Some comments on the governance model in the figure.
I do not see much reason to give administrators of servers so much power. I am admin myself of a small instance. That is not much work. Maybe because the users are nice; I never had to moderate. There are people who self-host an instance with only themselves. I see no reason to treat them differently from normal users.
Moderators of large instances put in real work and have an interest in building the Fediverse in a way that makes their work easier, while keeping their herd happy.
Equating accounts with people is not always accurate. I legitimately have four accounts, that is not being a sock puppet. One as human; partially in German. One as scientists; only in English. One for an open science project. I also maintain a feed, which is work in my case, but in other cases will just be a blog feed. How do we handle feeds? The open peer review system I am working on has one AP account per article, there are humans doing the peer review, so it is not a bot or a blog feed, but there is no one to one relationship between people and accounts.
A government that does not distribute goodies somewhat proportional to power will be toppled. Developers have well above a third of the power, giving them less than a third is asking to be toppled or ignored.
It looks as if the current system would debate any question brought up to by any user. That will be much more than anyone will be willing to handle. There must be a prioritization. Especially if we work with randomly selected people; part of citizen’s assemblies is a long phase of people informing themselves and consulting experts.
Asking people to register for the lottery will lead to strong self-selection (oligarchy). Is it possible to make a census of the Fediverse? Then we could invite people who have been selected by the lottery, which greatly reduces self-selection.
Good points, some have been answered all ready some need some thought. Will post replay soon
We don’t decide who should get the voice from each instance - this is up to each individual instance to decide in their own way who talks for them. We just use the admin check to tell if the person registering the account is a “representative” of the instance - we can open this up to mods as well. So if instances wants one of their user accounts to be the voice then they would have to bump them up to mod or admin.
The idea is to ALWAYS moderate as much POWER downwards to federate responsibility, fedivers native styaly.
The would be other automatic checks
- was the instance online last year
- is the instance used - check for recent posts etc.
- am up in the air if single user account instances count - this could be an option that can be chosen by the body - I would start with default ON that they count.
- ideas for more #KISS codeing approaches?
Then the is the human community - would then be user/stakeholder flagging as an option to question if an instance belongs. (TOS)
When an instance registers it would pop up on the body activity feed before it becomes a member giving people time to flag it/discus it if needed.
At the #OMN we have to work quite hard at moderating and removing SPAM and dealing with right-wing attacks on our 3 instances, so it is work - the instances are the WHOLE of the fedivers so deserve a voice that they currently do not have.
We simply ignore this issue, if you have multable USED accounts and be a instance admin and run a support group or two you will have multabe accounts in the lottery.
AGEN if people start to take the piss and game the system the human flaging comes in to mediate this issue. Secondly if few people are taking up the roles the solution is to be more human to get more people involved - this will dilute the problem.
- If this controlling few get shitty sighn up more people then flag them out - its politics and if everyone gets shitty the community can solve this issue the same way.
The solution is ALWAYS more people - the lottery will shift bad groups out if fresh people of goodwill join.
We keep it #KISS one user is one activertypub account with basic checks to see if it is active and human then relay on flagging for blatant abuse.
An example the lottery might choose 4 of your accounts as stakeholders in one go… then it’s up to you to resign 3 of them, so 3 new stakeholders can be choicen - and yes you could keep them all as sock puppets, but you would likely be found out and flagged to be removed so not a good long term thing to do.
My though is we don’t HARD code this and keep it as a tradition, but this would be up to the group, we just set the default.
That a good point about goodies - the group can setup a treshery group using openclective or something less NGO and then use the tools to provide funding and pay expesise as needed.
The instances are NOT the developers - the codebase have limited say as affiliates - so mastodon would have the same voice as any other code base that jumps the basic criteria and gets excepted by the body as being a part of the fedivers.
This body is mediating the shifting of power from the .01% to the 99.9% who make up the fedivers - it’s a compromise that the body itself can decide the balance of this shift.
WE JUST SET THE DEFAULT and provide KISS tools and structure.
Yes the power is not the normal power of courts and police… this might or might not be an problem.
Any user can put in a proposal that would appear on the activity stream of the body - its then up to the body to actually do something with it or just ignore it and let it slip off the activity feed out of site.
In this way its LOSSY.
If the body thinks that a proposal has merit its past to a group or a new group is formed, and it works its way through the process. The groups do the “long phase of people informing themselves and consulting experts.”
SPAM would be dealt with by flagging and standard fedivers moderating tools - remember the whole steam is based on activitypub so is native to the fedivers.
DEV talk The groups can be coded in two ways, we can create a new forum codebase or we can do the whole project as plugins for an existing one like discourse.org which is likely a better choice - this is up to funding and coders.
