Not sure the $4,000 figure is an accurate reflection of what the reserves actually are as a budget that captures all the expenses for sustainably running the platform as it scales has yet to be created. Living wage might not be attainable at the moment but a system for tracking critical admin hours and providing partial compensation should be.

Wow. This got traction fast.

Seems like potentially good idea and the direction social.coop was heading in.

I recommend we prioritize getting a few outstanding social.coop organizational changes in order ASAP while this conversation unfolds.

#1 - The Admin Ops discussion needs feedback so can be up for proposal vote. https://www.loomio.org/d/URY9AKci/creation-of-admin-ops-teams

#2 - A fully formed allocations based budget should be put in place to make sure that the base financial needs of the social.coop (including a funding for remuneration of operation critical admin duties) is provided for. Discussion in progress but needs feedback in order to create a proposal. https://www.loomio.org/d/cIA3pK4a/looking-at-the-long-term-financial-vitality-of-social-coop-

This looks great! Helping move Mastodon in the direction of platform cooperativism would be a big step forward. Like I said elsewhere -- I feel like we need something like a retreat to discuss strategy for social.coop. Because it seems so great, and in the spirit of scientific inquiry, I would also like to hear from @mayel all the best arguments AGAINST this idea.

So I had actually been weighing the idea of going to a third-party for some of the hosting work as part of the Tech WG strategy, so this is a little different than what I'd imagined, but a very interesting opportunity.

My initial sense would not be for social.coop to take it over directly, but form it as a "shared services" co-op owned by the instances that use its services, with social.coop as a founding member. If we wanted to get more complex, it could make for a very interesting multi-stakeholder co-op model, with the staff owning one half of the hosting company, and the instances owning the other half (shared services co-op), but the "second-tier"/shared services model would probably be the simplest.

I'd be happy to get on a call to discuss options and possibilities, if @mayel can find a few dates/times that work for him and Alice and builds a schedule poll around them.

Does social.coop have a legal entity, or is it an unincorporated cooperative? (excuse my ignorance on this, I'm new here).

Unincorporated, but we have the ability to issue and pay invoices through our OpenCollective host acting as fiscal sponsor.

That is in the process of changing now that we are in IoO and working with Sarapis.org @ntnsndr @thomasbeckett

Right, the fee will be 5% rather than 10%, but we'll retain the same abilities, correct?

My understanding is free, but @ntnsndr has the latest info on the Sarapis front

IIRC currently OpenCollective takes 5% as a platform fee, plus 5% as "host fee" for the fiscal sponsorship.

You took the words out of my mouth. +1 to this.

I like the idea of social.coop "seeding" a new shared services cooperative to take on hosting. Starting a coop sounds like it could provide the right environment to solve the questions of how to pay people, make the decisions we have to make, and things like that. It would also mean that social.coop could spread the cooperative model, and participate democratically in decisions about the server, without centralizing power too much by taking over the servers of a bunch of other instances for itself.

I'm not a huge fan of the idea that a service can be owned by other services—I think the owners of any collective formed should be the humans running and/or using it.

Federations have one set of problems (alienation inherent to layered representative democracy), co-ops with beyond-human-scale memberships participating in mass democracy have a different set of governance challenges.

One approach might be to require instances hosted instances above a certain size (say, 1k members) to be run in alignment with the Cooperative Principles? So ownership and control of the network ultimately is structurally rooted in a broad base of people?

Welcome to @alice133 who is now on here! Would love to here your thoughts on the conversation, and please let us know if/when we can have a call or some other form of chat soon.

I like what I've seen so far, it finally looks like a good future for MaaStodon.

Timezone?

Loomio should be showing you the time in your timezone.

Neat

Thought I should throw in my two cents since I rather accidentally backed into this :)

My thoughts:

• I like the idea of having hosting run as a separate co-op or sub-co-op (if that is a thing)
• It could definitely be a step toward #ReplaceTwitter (vs. Buy)
• Keeping co-op members separate from regular clients is probably best as well
• Definitely would want to put social.coop instance here (dog-fooding)
• While this URL is set up for Mastodon, no reason the hosting coop can't expand to GNUsocial, Pleroma, etc. hosting - or any other hosting that makes sense for coop structures
• This would help insulate this entity from issues if Eugen does something drastic or unexpected w/ Mastodon
• Would be a natural for other *.social.coop instances, or even just *.coop
• Definitely need to pay admins/devs - probably on contract basis for now, then utilize consultation or on-call structure after that

How can this be operated to pay people fairly? I could help develop financial projections. This might also be a good multistakeholder co-op, between worker/hosts and instance/hosted.

Is general server hosting a potential service offering, too?

