Nick Sellen

I have a feeling the regeneration proposals/voting would have come to a better outcome by using the ideas outlined by @douginamug in (summary page)

There seemed to be many good intentions and ideas, but not the right structure to turn them into an acceptable proposal.

@nicksellen @douginamug

Interesting. Have you tried the Kanthaus approach or Ukuvota?

@Matt_Noyes @douginamug I'm peripherally involved in both of those projects. I think the tooling lags the theory still but a promising direction.

Sounds like it might get implemented in loomio too. See

@nicksellen @Matt_Noyes @douginamug

Neat! Back in 2013, when we made the first draft of Bylaws for, we proposed likert-scale style -3 to +3 score voting for voting on proposals, with requirements of *median* and *mean* minimums for different types of votes to pass.

Unfortunately, we didn't get it in action and don't have tooling for it yet. But I've still wanted to see the ideas in action (and we have more nuanced details in the drafts…)

@wolftune @nicksellen @Matt_Noyes I was just looking over the page before you posted this :)

Ooh, medians sounds complicated! I remember this being interested regarding the gruesome details of utility mathematics.

Multiplying negative scores by 3 and then calculating the mean has, so far, been fine for me/us.

@douginamug @nicksellen @Matt_Noyes is not fully in place and we plan to move away from multi-stakeholder to simpler structure, but I think you may find the ideas we worked out useful.

Median-minimum makes sure, e.g. that something fundamental (like a Bylaws change) doesn't pass with a majority opposed but just not as strong as the proponents.

*different* median or mean thresholds is like requiring majority vs supermajority etc.

More flexible than arbitrary ×3 idea…


median vs mean isn't about utility mathematics (we're skeptical about that stuff).

median just measures that a majority is above some level. It's not complicated.

So, for certain measures, could even say simply that the majority needs to be -1 or above so that you don't have a situation where the majority vote -2 but are overruled by a minority voting +3.

That reduces the incentive to strategically exaggerate scores without the way ×3 of negative scores weights negative.

@wolftune Yay! Someone who's interested in this. Will tag @Ultrademocracy since they have a stated interest in the topic.

Before I go further, did you have a look at the slides I made, specifically the ones about including 'status quo' and 'further discussion' options by default?

@douginamug @Ultrademocracy I was already reading the slides as you wrote. So far, I agree with everything.

Side-note: it took a while for me to buy-in, but I now support STAR voting *slightly* over score for electing representatives or other selections from multiple-choice options (as opposed to yes/no on passing a proposal).


@wolftune @Ultrademocracy Wow, strong image.

I've come across STAR, and on first look seems promising. I think a lot of the CES people like it too... hang on, you're a CES person! Ah yes, it all clicks together :)

I've been almost exclusively focussing on groups which are post-conversational (i.e. >8) but pre-anonymous (i.e. <150) As such, I'm assuming relatively high levels of cooperation and haven't focused on elections too much (which a law unto themselves...)

@wolftune @Ultrademocracy based on the position that 'keeping things the same' is in fact an option and not the-natural-way-of-things, cut-offs (whether mean or median) don't make sense.

As for 3X being arbitrary, well, it is :) But so is a 2/3rds majority, or a median cut-off... these things are socially intuitive, I think there are some studies, but I'm more into direct-testing ha.


@douginamug @Ultrademocracy I agree that the settings are somewhat arbitrary in that there's a judgment call for how far to weight the competing factors (e.g. status quo vs change).

"keeping things the same" as an option is a presentation detail. It can't be rejected at the same time as rejecting alternatives though. It's just just like any other option.

@douginamug @wolftune

The multiplication factor for the negative scores is actually the one aspect of your slides that I thought was weird and useless. I get what it's trying to do (emphasize avoiding really pissing anyone off) but it doesn't really do anything. It's logically equivalent to using an oddly staggered scale like 0-3-6-9-10-11-12.

@Ultrademocracy @douginamug

It is indeed *more* straightforward to weight against changes by simply requiring a higher mean score (like 1 or 1.5 etc).

That's independent from using median (which is for ensuring majority or supermajority support, if desired)

@wolftune @Ultrademocracy Do you mean:

proposal with highest mean score selected if greater than status quo by x, else status quo


@douginamug @Ultrademocracy

I don't yet fully embrace the idea that all proposals are choosing between multiple competing ones with "status quo" as an option. I mean, that can certainly make sense…

But my view of this sort of negative to positive range is strictly for accept/reject votes of a single proposal.