Will let an old dead imperialism answer that one:
‘Many forms of Government have been tried, and will be tried in this world of sin and woe. No one pretends that democracy is perfect or all-wise. Indeed it has been said that democracy is the worst form of Government except for all those other forms that have been tried from time to time.…’
Winston S Churchill, 11 November 1947
May need to highlight - am using a bit of abstract language as this codebase is not JUST a tool for the fedivers it can be used to democratically run any structures that has stakeholders.
An example let’s look at it running our local street market Chatsworth Road Market - Chatsworth Road
The stakeholder are each stallholder - put in the stall registration number.
The users are people who shop at the market - just sighn up.
The affiliate groups are the local counceal who licence it, the neo-liberal events company who manage it, the local shop owners association who wont a say.
In this case the codebase is a progressive web app and app ifyed with a rapper to run on mobiles.
Then all the issues and running of the market can be moved from the .01% ( the event company and counceal that gives them the licence) to the 99.9% who actually run the market – yes the would be a political fight which would likely end in a compromise – the power might end up 40/60% which is still a good outcome.
This approach and codebase will scale sideways – lets look at this example you can have streetmarkets governed city wide – in this next level each of the markets becomes a stakeholder the users are users and the affiliates are city wide orgs and groups. Rise and repeat, the whole thing works #4opens
The instance stakeholders is one admin/mod account registers the instance, after that they will get a message that instance is already registered. Important to be clear “we the codebase” play no role is HOW an instance decides who represents them - power is federated down words at every opertunerty.
It’s a democratic body so if one big instance pushes all their member to join then the members of that instance will have a larger say. The solution is grassroots the other instances get their members to join up. In democracy defining the bounders of the “body” is a very hard thing to do, we simply avoid this.
Yes, If there are few registered instances the effect of this will be to encourage people to set up and run more instances as this will be the shortcut to get a voice in governance - as you say a good outcome.
The shaping of the “body” comes from a long history/expirence of horizontal activism. In this the is an unspoken tradition of “those who do the work have more say” .
The bulk of the voice is those who run the fedivers - the people who run/support the instances. The people who build the tools also get a say, pretty much garrented by their smaller numbers as do support orgs and events. the users who will be spread widely get a say - but this power is diluted by the much larger numbers involved, but their voice is there and strong to bring in need fresh views and agenders.
The current plan is #4opens to build a production/coding team, go for funding, expanding the coding team, build the tools, test the tools, roll out this test to a real user base - likely not the fedivers to test.
While these last stages are happening we can try and ratify it though some process on socialhub or just roll it out #permissionless and see if people use it, both are fine by me.
Important largely what am describing here is simply the ideas for the default config of the “governance” codebase.
The actual function would all have sliders - you want to change the proportions or number of the stakeholder groups: add groups and slide here. It’s up to the body members to use the internal tools to adjust these as they think appropriate #4opens
A lot of this conversation is about the power of the default (bad article Default effect - Wikipedia) the actual running might end up looking different for different groups.
I am happy to discuss these guidelines and use them as inspiration. But there are several ways to skin a banana and some of them are a little better proven than others. Starhawk & al. did a thorough and long-term work (practical as well as theoretical reflection), developing social permaculture into an existing ecosystem per se, not a reaction to toxic processes. There are communities out there who we can learn directly from, on a very pragmatic level. Reinventing social networking is enough challenge; we do not need to reinvent the wheel to boot.
Yep, the is a long history to this stuff, when spiky and fluffy both push/build social change/challenge things do happen. We are putting this history into code:
" Distilled grassroots radical governance is a good fit for the fediverse. This working practice comes from 30 years of building from The Tyranny of Structureless tick box list "
Code being quite “anti-human” this is an interesting challenge, how to get the humane “mess” in a coding process that is based on being “exact” and in control. What’s your thoughts on this?
Humane is not mess. It is complex and emergent. I would recommend coders to start from acknowledging they are humans as well. As Starhawk says:
We each carry the imprints of our early experiences, and often respond to current situations with the negative patterns of the past. We hold onto painful memories and anticipate future hurts. When we come together in community, our own needs, goals, and communication patterns often clash.
Moreover, we are embedded in larger systems that do not encourage beneficial relationships. Our overarching economic system sacrifices the good of people and the earth to the goal of achieving short-term profits. It maintains itself by fostering systems of prejudice and exploitation—racism, sexism, classism, heterosexism, ageism, ableism—all those constructs that separate us and elevate some peoples’ good over others. Those systems affect us deeply, often unconsciously, no matter how much we might deplore them and struggle against them.
Also, one of my favourite simulation says:
What the game is, defines what the players do.
