Agree with statements instead of adding new

I think the original idea is a pretty simple implementation, if we allow users to select a 'parent' vote when they cast their vote, similar to the way a comment or a group can have a parent. Then, we could also group them by a parent vote, assuming that all of its children 'agree' with the parent. By clicking on 'Agree' next to a vote, we take you to a new vote form, with that vote set as your parent vote, and a matching position (agree, disagree, etc.).

So I'm biased because there's a nifty, elegant implementation. I wonder a little about how many people would derive value out of it, though.

I do agree that as we look toward supporting slightly larger decision-making bodies, we'll want to take a harder look at clumping together different 'factions' or like-minded viewpoints.

As @madsringblom mentions, "adding more complexity always raises the bar for participation". This suggests that added complexity will ideally be optional.

As @jameskiesel notes, supporting "larger decision-making bodies" will require the ability to clump ideas together. I'd add that the clumping need not be by "‘factions’ or like-minded viewpoints", but rather by idea thematic.

What I mean: in Jame's comment immediately above, there are 3 paragraphs which map out to 3 ideas. I might "like" 2 of them, but not the third. So what does the "like" tell us?

Instead of "liking", I'd love to add separate content on each of these 3 ideas. I'd like to see a form of "threading" either by indentation or by adding meta-keywords to paragraphs so I can see all the comments associated with a certain keyword.

But if we want to really support for formal bodies, I submit that we will want to be able to support formal argumentation methodology (see http://en.wikipedia.org/wiki/Argumentation_theory). For instance, an given post can be a Warrant, which can be substantiated by Evidence, countered by a Rebuttal, supported by Data, etc. These components of argumentation help parse the interchange and force the user to be clear about the nature of his/her contribution.

This would be better than a general "like" because what is being liked, and why it is liked, is very ambiguous at best.

@jeandaniel I need to delve into that link but supporting argumentation theory sounds like a great direction to investigate. It speaks a bit to my question of how Loomio.org defines a decision.

@BenBurton About how to define a decision. That would depend on what Loomio can support AND on what a group has agreed on as a norm.

For instance, a group may decide that majority vote is good enough. Another group may decide that they want full consent. How a group decides to make decisions is a key aspect of group formation.

Loomio could help that along by having a field "Proposal Acceptance" that would default to "Majority Vote" but would also offer other variants, such as "Full Consent", and including an "Other", where the proposer can write in a group-specific decision-making norm. It wouldn't be tied to any other logic. It would just be an information field to make the norm explicit.

@jeandaniel Well it kind of goes without saying that if a group decides to use Loomio then they've agreed on that as a norm. ;)

@BenBurton Well, I guess the decision to use Loomio is certainly part of the group norm. But beyond that, Loomio doesn't constrain the group to a particular criterion that automagically determines that a decision has been made.

As a proposer, I can Publish an Outcome even if not everyone has voted or supported the Proposal. My previous post was to explore how Loomio can help the group define for itself how it wants to use Loomio in the group.

Hmm, I definitely like this idea but am wondering if there's too much overlap between "Like" and "I Agree"/+1 or similar.

@madsringblom Are you satisfied with the "Like" functionality? Is that close enough?

I really like this idea. In order to scale loomio to larger and larger organizations, there needs to be ways to aggregate opinions. I think this is a good first step.

I don't see "Like" being useful at all. In general I think the discussion panel is too divorced from the actual proposals. You can't easily see who is making arguments on which side of the debate without reading over the entire list of comments (which doesn't scale).

@dnephin I liked your comment, because as yet there is no other way to "support" or "endorse" or whatever else people may want to do with it.

Democracy.os have a mechanism for rating if the comment "contributes to the debate" or "It brings the debate" by which I guess Google Translate means it detracts or argues against the proposal or discussion topic.

This doesn't address long posts with multiple "points", "facts", "comments" that other users may want to agree with or promote mixed in with others that do not.

I have been giving this some thought but still not any closer to anything that really addresses this.

I think it is definitely a challenging problem. Before I heard about loomio I had done some work on this project, which had basically the same goals as loomio.

My idea for solving this problem (of being able to scale discussion making) was:

• keep the discussion and the "decision" together, not as separate things. Every comment had to have some "position" (support, agree, challenge, disagree, flag), which is very similar to the proposal portion of loomio.
• all posts would be responses to a specify piece of data or sentence in the proposal. Allowing people to disagree with some portion of a proposal and agree with other portions of it.
• handle some of the complexity by exposing all of the posts as more of a graph instead of a linear list of posts. So every post could have comments on it, forming a directred graph.

The idea with the user interface is that instead of seeing a chronological list of posts, or something like reddit where you can see the nesting, you would instead look at posts in the context they belong. So if you clicked on a proposal, you'd start by seeing some posts for each position, and if you click on one of those you'd see the parent context, and all responses to the post, and so on down the graph.

Of course all of these add lots of complexity (to both the process and the user interface), and probably too theoretical and not designed out of organic use. I still hope that some of these ideas may be useful to loomio in the future, but we'll have to see.