A few suggested operating principles for how such a project could work:

• Critical system expenses (hosting, backups) are the top priority.
• Fair remuneration of core sys admins for key administrative tasks (e.g., upgrades, installs) is the second priority.
• As a whole, the platform should always invite volunteers, as per the existing initiative to create a volunteer ops team.
• The co-operative should actively work to mitigate against technical over-centralization (e.g., all instances with the same hosting provider, same location). Replicating the setup with multiple hosting providers should be a medium term goal, to enable spreading out instances over multiple hosting providers, if this is not already the case.
• The co-operative should give all instances the option to switch to a co-ownership model like social.coop, but this should not be required.
• We are collaborators and should strive to help make meaningful connections between instances, to provide useful metadata to potential users, and so on. Shared resources like the social.coop wiki can be leveraged across participating instances.
• Maastodon.net appears to have a solid policy against hate/harassment. It is imperative that this policy is maintained and enforced, including in public messaging. Indeed, ideally we would make it easy for instances to simply inherit a reasonable code of conduct as a starting point.

These are just some initial thoughts. I love the idea of a large synchronous conversation, though timing will be tricky for me if it's during my work week. :| In any event, this is the most exciting proposal I've read since joining social.coop. There are lots of things to figure out and absolutely risks to think through, but please please let's keep the positive momentum going. :)

This is very exciting. Thank you, @mayel! This is an opportunity to move the social.coop vision forward in showing how the co-op model can really super-power federated networks. However, the logistical issues at play here are quite considerable, and I worry that we're still struggling a bit to make our own small instance functional. This will require a rapid curve of professionalization and accountability—especially as it means switching from an affinity-based model to one that includes people who are not necessarily avid cooperators.

Nevertheless, I am extremely hopeful about this and believe it is an occasion we should make the effort to rise to!

I'm sorry I can't make any of those times for the call. But I'll be watching this thread enthusiastically. Please let me know how I can be of help.

@davevnd3jr Loomio should be showing the times in your local timezone

An option that should be on the table here is to partner with an existing tech co-op to help manage the infrastructure.

If anyone knows of such a tech co-op that may be interested, I'd be very happy to speak with them (about the SaaS, about the possibility of them doing the maintenance of social.coop's tech infrastructure, and about me and possibly Alice and others joining them ;)

Most of the tech worker coops I've seen seem to be groups of friends in one area rather than internet strangers...

In the US: https://techworker.coop might be of help.
In the UK: https://www.coops.tech
Elsewhere?
I'm thinking about this option primarily as a means of quickly bringing to bear some solid backing to support the work that yourself, Alice and others are doing, often on a volunteer basis.

what if we hosted it ourselves? for example, I have some servers sitting around in my room wasting 98% of their cycles, and I would love to put some of those to work. I'm sure other people have unused hardware sitting around too that they could either donate/sell to co-op and/or bring it with them when they join the smaller sysadmin workers coop and then manage it with everything else. we (should) own it!

@gregcerna what's your connectivity / bandwidth like?

I am now chatting with Chris from https://www.webarchitects.coop/ (I was put in touch by a mutual friend) about this option (of working with an existing tech workers coop for this)

300 mbps up and down, but I live in a housing cooperative, so I share that bandwidth with the rest of my housemates lol. it's just a residential connection but it's a solid one, and I've been running a toooon of services behind a reverse proxy for almost a year now, and my isp has yet to notice and doubt they would anytime soon. if that's not enough bandwidth, one of our other houses like two blocks away has a gigabit fiber connection (the lucky bastards) and it would be straightforward to move my machines there.

Say hi from me.

Web Architects seems like a great candidate for this, since they're already building out co-op cloud services like git.coop

got an email that said the call is on monday the 7th. but then saw a comment that its happening on a tuesday, is this because of conflicting timezones? I am in berlin.

Not sure what happened but it should be 05/08 at 7:00pm, in Berlin, Germany.

Love it love it love it! Way to GO!

I've been offline for a few days so I missed this, but this is a really exciting proposal. I agree with @matthewcropp's suggestion of seeding a "shared services" co-op—I think this would allow social.coop to focus on spreading the co-operative model, while allowing MaaStodon to focus on hosting.

So the chat about this proposal is happening on Matrix in about an hour. Alice from MaaStodon will be joining us, and maybe also Chris from https://www.webarchitects.coop/

There is a collaborative agenda that needs populating here: https://oasis.sandstorm.io/shared/l6owJmSrlcNlOGi1lr2npBMLxsLBLQQUhhLqL8RlKdo
18:48

It would be great if people (whether you can join the meeting or not) could add important points / questions / suggestions there so we can reduce the noise during the scheduled chat by following the agenda.

Here is the agenda for the call, in an ODT doc.

Couldn't figure out how to export the chat, so here's are the notes from our 5/8/2018 meeting.

Gender check self-criticism -- the discussion participants were nearly all male, as far as I can tell. We need to be more intentional in our organizing of discussions.

Thank you for raising this.

I copy/pasted the chat into the attached .odt file also into the pad that was used for the agenda. I'm spreading this thread among co-op minded people who may have necessary tech skills

should note that its not at all clear when people were quoting others - so bear this in mind if reading

is there any way you could post the pdf version as well? my phone can't open odts. also, are we putting these in the wiki?

@gregcerna heres a pdf of the copy/pasted chat text from #socialcoop.matrix.org

Quick update - I'm gonna be posting an initial poll here based on the chat we had, but I first want to catch up on all the related discussions and proposals in Loomio (such as about Tech Ops team) in order to take those into acount in framing the proposal, so please bear with me!

