Number of users in a group
Which is the affordable number of users wich make discussions work? Do we need a limit? Usually tools gets overwhelmed by comments overdose.
Nice : )
Ok it is removed. 🔨
There is a topic floating around about how larger Loomio groups share their learnings, which I'm following and hoping to take some kind of action on when the time is right.
Hi Robert, the FAQ text was introduced again :(
Too many users currently cause problems with the amount of comments. E.g. look at the introduction forum. I simply can't write there because the comment field is always getting moved down if I scroll down.
Is there no way to sort comments new to old?
You mean an option which allows users to sort comments so that the newest comments in a thread are at the top of the page, right?
I think that would be somewhat helpful for the problem you're indicating.
Another issue with busy threads is that it's hard to follow who is replying to who. One thing which could help tons IMO, but may be hard to code, is if Loomio could create visual navigation clues to link "Replies" to the comments which they're replying to.
Each comment has a unique id. It should be as simple as making an "a" tag with that as the href/url. Though you may want to use the whole url in case someone copies it.
Exactly. Actually I think it is a grouping issue - comments threads by topic would be helpful.
Coursera forum with thousands of participants suffer a similar fate, so hard to scroll down to finally make your comment. Comment replies there are indented, that is helpful at least.
Another aspect: look at chat applications Whatsapp, FB, etc. If you have more topics in the chat, you're lost. People often forget to answer questions that are just a few lines above if some other topic is also discussed in parallel.
I just reviewed it.. The number is not a technical limitation, but rather a very rough indicator about what kind of processes we've seen be successful with the tool. It's more about an order of magnitude than the number itself.
I would back that up by saying that you should consider designing your engagement around groups of manageable, knowable numbers of people. Depending on the use case it can be a good idea to divide bigger groups into smaller ones each discussing the same topic, the outputs of the discussions are then reported on to all members.
But there is no hard rule to this though, and there is certainly no software defined limit on the number of members that a group can have.
I'll bring up changing the FAQ with the team tomorrow.
@josircardosogomes.. Did it? I can't see where.
@josircardosogomes .. sorry about that. Fixed now.
BTW, it must be out of the translations too. However, since the help section isn't in transifex anymore, it'll take a little longer to be fixed.
I've taken it out of all the translations of the help manual too.
🚀 that was fast : D
Ability to sort newest-first sounds useful. Also threading . . . many web setups for managing discussions allow you to opt between threaded and unthreaded, as well as to expand and collapse groups of replies to a post. All that stuff is probably useful.
In the long run, unwieldy group sizes leading to unwieldy conversations is going to happen. Rather than solving that through delegation, I think it would be better to solve it by dispersal--hiving off manageable sub-groups to deal with different concerns. Keep things horizontal rather than vertical.
Thinking of which, it seems to me that Gaspar's proposal for sorting and the discussion emerging about threading really constitute new, separate proposals and discussions from what this ongoing one seems to be about. And at that, proposals that are rather more concrete, useful, practical, and decidable than what was previously under discussion. Should we start something new, or perhaps a couple of somethings, to deal with them?
New feature: convert comment to new topic 😃
That also sounds handy.
I think we should go beyond the plausible limits of a Loomio discussion and face the real problem of a social debate with hundreds of millions of people involved. The construction of such a debate, is possible, would completely change the paradigm of the social narrative, giving (not returning) control to the people.
I have been working for a while in this concept and I think that we should have a sytem by which opinions and ideas are weighed by the number of people who support them. This is not the same that likes or dislikes, because if you are reading through the thread a contribution with one thousend likes takes as much readers attention as the following, may be with the same amount of dislikes.
There are a number of concepts to be explored. The first one is the grouping of people by afinity with respect to the idea at stake. Their position can be discussed in subthreads. The main thread is build by the agreed subgroup opinions, be it by consensus or delegation. The length of the thread would be limited to a size proportional to the number of group members, and each subgroup would have a corresponding proportion of "words" in the thread.
Subgroups reaching consensus would take up more of the debate, truly representing the formation of majorities. Groups would then tend to negotiate their position with other groups in order to be heard by the rest. Ultimately this would lead to the construction of the social narrative, including discrepancies.
One of the ideas I find most attractive is a system of delegation in which the people who delegate, empower someone, will be informed automatically every time their voice or vote is used, and they can withdraw their support. I.e. delegation implies trust, but this system is treason proof, because ultimately is the user who is in control.
At any rate I believe THIS is the social algorithmic problem that we need to solve in order to achieve true democracy. How can we ALL discuss constructively? How can we make decisions TOGETHER?
Love this mode of thinking.
I had an epiphany with @joshuavial and @alanna the other night and it is that you can split delegated read and write. Practically this would be the ability to delegate to someone to listen (read) and report back to you/the group and than speaking (write) for the group in a representative way.
Why this matters: i think (please debate this point) there is a dependancy on delegated reading from writing. Ie you cannot delegate your vote without the ability to recieve information back. Where as i think you can have delegated reading. In fact though functioning with a lower trust point i think this is what journalism is/used to be.
For example i am very keen to delegate the "following of" some loomio threads to others with the expectation that they will provide the information i need. Obviously i would like to delegate my vote. But that can be phase 2
I believe delegation is essential in order to achieve large scale democracy. Write delegation implies de-facto read delegation, since the contribution that anyone may make on your behalf must be based on a previous read /documentation. The bottom end question is that in a large system no one can read everything, so, more than read delegation what you need is systematic REPORTING.
