Proportional representation

Please explain?

Is there really evidence that STV is as good as proportional rep?

In state socialism, there could still be parties based on factions. There were factions in Democratic Centralism. Most liberal democracy parties are all just factions of the bourgeoisie anyway.

I would say that the flexibility of unaccountable parties actually prevents splitting. When they're unaccountable, they can just lie or constantly change positions to get more people in.

What type of disagreements?
politcal parties are liberal bullshit
Could you expand on these scenarios?

Fuck no, the party ina single party state should work as a fourth power not a private group of interest.

I am Greek and your pic triggers me!

It's actually more proportional than most of the existing party list proportional representation systems. It's less dependent on the way the vote is divided and does not discourage voting for smaller parties. I'm pretty sure a mathematical proof exists somewhere, but it's pretty obvious even without it.

The main argument against STV is a complicated, poorly scalable voting procedure, although, as pointed out, there are ways to greatly simplify it.

Australia used to have a party list based variant for its senate elections, but small parties kept winning seats thanks to it, so bigger parties went "what the fuck, we need to stop this" and removed it.

The next election, the minor party vote consolidated and they won more seats than ever, so there's that.

One product design versus another, one location versus another, one leader or philosophy versus another, working conditions, hiring, deadlines, hours, corruption, environmentalism, etc.

Maybe not political parties as such, but look at the history of amateur/free/open source projects online. Egos and "artistic differences" commonly smolder into splits and forks, plans often fail and have to be worked around, planning itself is often a fractious process, and communities bleed members or get flooded or attempt to recruit from odd places. And that's usually without the further element of peoples' livelihoods at stake, let alone the life-or-death decisions that a lot of meatspace activities have.

Scenario A) You are a cop, firefighter, or whatever, under a liberal democracy. You are really angry about something, maybe pay, maybe equipment, maybe regulations. The bureaucracy over you is unresponsive, and the voters are unaware of or indifferent to your plight.
Scenario B) You are a factory worker under socialism or in a coop, part of a much larger entity than your factory. You want better conditions, and are at the end of your rope, willing to resort to strikes or seizure of the factory. The broad majority of the organization doesn't think you have it that bad or doesn't particularly care about your concerns.
Scenario C) You are a member of a council, syndicate, communist party, or other socialist democratic body. You and some sympathetic fellows (possibly not all a member of the same subdivision as you) are developing a new policy proposal that not everyone agrees with at the moment, you must acquire and use resources to openly research and promote this idea separately from consensus.

Math.

When you vote for single-seat elections with standard plurality voting, a voter group that is bigger than half of all the voters can determine the result without even taking into account how people outside of the group vote, just by all in the majority group voting for the same candidate. Some of the voting methods (e. g. Instant Runoff, Schulze) that use ranked ballots have the feature, that if a majority ranks a subset of candidates above the rest, the winner comes from that subset (within the supported set, the ballots can agree or disagree in any way, they only agree in putting the set above the rest). This is called the generalized majority criterion (also sometimes called mutual majority).

Suppose single-mark ballots with voting for individual candidates is used for multiple seats, say three. If there is a group that agrees on a candidate, how big does the group have to be in order to get that candidate elected, no matter how people outside that group vote? The answer is: just above 1/4, because there can't be four groups with that strength. Likewise, for five seats the value is >1/6 and for 99 seats it is >1/100. There is a criterion similar to the one above for proportional elections, that groups of voters of a certain size who rank a subset of candidates above the rest will get a seat. And if the group is twice the necessary size for one seat, they can get two seats, if three times the size, it's three seats, and so on. That's Droop proportionality for ranked ballots.

t. Zizek
local democratic assemblies can be sufficiently alienating if you just don't get involved.

MMP has constituents and proportianal representation.

So this whole discussion is a false dichotomy.

youtube.com/watch?v=dhnLqAUdYKk

Any proportional representation system that uses some form of automated runoff system (rather than discrete runoff elections)–instant-runoff or "ranked choice" voting for instance–fails to overcome Duverger's Law due to tactical voting and results in the same two-party dominance in existence under first-past-the-post, winner-take-all systems. Australia is a good case study.

What is necessary is a proportional representation system through either approval voting or score voting. The only problem is that those may computationally intensive enough (proportional approval voting anyway) that they necessitate computerized voting, which creates a massive set of vectors for election fraud.

Fuck computations, use a random ballot system instead.

Gee, who'd have guessed.

I have no doubt in my mind that randomized representatives would be more democratic than the system in place now.

Not entirely without precedent:
en.wikipedia.org/wiki/Sortition

I asked in a previous thread, that I faintly remember an author who analyzed all sorts of election systems under criteria like fairness, representation etc., and came to the conclusion that it's impossible to satisfy all these criteria. But I can't remember his name. Anyone as any idea?

The only two functions of a party are to reach power and to hold on to power. It is, by definition, a group of interest, and whether it's private or not really doesn't matter unless you can somehow get most of the entire population to participate out of their own volition. If you want a fourth power, you have to create a fourth power with its own explicit mission and proper democratic and socialistic structure.

You might say "just because it has 'party' in the name doesn't mean it has to act like one", but looking back on 20th century history, well, it will act like one.

Isn't STV the shit one that just bogs down into electing moderate candidates that don't offend anyone?
(Which all systems do to a degree, but compared to MMP STV seems particularly egregious.)

Like we do today, people negotatiates or sgrees too disagree

I'm not a programmer so I am not used to the particularities of libre software development, I will atribute such conflicts to autism.
But getting real, I paint and do music, never worked with any people but I imagine things wil go like "Hey wanna jam" *they jammed* and they did because each one knew a common musical idiom and what part should their instrument fill

Never heard of a firemen strike but police strikes are a thing. either way all the scenarios you're proposing could not happen in a gift economy

making the party and explicit fourth power is what I want

Unless we're talking post-scarcity (in which case socialism/capitalism/monarchy/tribalism/etc are completely irrelevant), there are going to be serious disputes over how to get important things done using limited resources, opportunities for graft to take root, and lives on the line.