By the way, some of these other proposal are happening in the various working groups, so for any members who are interested, you might want to join them to participate.

which working groups are these proposals in? I can't find them

There are a few, I see one's in Governance, the other Finance, and maybe more.

so quick update: Eugene is moving mastodon.social off of being hosted by maastodon. So the main instance that we would be taking over control of is no longer present. Has this conversation gone elsewhere? seemed to die after the group chat....

I can confirm Mastodon.social has now moved to their own servers.
It did get quiet since the group chat; I'd love to see some development before the end of the month.

I missed all of this while away of holiday, it's a really exciting prospect! Just now trying to catch up on the backlog.

MaaStodon is effectively closed, it shouldn't change much related to this discussion, but for the time i will be responsible of it it will stay in that state.

I wasn't part of the call, and so perhaps I don't fully understand the proposal, but I personally don't want to support a large instance of users (e.g. Octodon) who are not themselves contributing to the coop. I would prefer to see us grow social.coop organically, and go through the technical and administrative growing pains, without doing a big infusion of users which could amplify some of the problems we will encounter. Bigger is not better.

I feel the same way. It could be nice to have multiple cooperative instances either under the social.coop group or (organizationally) federated and invite other instances to join if they're using a cooperative model. If octodon.social was interested in moving to a coop model, I'd be happy to help them with the transition.

As I understood it, hosting other instances as a commercial service would benefit the coop by funding the infrastructure (and potentially more), while being able to host other coop instances.

I think I like the proposal even less now. I didn't get into social.coop to help start a business.

I like Alice's understanding; it is a bit more like a workers coop or producers coop model than a consumer coop model. The big cost, in time, skill, and money, is the hosting and admin. It makes sense to pool instances to economize, rather than each instance try to solve that problem. It also makes sense to make that hosting and admin cooperative -- something we can do, and from which we benefit.

It seems to me that the big assumption being made here is that pooling resources is going to make things easier--that economies of scale are going to kick in when we are managing 100 instances with 100,000 users instead of 1 instance with 1000 users. That seems pretty debatable if you ask me.

I would prefer to see social.coop focus on getting one instance working with a team of co-op member admins with a code of conduct folks can live with, and work from there instead of trying to manage the complexity of multiple communities, with different values, all using the same infrastructure.

Though I do believe a degree of economies of scale is beneficial and feel Social.coop can and should facilitate many different actives where ever there is enough member support and will to voluntarily manage, I'm completely on board with your prioritization of activities (especially involving funds) around making this instance operationally sustainable. There is a in-depth discussion inside of the Finance working around with some ideas that I believe could accomplish both. Would love to have others weigh in.

As far growing social.coop to a stage where it had 1000 paying members I think that would be an amazing goal and achievement. Currently we have around 150 paying members, 1000 registered users, an unknown amount of them active.

I think we should give @mayel, @alice133 and Co. the space to develop their proposal - and the pros and cons for social.coop, the scalability of social.coop and Mastodon and the spreading of co-operative governance to the Mastodon eco-system more generally.

I think at that point we can have a more grounded conversation about what role social.coop could/should have in it. Until then, we get into the dangerous territory of discussing a proposal without any concrete facts.

@mayel are you guys coming towards some sort of working proposal for how things would operate etc.?

If I were you guys - I would also craft your proposal so that it wasn't just dependent on social.coop (the hosting co-op could be a viable option with our without social.coop's immediate participation). :)

And props again for even considering this @mayel and @alice133 - and for doing the thinking work on it. I'm very inspired.

Thanks @samtoland :)

I personally haven't had much time to think about this or discuss it further since the big chat-up the other week (things have been crazier than usual, as I'm in the middle of taking on a new gig). I also haven't had time to catch up on other possibly related discussions on Loomio.

I still think there's something to this (maybe as a new workers co-op or multi-stakeholder co-op separate to the main social.coop - or by combining our forces with an existing co-op like Webarchitects if they're willing), and that it could help solve many pain points for social.coop, for @alice133, and for @victormatekole and I. For that to be the case though, it seems we'd need some new people who have the time/energy to step up and help make it happen, and also help with operations/sysadmin of the new organisation in an ongoing way, as it seems most of the existing people are a combination of busy and burned out.

Hi, regarding Mayel's post:

I still think there's something to this (maybe as a new workers co-op or multi-stakeholder co-op separate to the main social.coop - or by combining our forces with an existing co-op like Webarchitects if they're willing), and that it could help solve many pain points for social.coop, for @alice133, and for @victormatekole and I.

We are definitely interested in exploring this further as and when it is suitable to do so. It does take quite a lot of time and energy to setup and maintain co-operatives and if we can save some effort and help the project continue and make it sustainable then we would be happy to help.

I think where we stand is that social.coop needs to a) find consensus in our current discussions of organization (operations teams) and financing, b) get the tech ops team/working group organized, so that the work can be spread and coordinated, and then c) come back to this really important discussion. So, I hope Alice and Chris will not give up on us.