Our problem today isn’t just that people are losing trust,
it’s that our environment acts against the evolution of trust.
That may seem cynical or naive – that we’re “merely” products of our environment – but as game theory reminds us, we are each others’ environment. In the short run, the game defines the players. But in the long run, it’s us players who define the game.
So, do what you can do, to create the conditions necessary to evolve trust. Build relationships. Find win-wins. Communicate clearly. Maybe then, we can stop firing at each other, get out of our own trenches, cross No Man’s Land to come together…
If there is one thing I would like to contribute to the whole governance conversation, it is that we stop separating people of fediverse into this-or-that group of stakeholders, those who moderate and those being moderated and whatever.
If we take responsibility to design and build the environment, we need to build it for everyone. Empowerment, equity, beneficial relationships – this is our goal. Focusing on cohabitation, communication and cooperation, instead of shooting bad guys. Being practical: the goal is not to design an environment to make everyone happy. More workable is to spread unhappiness equally, so nobody carries more burden than others.
We are not gardeners. We are plants, trying to run our garden, That’s the root of it.
At the moment the main issue in this thread, I feel, is that we are all in agreement wrt the fundamentals of the concepts, but are repeating these points over and over to each other with different words.
What we have defined here and in some of the related threads are a whole bunch of basic principles. Though we must not create tech for the sake of tech, these principles are there as first inputs to avoid that. And the translation process can start.
Repeating more of what we already know (click to expand)
Other than that, if I look at the real world, then it is in fact complex mesh this-or-that group of stakeholders. And that is perfectly fine. If your daughter attends school, you become a stakeholder on school policy. And if you are a coach of a hockey club, and children don’t become members anymore because school requires them spending their free time on homework, then you are a stakeholder in both school and sports club ‘communities’ and raising your voice to change governance.
Similarly moderation is also everywhere in real life. The ‘covid assistant’ in front of the supermarket saying “Wait a minute please, there are too many customers inside now.” moderates you. So does a traffic light, or ticket you drew when waiting in a queue.
We already have lots of “normal” power politics and invisible hierarchies in tech http://hamishcampbell.com/2021/04/24/governance-in-open-source-projects/ the plan is to mediate these existing issues with trust building democracy in a NATIVE fedivers way.
I know Starhawks work for a long time on the ground, its a good fluffy path to social change.
The gardening metaphor is a good one.
An important word that needs some thought is “permissionless”
The body is made up of three different, balanced groups:
- Stakeholders - the people who do the work, who run/mod the fediverse
- Users - the people who use the tools/services, who use the fediverse
- The Affiliate Stakeholders - the people who commit time to support the work of the fediverse
This is a very broken web we live in so let’s clarify issues. The names can change, they are placeholders
Anybody can become a stakeholder, in the case of the fediverse this is setting up and running an active instance - could use mastohost for the less technical to do this or a home hosted instance on a old laptop.
It’s simple if a user wont to become a stakeholder setup and mod an instance.
Users are self-explanatory, they buy in but don’t have time or focuses or inclination to run a part of the fedivers.
The Affiliate Stakeholders are a little more complex and are thus treated differently, it’s up to the body itself to decide if the play a active and useful role.
Nothing in this is top down, nothing is elitist, nothing is discriminatory, nothing is undemocratic. Its #KISS and looks safe to the “normal world” while at the same time being native to the fediverse and its roots.
All the coding is #4opens and based on #activertypub
We need a grant to make this happen who is helping to write the grant app.
Starting work on funding for the fediverse #openweb apps:
Online governance - openwebgovernancebody - Gitea: Open Media Network need a non-political implementation maybe from Europe, the example I use is a street market? Then we have the fedivers
need 3 default configs - closed/normal/open
we need a few mainstreaming fluffy examples of use.
Home - MakingHistory - Gitea: Open Media Network
Home - Open-Media-Network - Gitea: Open Media Network
We need a non-political use case, then we have #indymediaback (to political) and the resistance exabition (to political)
this is about the samaticweb but hard to talk about this as the current projects are way to top down to be relevant so not sure how much to push this side.
Home - 4opens - Gitea: Open Media Network then this is the foundation of them both.
The #openweb and WWW worked because they were #KISS we need to do the same, simple/human understandable at each step.
Then we would need a good coder for each project. So we are talking about employing 5 people for a year. Then get fallow on funding in 6-9 months to expand the crew and extend the funding.
We apply for funding for both projects so we need 3xhalf wages and a full wage for each project.
We use Open-Media-Network - Open Collective for the democratic holding of the funding.
Have code hosting at OMN - Gitea: Open Media Network
This might be interesting and relevant: