neal | # 811 | 2011-11-29, 21:52 | Report

If you bothered to read my post, you'd have seen that it changes the results in about 5 categories.

Helmuth | # 812 | 2011-11-29, 22:11 | Report

Oh, sorry. I overflown it and tought it wouldn't change the order of the first 3 places.

You're talking about your Post #805 ?

Hmm... I don't understand it... again... so... I just overgone it and thought the experts, you and Wonko, are talking there to each other in your own language.

Perhaps you could try to explain it a little bit.
What is in the first column? I read there -1, +2, -4, +3 - are this the vote numbers or the places? What about the Categories? And what does at example this mean: "@@ -25,11 +25,11 @@"

Please to take into account: I'm not a native speaker, not a mathematician and over 30. Not the first time I feel myself very stupid at this thread but I'm willing to learn.

Wonko | # 813 | 2011-11-29, 22:22 | Report

Nevermind
Essentially, it's a "diff". That's some way to display differences in files or other sets of data.

If I understand the format neal chose correctly it means that the numbers in the front indicate the place/rank (1 - first, 2 - second, etc.).

The "-" and "+" signs indicate what's removed and added. So something like:
Means that the first entry stays the same. While the latter three differ between the two discussed methods for voting in the way that:
Entries 2-4 are remove and replaced with the ones marked with "+".
So
becomes
The problem we are now facing is that the two different methods for counting the votes yield in differences in some categories. I think neal mentioned that there are changes in 5 categories. As what I can tell there is even one change at one first placed candidate.

---

Last edited by Wonko; 2011-11-29 at 22:25. Reason: Fix typo.

Helmuth | # 814 | 2011-11-29, 22:37 | Report

SIGH!

Thanks Wonko for your explanation.

I'm preparing myself now for bed and have to talk a little bit to my toothbrush before. Perhaps I have a good idea for a meaningful reply this evening... or perhaps tomorrow. I don't know. I have to think about it a little bit before.

Zehjotkah?

mmlado | # 815 | 2011-11-30, 05:52 | Report

Let's make a voting about which voting system to use. Just joking.

This should have been decided before the voting. It will be a good lesson for the next... competition if there will be one.

The final decision rests with Zehjotkah, and whatever he decides there'll be unsatisfied people.

antman8969 | # 816 | 2011-11-30, 05:54 | Report

oh man!!! Just flip a coin and call it a day. This should have been done three months ago already...

zehjotkah | # 817 | 2011-11-30, 08:52 | Report

I haven't voted and I haven't looked into the potential results, either. I don't care *who* wins, I just care that there is someone who wins.

I (mistakenly) understood that only borda would solve the problem with ties.

After my last email to nemain

this is the last email I got from Riku:

I don't know how to explain them how to solve the problem with the ties.
As announced I still would like to use the STV system, because I think it is the fairest one, but I haven't thought about the ties.

As prizes we have:

1st: Nokia N9
2nd: Nokia N900
3rd: 50% of the money for one category
4th: 30% of the money for one category
5th: 20% of the money for one category

So we need a rating without ties for at least 5 winners.

---

Last edited by zehjotkah; 2011-11-30 at 08:54.

Wonko | # 818 | 2011-11-30, 10:03 | Report

As far as I can tell, neals method solves the "tie-issue". Similarly to the other method everyone can download OpenSTV, the *.blt files, and neals script and do the evaluation by himself. It is also possible to do this manually. I actually did this for one category to see what's going on and it seemed to work reasonably well (at least for this category).

With respect to the requirement of having 5 winners: in some categories of the results neal uploaded appear only 4 entries. But that's due to the nature neals script works (It only puts out n-1 winners and leaves the last one out.). In the case where only 4 winners are shown in neals results the 5th winner would be the one who is left.

Having said that. Would it be agreeable to simply use the results as published by neal (with the addition of the 5th winner in case of the "4 winner categories") and send these to nemain?

zehjotkah | # 819 | 2011-11-30, 11:36 | Report

for me: yes.

(but only if he adds the last one in each category...)

erendorn | # 820 | 2011-11-30, 12:15 | Report

As a non voter, non coder, contributor and someone who didn't look at the results, I support neal's solution, out of consistency with council voting (e.g., same winner for a one seat election).