For selecting among multiple options, I lean more toward some sort of Proportional Rep, e.g. RRV or STAR-PR, with 0-5

@wolftune @Ultrademocracy Yeah, I'm not 100% sure about having the status quo completely without privilege, after all, it was already chosen so should have some merit.

However, giving privilege to the status quo makes an organization inherently conservative. Perhaps that's OK, perhaps not. Situation dependent I guess.

Roberts rules of order style, binary-choice majority gives no privilege to the status quo, interestingly :)

@douginamug @Ultrademocracy

I don't understand the RRO point. Requiring a majority approval to do something new and *otherwise* status quo remains *does* privilege the status quo…

per other comment: neutral facilitation has nothing to do with centralization, and, considered as an ideal (versus practical reality), it's great even in with just two people conversing — it has nothing to do with post-conversational, neutral facilitation is *for* conversation.

@douginamug @wolftune

A while ago I came up with something I called the "score bump-up method", which is a STAR-like runoff between the two top-scoring options. You look at each voter's scores for those two options and "bump up" the higher of the two to the maximum, without modifying the lower score. If it ever overrides the plain score winner, it's because the second score winner was less polarizing and more amenable to the people who didn't really like either of the top two.

@Ultrademocracy @wolftune It sounds good, but I realize my head is in a very different place...

I've been involved with a lot of grass-roots, voluntary activist-type groups.

They start small and are typically informal, or formally adopt consensus (unanimity of non-resistance)

They begin to face issues when they get bigger: cognitive/temporal/spatial limitations.

All my work/thinking has been in this direction. Good-hearted people, but too many to talk together.

@Ultrademocracy @wolftune You're completely right about the numerical equivalence being 0-3-6-9-10-11-12

I've been focusing on non-anonymous voting in groups of people who want to cooperate but are too big to chat things out (i.e. >8)

In these cases, voters risk social capital by using the scale as a 'rational, self-interested individual', instead of as intended.

As a group gets bigger, the factor would need to decrease and for very big groups, no factor at all 🤷 Make more sense now?

@douginamug @wolftune

I have a different mentality. I don't like systems that rely on social pressure for their basic functioning, and one of the most reassuring aspects of score voting is that it works fairly well even when all the voters are strategic.

Scrutinizing everyone's ballot for "strategic voting" is a dangerous road to go down, especially if you accuse people of being "strategic" for using the minimum score at least once.

@douginamug @wolftune

If social pressure is involved, it should be directed at the person who made an outrageous proposal to withdraw it before the vote even happens.

@Ultrademocracy @douginamug

In the STAR voting developments (I was around in early states before the name solidified etc), I grew in my understanding and shifted from seeing 0-5 scores in score voting as referenced from bad to good. I now realize that it makes more sense to embrace (and label) 0 = no support, 5 = max support. It's how much of your vote weight to give, relative only to the pool of options. I.e. "normalized" scoring.

Seen that way, there's even *less* worry about "strategy"

@Ultrademocracy @wolftune I feel like we may have a gulf in experience.

Have you been part of a small to medium sized voluntary group? If yes, how did they make their fundamental decisions?

At this level, social pressure is inevitable. Indeed, social pressure almost always exists: I believe it's a stronger force than the police most of the time. (I haven't read any Foucault, but I think he writes about that a lot.)


I realize this was to @Ultrademocracy

I agree completely that the idea of no-social-pressure is naive. Tyranny of Structurelessness etc.

Incidentally, small to medium-size groups I've been in always did fundamental decisions via consensus+fallback (and I've never been involved where it came to a fallback vote, though I know of cases).

Given human reality in small groups, I think *neutral facilitation* is key rather than a voting structure.

@wolftune @Ultrademocracy And the discussion can be fun! Or an all-out war-of-attirition ;p

With post-conversational groups, I agree 'neutral facilitation' is best. But that is hard, and a point of centralization... Still, I do it sometimes.

I'm trying to encourage consensus groups to use this positive/negative scoring, even if not to vote, just to get everyone heard and overcome the biases/privileges of conversation.

Sign in to participate in the conversation is a a coop-run corner of the fediverse, a cooperative and transparent approach to operating a social platform. We are currently closed to new memberships while we improve our internal processes and policies, and plan to re-open to new folks when that work is complete. [9/2/2018]