Thoughts on anonymized compos
category: general [glöplog]
I lost to Firg at one Function demoparty. That wouldn't happen in germany ;)
Maybe we're going about this the wrong way... Maybe we should de-anonymize everything? Make individual voting public. Wouldn't that at least deter blatantly down voting? Best of all that would give us data to research voting trends! Says a mathematician while rubbing her hands :D
that's potentially a very neat idea, at least if the data was used for sensible discussion instead of some weird drama crusades on pouet. (i'd still proudly self-vote, fuck peer pressure)
Quote:
Maybe we're going about this the wrong way... Maybe we should de-anonymize everything? Make individual voting public. Wouldn't that at least deter blatantly down voting? Best of all that would give us data to research voting trends! Says a mathematician while rubbing her hands :D
That is a horrible idea, tbh. I don't want to have to defend my choices to everyone and his hamster. In fact, I don't even think anyone has any business with whether I voted or not at all.
For the mathematician, I think making anonymous individual votes available would be enough?
live voting statistics would be neat tho, not personalized of course
ive seen some interesting charts so i guess its actually possible with wuhu?
ive seen some interesting charts so i guess its actually possible with wuhu?
It's also possible in PM which is where Wuhu had the idea from. That being said they both do little to solve the problem :)
Quote:
Gargaj: Come to think of it, we don't need to know the extent of the problem. Even if there are only a few that actively down vote, it can have a pretty big impact on the scores. A quick count of this years revision results gave 17 candidates for top 3 that had a margin of 25 points or less, and 5 of them had a margin of 5 points or less.
This is a point I forgot to address earlier - if we mean all 17 compos, then a part of them are anonymous and the part of them aren't, so what's our conclusion here?
Quote:
Quote:Maybe we're going about this the wrong way... Maybe we should de-anonymize everything? Make individual voting public. Wouldn't that at least deter blatantly down voting? Best of all that would give us data to research voting trends! Says a mathematician while rubbing her hands :D
That is a horrible idea, tbh. I don't want to have to defend my choices to everyone and his hamster. In fact, I don't even think anyone has any business with whether I voted or not at all.
That would drastically reduce the number of people voting.
Quote:
They are quite small indeed! The PC 64k compos at Revision for 2015 and 2016 are great examples:
I think most people will agree that the Conspiracy vs Mercury battles these years have provided us with intros of exceptional qualities, where it's close to impossible to choose a favorite. So with the race being that close you would think that the order would be what's tipping the scales, right?
However, in 2015 the top 3 places were the reverse order of the compo show (Conspiracy played third to last and won, Mercury played second to last and placed second and Approximate played last and placed third). This year, Conspiracy was played last and placed second, while Mercury was played second to last and won.
If anything, you would think that being played last is a disadvantage. Or that the order means less than you think for the eventual outcome of the compo.
I am not going after statistics here, maybe Mercury would have won this year with more points or not all all. We do not know exactly how voting is influenced. But I think we can agree that is has "some" impact.
Now imagine for a moment the 64k result would have been the other way around and Conspiracy would have won by 2 points. Since the last position is assumed to have an advantage, I am sure Mercury (or some supporters) would have started a discussion about the order. Using (controlled) randomization would put you in a safe position here.
This does not make sense. Both Mercury's and Conspiracy's prods were highest-quality winner material that you can already argue that the order as it was shown already was randomized in some way.
Quote:
That is a horrible idea, tbh. I don't want to have to defend my choices to everyone and his hamster. In fact, I don't even think anyone has any business with whether I voted or not at all.
First of all, just so it's clear, this was not a suggestion! It wasn't more than, "hey, we haven't explored in this direction yet!" *wandersofaimlessly* Anyway, I don't see why you would need to defend your choices. If anyones hamster asks, you could just tell them what you like is you choice and not a discussion. The point is that people are far less likely to do something they know isn't right when some might be watching. Of course I see the problem with added peer pressure. Some might find it hard to chose freely in fear of being mocked in some way.
Quote:
For the mathematician, I think making anonymous individual votes available would be enough?
Of course you're right! Some additional attributes would be necessary though, depending on what you want to compare.
Quote:
I am sure Mercury (or some supporters) would have started a discussion about the order.linky
yeah, because we totally would ruin one of the most awesome days we could ever have by complaining about our friends taking a deserved first place. complaining about some numbers. yep.
I prefer the theory that conspiracy lost because some fucker didn't vote - because someone's live voting system prevented him from it (by closing the compo before you could vote for the last entry) :D
Would be interesting to see not just the sum, but also the number of votes for each entry (/the average vote). For instance it's obvious that saturday compos receive many more votes (probably due to people catching up on voting sunday afternoon).
Would be interesting to see not just the sum, but also the number of votes for each entry (/the average vote). For instance it's obvious that saturday compos receive many more votes (probably due to people catching up on voting sunday afternoon).
Quote:
This is a point I forgot to address earlier - if we mean all 17 compos, then a part of them are anonymous and the part of them aren't, so what's our conclusion here?
This isn't a comment on anonymity, but whether there's a "real" problem with name voting. I wouldn't conclude anything, but it suggests to me that even if there's only a few people going about down voting, it will have an impact on the results. That even if it's not a big problem (not a lot of people doing so) it can easily have consequences, and be a "real" problem.
That brings me to an interesting question that I only just remembered - at Solskogen last year we had an experiment that the voting range wasn't 0..5 but -1..5 so that you could explicitly express negative feelings towards prods, since there wasn't an option to "abstain", so to speak, since voting 0 was also the lowest possible score. After looking at the averages I found out that there were cases of "vote -1 on everything in the compo, vote 5 on a specific prod/prods".
So the question is, even when you have some sort of red flag system for blatant vote abuse, what's the correct course of action?
So the question is, even when you have some sort of red flag system for blatant vote abuse, what's the correct course of action?
Toss the extreme outliers? If you can't vote sensibly...
Less systems and rules please. It's not a nail, put your hammer away.
By the way. It's better to -1 vote only those that you think are your most serious competitors, not everyone! Even a drunken scener can figure that out. If you aren't given the possibility to down-vote, then up-vote someone who you think might overtake your competitor but not you. Also, lobby for your entries.
By the way. It's better to -1 vote only those that you think are your most serious competitors, not everyone! Even a drunken scener can figure that out. If you aren't given the possibility to down-vote, then up-vote someone who you think might overtake your competitor but not you. Also, lobby for your entries.
Make no mistake, I'm not advocating anything - as I said before I don't think it's a problem that's serious enough to act on (which was my opening statement), but it's definitely worth having a thought about.
stream had a -5...5 point voting system. I kind of liked it, though it might feel too blunt for some peoples tastes.
And as a result Stream had a lot of compos with very little prods that ended up with positive scores, I think even some where all entries were in the negative :). Not that it really matters that much, but some people have said it was discouraging.
(Yeah, case in point: the 2014 photo compo for one)
Quote:
yeah, because we totally would ruin one of the most awesome days we could ever have by complaining about our friends taking a deserved first place. complaining about some numbers. yep.
I am not pointing fingers at anyone. If I sounded that way: Sorry!
But everybody has his fans. And sometimes some fans tend to become fanatic. You cannot control that.
Quote:
(Yeah, case in point: the 2014 photo compo for one)
That's depressing. :(
And definitely discouraging.