@Ninjatrappeur
À savoir, le livre « #MajorityJudgment:
Measuring, Ranking, and Electing » <https://mitpress.mit.edu/9780262545716/>, <http://zlibrary24tuxziyiyfr7zd46ytefdqbqd2axkmxm4o5374ptpc52fad.onion/book/1208115/a703c9>, publié chez #MITpress en 2011 par #MichelBalinski et #RidaLaraki, démontre formellement que le #JugementMajoritaire est l'unique méthode de #vote qui évite le paradoxe de #Condorcet, le paradoxe d'#Arrow et contre au mieux la manipulation stratégique.

TIL, during the last months there was a #Debian proposal and experiment to use #Belenios <https://www.belenios.org/> for voting:
https://www.debian.org/vote/2022/vote_001
That proposal was not accepted, yet I'm rejoicing to see this great algorithm/software being noticed by fellow libristes :)

On that topic, sad to see that #Debian is still using a buggy #Condorcet-like voting method instead of the mathematically sound #MajorityJudgment :\

#Math #MichelBalinski (1933 – 2019) #MajorityJudgment
> Chapter 1: Early Life
> Chapter 2: PhD at #Princeton
> Chapter 3: Academic Appointments
> Chapter 4: From Convex Polytopes to Apportionment
> Chapter 5: Ranking Wines and #Voting in #France
> Chapter 6: Early Influential People
> Chapter 7: Work in Districting
> Chapter 8: Impact on the Field of #OperationsResearch
> Chapter 9: College Admissions Problem
> Chapter 10: Retrospective on the field of #OperationsResearch
https://www.informs.org/Explore/History-of-O.R.-Excellence/Biographical-Profiles/Balinski-Michel

« 2019 State of #Haskell #Survey results »
https://taylor.fausak.me/2019/11/16/haskell-survey-results/
#Sad that the #tally is not a #MajorityJudgment (compare medians of grades, not averages of numbers).
I've published an #Haskell library to do such tally 2 years ago, but nobody knows, because I have not blogged about it I guess, because I have not (yet) finished™ the CLI and Web tools, because I was unable to rally support, because other people supporting the #MajorityJudgment want to do #Python, not #Haskell or #PureScript.

@paulfree14
Well, #MajorityJudgment respects the… majorities.

For each choice, its majority-grade is such than at least 50% of voters agree to judge this choice at _least_ greater or equal than its majority-grade, and at least 50% of voters agree to judge this choice at _most_ lower or equal than its majority-grade.

Using medians as such is actually what makes the #MajorityJudgment resilient to manipulations: to strategic or cranky votes.

@douginamug @nemoudeis @HygieneMentale

@paulfree14
Concerning not bothering anyone unless their isn't a resistance, #MajorityJudgment can be used continuously and dynamically:

- it usually sets a default grade of "To reject", thus unless for a given choice more than 50% of voters vote, and vote some other grade, its majority-grade (middlemost) remains "To reject".
- it ranks, but does not _ask_ voters to rank or compare choices at all, hence it is independent of irrelevant alternatives.

@douginamug @nemoudeis @HygieneMentale

@paulfree14
Yes, however be aware that #MajorityJudgment's authors recommend to not learn it from #Wikipedia due to its errors (and I would add: lack of mentioning the importance of applied criterions).

Indeed it's a tool addressing only a few concerns within the decision process. Which can still be undermined by other factors like #gerrymandering, #disinformation, <you name it>...

Still, it's the best collective ranking tool I know :)
@douginamug @nemoudeis @HygieneMentale

@douginamug
Concerning #voting, you may be interested by the #MajorityJudgment which is judged by its authors to be “superior to any known method of voting and to any known method of judging competitions, in theory and in practice”.

- comicbook (fr) : « Vous reprendrez bien un peu de démocratie ? » https://lechoixcommun.fr/
- textbook (en): « Majority Judgment: Measuring, Ranking, and Electing » http://libgen.io/book/index.php?md5=BF67AA4298C1CE7633187546AA53E01D

@nemoudeis @HygieneMentale