Even in a reduced-scarcity economy, major capital goods (not to mention human talent) will remain in contention between mutually exclusive exploiters.

Honestly, I think simply naming it a party will be enough to deviate its purpose. Yeah I know it sounds a bit superstitious and spooky, but names have power, now more than ever, in a postmodern, informationally overloaded world.

Some law. Even if you keep first-past the post, if becoming a candidate is based on population lottery, are you certain that will result in a two-party system?

And I certainly don't believe in a general tendency of proportional STV leading to duopoly. The devil is in the details: How hard is for a newcomer to even get on the ballots (many countries don't have a write-in option and that's an interface/nightmare anyway) and how much work is it for voters. If voters only have the choice to either provide a full strict ranking or support a few pre-made rankings (instead of a more sensible option like allowing truncated ballots with your first choice providing a ranking for filling in the rest) and if there are only very few seats and if there is media concentration, you can't expect too much. That doesn't mean there is some duopoly law about STV itself.

You are probably thinking about the versions that compute some happiness score for all possible seat allocations. A much more simple version is to allocate the first seat like in the single-winner version of the voting method, then reduce the voting power of those supporting this candidate, then allocate the second seat, reduce the voting power of everybody supporting that one, and so on. This isn't quite as proportional, but if the total number of seats is odd, it's still proportional in the sense that is analogous to Droop proportionality for ranked ballots: instead of a voter subgroup ranking some candidates above the rest, it's about giving them max support and min support to anyone else.

Kenneth Arrow, who just died. Other incompatible combinations have been found.

No. Plurality at large (7 seats means you can mark up to 7 candidates) or approval without re-weighting would do that.

Australia, Ireland, and Malta are the three countries that all use instant-runoff voting to facilitate their single transferable vote proportional representation systems in national elections and all three still manage to have total two-party dominance. The incentive to vote tactically under instant-runoff systems is so powerful that it ends up not better enough than first-past-the-post in the ways that really matter.

He said it was impossible but over the years eventually reached the conclusion that score voting was "good enough".

There is literally no point in having representative democracy anymore. With the internet and blockchain technology with encryption, there is literally no point in having a representative anymore.

You're going to have to convince an awful lot of people to dispense with the secret ballot and open themselves up to voter intimidation first.

We have the technology to allow you to vote anonymously with end-to-end encryption. Bitcoin has public ledgers but your bitcoin address does not contain your identity. The only way you can be traced is if someone who knows your identity also knows your bitcoin address. And then they do an analysis of your bitcoin address online to see where you are sending funds and such. This is how the feds caught Dread Pirate Roberts, operator of the Silk Road (online black market for selling drugs, weapons, stolen credit cards, etc.).

Instant runoff implies single-member constituencies, and is therefore by definition not proportional.

Of the three countries you mentioned, Australia does not have proportional representation (except for the senate, which does have a huge contingent of minor parties), and Ireland quite emphatically does not have a two-party system.

That leaves you with Malta, which hardly demonstrates anything other than it being Malta.

Nope. You can still have secret votes on issues instead of people, and even if you couldn't, a few representatives are much easier intimidated than millions of voters.

rangevoting.org/TarrIrv.html

If you were to have blockchain voting, you would have to have some sort of system where the issuer is blind to your private encryption key at the time of voter registration (where they verify your identity documents and register you to vote). So that other people have no idea that you are the holder of that private encryption key and can't find out your voting history.

Some people will be stupid enough to post links to their votes online on social media for narcissistic reasons. That can't be helped.

Instant runoff isn't proportional.

If you lose your private encryption key, you are fucked though. You can't vote. And if someone else gets their hands on your private encryption key, they can vote on your behalf. lol. There is no "account recovery" option with bitcoin. If you lose your private keys, you are fucked. You just lost all your money.

Read the title of the chart: Single-Winner Election Methods. Your opinion is not informed by real-world usage of re-weighted approval or re-weigthed range (as there isn't any). So you are not comparing real-world STV with that, you are comparing real-world STV with what you merely believe about these other methods without experience. There are some plausible claims by the range voting guys regarding the math and logic of instant-runoff and so on that have some purchase even without real-world testing, but these claims are based on the single-winner versions. You make a leap from the single-winner method analysis to what you believe about how the respective methods work out in their proportional versions.

Quote from the page:

It's nice to see you're still around election systems expert-kun. Do you think proportional approval voting is good enough for the real world or should third parties really be advocating for proportional score voting?

The former.

One issue that election-reform enthusiasts usually don't give enough weight to is voter fatigue. The reform-minded people are usually far more interested in politics and the act of voting than people on average, so when they discuss among themselves, they tend to have very optimistic expectations about how much people can be arsed.

Yes, that's the same reason I advocate for approval voting myself.

There's probably something to be said for recall, which would pressure politicians to stick to their promises and perform well. Basically, if a petition to recall a particular person/group in power gets enough votes then an election is held. Lenin said all officials should be subject to recall at any time

The revolution will come not in the streets or at the ballot box but in the fields and factories. With a powerful enough labor union controlling the means of production, they can threaten a general strike to force through political change and make the politicians answer to the workers.

If that is supposed to mean an election for the entire parliament, it follows that the bigger the parliament is, the more likely it is that some people want one person in it removed, so the frequency of elections increases. Unless you increase the threshold for triggering the new election.

If it supposed to mean just an election for a particular seat within parliament, how do you square that with proportional representation?