If you imagine a topic being discussed by one million people there would be a large number of subthreads (subgroups) that no one would be capable to keep up with. There is a need for read delegation, for a way to sum up the content of the subthread discussion and let the key points come up to the surface and become visible.
Now, a reader who is not active in your thread cannot trustfully delegate on the reporters, because the to lack of time required to check on them. However, active members can control the reporters without effort. They read the thread, so, they can at a glance approve or disapprove the report or summary. May be what we need is a “thread summary” function, where summaries can be exposed and these summaries be voted. Delegated writers could contribute all the votes they have been trusted with to support or elaborate these summaries.
If you imagine a tree like subthread structure where the top trunks /branches, are only elaborated by delegates with the strongest support you would see the most popular ideas coming up and create the social narrative upon which ground laws and democratic decisions.
At any rate, this may not be the solution, but it is definitely the problem.
Dear Daniel, we have been working on this theme at our collective, Guilda. We are working on a prototype that we call Entrust. It addresses the point of exchanging delegation for curation, within circles of trust.
Thinking about reporting and summary features in Loomio is really interesting for me, thanks for taking the time to put your thoughts into words!
I agree that delegation, in some sense, is important. But when we think of "delegation", we tend to think in terms of people at "upper" levels having the power to decide things on behalf of people at "lower" levels. The people at "upper" levels get to decide about things with broad scope (on a broad range of issues), and people at lower levels can decide either nothing, or get to decide about things with only local scope (on a broad range of issues).
That, I don't think is necessary. The problem is that there are too many things for everyone to decide on all of them. The problem is NOT that only a special breed of decision-maker person should be dealing with the big things. So it can be solved by distribution rather than some sort of levels, which become a form of hierarchy no matter how well-vetted.
So if you've got a really big organization using Loomio-style decision making, you'd have a bunch of groups handling different aspects of organization-wide policy. Most people wouldn't be in most of them, but everyone would be in a few, and there would be some overlap. In a really big organization (nation-state?) there would be sub-groups nested within broader groups, going more and more specialized. I don't think there's a need to keep everyone constantly informed about all the decisions they all make--leads to information overload, again, and people ignoring most of what's in their inbox.
But you do need a mechanism to stop little decision-making groups from doing whatever they want and becoming little fiefdoms, bringing unaccountable power in the back door. I envision it working vaguely like this: First, as usual for Loomio style groups, membership is self-selected, so if a group starts doing weird stuff that impacts the rest of the organization a bunch of people can join and get a voice. Second, possibly there should be a certain amount of duty to be in a few groups randomly, so there are always a few people in every decision-making group who aren't there because that's their special interest, and so have a different perspective. Third, there should be an option like "blocking" which says a decision is too big and its impact too broad for the group to make it on its own, and so the proposal should be kicked out to the broader organization or at least a bigger subset. This should only take a smallish minority to activate. The principle then is broader, larger groups override narrower subgroups. Egalitarian principles are retained without everyone drowning in information overload.
This still brings us back to the need for, some of the time, decisions to be taken with a whole lot of people involved.
With respect to delegate, I do not think one should get tangled with the semantics. A delegate is also a representative, some one who acts on behalf of, and it is in this respect in which I use the word. However, the second meaning is quite interesting, because I believe that the decision makers should be understood as people who work for the community, for the people who have delegated them with their share of power, and hence endowed with the ability to decide. In my head the voter is "top" and the president is "bottom" of the decision making power piramide.
I totally agree with your idea of the subgroups. In particular I like your idea of kicking a problem out to the larger group, which I had not think of.
One of the key points for me is that in a nation state size group, most people do not want to spend much of their time in the running of the "res publica". However, they have the right to, and will eventually want to, have a say in a particular issue. For the rest of the time they can either:
1) delegate in absentia, i.e. trust that the group of people dealing with the subject are reliable
2) delegate explicitly, i.e. chose to be represented by someone who:
a) wants to spend more time on “politics”
b) has an alike point of view in that particular subject.
Indeed, in a really large group you can envisage a tree like structure with subgroup and subgroups of subgroups which may allow everyone to have a say and be active only as much as they want.
On the other hand, this introduces the fact that you should not need to chose among parties all of which you may disagree partially with. You could chose a different party, individual or group in every different part of society.
Oh no i just made the new thread https://www.loomio.org/d/WnsJ3SRN/delegated-read-write-for-groups-large-scale-democracy
Fascinating perspective @purplelibraryguy ; thanks for sharing! I probably agree on 90% of the stuff you said.
In contrast-- not necessarily contradicting you-- I'm quite critical of "collective identity" per se, organization-wide policies, and the ways that pepole use pooled resources. I think most shared objectives can be pursued in what I call an "intercommunity" (inter-networked) way. I think big groups are really important, but should be as simple as possible. (Probably, for the same reasons that you advocate delegation.)
I suspect that while we're defining or identifying things differently, the functional nature of the sorts of things we're thinking of is not so far apart. I myself am a fan of decentralization--but there are things that need to be thought about on a broader level. It is a bad idea for every community to have a different gauge of railroad track. It is a bad idea for ten different communities in a watershed to have ten separate ideas about water use. So on and so forth.
Well said.
Making a distinction between the tool and the method I think is very important. It is not to say we cannot have an opinion on how the tool we create should be used, rather that others are not "wrong" for using it differently.
I also think we are moving into a conversation on some sort of delegated democracy rather than "number of users in a group" and think this would be better in a new thread. So am going to start one and reference your comments here.
Thank you!
But I will wait for the FAQ to be modifed. After the FAQ modification, I think they will be convinced :)