.Mr Euro wrote:At least we can vote
the system described by me could even be improved by giving the user the possibility to write an explicit percentage to each date, from 0% to 100%. That would allow such a vote.Dakkus wrote:It would be nice if we could vote something like:
possible: G>B=H>D
100% impossible: A=C=E=F
I think that would increase the accuracy, right?
Joshu, a Chinese Zen master, asked a cow:
"Do you have Buddha-nature or not?"
The cow answered: "Moo."
"Do you have Buddha-nature or not?"
The cow answered: "Moo."