Discussion: Support the Anti-Meta Fedi Pact

@Robert Guthrie thanks!

Social.Coop is already a multi-platform organization: Mastodon, Loomio, Open Collective, Matrix, Git.Coop, Meet.Coop, Nextcloud & Jitsi, via Mayfirst.coop, others? So there is precedent for this. The admin demand involved is what motivated a recent discussion and vote on a proposal to form an organizing circle. https://www.loomio.com/p/OI8kEjVz/social-coop-organizing-circle-proposal So, we need to get the organizing circle up and running to be able to organize the labor involved, but no reason not to discuss it!

As we take these discussions forward, it may be useful, for those not deeply familiar with the consequences of de-federation, to unpack a bit more the implications of the decision in plain language.

E.g.

• 'immediate defederation, no conditions' = no messages will be exchanged with threads.net / users of threads unless a future social.coop decision reverses this policy; whereas

• 'federate but be prepared to defederate on certain conditions' = social.coop users will be able to follow and see content from threads.net users, and have their content seen and interacted with on threads.net. If threads.net actions on data use or moderation violate the principles developed by social.coop, then social.coop admins would be empowered to cut off threads.net, and social.coop users would lose their threads.net followers / following, and ability to interact across the two.

@Tim Davies It sounds like the first is SUSPEND, and the second is do nothing? What about LIMIT?

I regret that I have not had time to participate in this vigorous discussion. From what I have gathered, I have an additional proposal: whatever decision is taken on this issue, that it be time-bound, and revisited in 12 months. Let's take a wait-and-see approach to the Meta situation (either way), and revise strategy appropriately.

@Thomas Beckett

This approach means allowing FB to track everyone for a year before defederating.

FB will NOT delete that data when requested.

It's why they didn't launch in the EU, as FB's normal M.O. will be in breach of the GDPR.

You do know that FB tries to track everyone, not just the people who use FB, but also all of the people that are connected to them via their social interactions even if the people being tracked are not FB users, yes?

This pattern was one of the reasons that the GDPR was created.

My current guess is that FB will keep all of the data, and continue to track everyone, but they won't make the data available to anyone who hasn't signed their NDA.

This is the pattern of behaviour that FB has been following for years.

The tracking data is where FB's primary profit-margin comes from, and they won't let go of that without a fight.

Can there be more clarity around whether this means a LIMIT or a SUSPEND?

A LIMIT on Threads means that nobody on Social dot coop will see content from any Thread users unless they explicitly follow them. This means that if somebody from Threads hurls hate speech at you, you will not see it.

A SUSPEND on Threads means that users of Social dot coop can’t follow people on Threads even if they want to, and users on Threads can’t follow users on Social dot coop even if they want to be followed.

With a LIMIT, any individual user can still personally block the entire domain if they want, and will never see any content from Threads, and will not have their content visible to anybody on Threads, and Mark Zuckerberg cannot access their personal data or their precious bodily fluids.

I’m fine with a global LIMIT. But personally I know there will be people I want to follow on Threads. Preventing social dot coop users from following people they want to follow feels like an overreaction. “EEE” is not a good enough reason — this is political posturing and won’t affect Meta in the least. Even though I personally strongly dislike Meta and wish everybody used Mastodon instead, this is not realistic.

@jsit I assumed implementation details should be worked out after deciding our intentions.

@Brian Vaughan But these are very different moderation actions. I’m not sure how you can decide whether to do “one or the other” of them without knowing which one you’re talking about.

@jsit I had first run a sense-check, asking if we should collectively sign on to the Anti-Meta Fedi Pact. A slim majority favored that, but the discussion led me to think that there might be more support for defederating from threads.net under certain conditions. So I tried to articulate what I thought was a reasonable range of conditions for defederating, from pre-emptively, to not necessarily doing so at all. That was already a lot to jam into a single vote; I thought that if we defederated conditionally, we'd have to make do with a very general sense of what sort of conditions, coming from the discussion attached to this vote, and set up some sort of committee to determine those conditions in detail. If we did, after all, vote to pre-emptively defederate, then I would have proposed a formal vote on signing the Anti-Meta Fedi Pact. I wasn't aware of the difference between a LIMIT and a SUSPEND, but important as that distinction is, that isn't the main question to be decided here. The question here is whether we, collectively, believe that Meta's threads.net is, definitely or possibly, an extraordinary threat.

@Brian Vaughan The leading poll option reads: “Social coop will defederate threads dot net.” It’s not at all clear what this means if limit vs suspend isn’t clarified there. How can one vote on “defederation” if one doesn’t know what that entails?

@Jay Defederation means suspend.

@Django How are people voting in this poll supposed to know that?

@jsit Defederation means severing federation. As far as I can see from reading the documentation, LIMIT doesn't sever federation. So it's not defederation.

As this thread points out, limiting won't help at all.
https://octodon.social/@siege/110666824291864135

@Brian Vaughan I think you are being disingenuous. People use “defederate” ambiguously, and it’s not made clear here which meaning is implied.

What your linked thread points out is that Threads users will still be able to see your posts on social coop even with a Limit. This is true, but they can also see your posts if they are logged out and looking at your profile.

Again, anybody who is individually worried can Suspend Threads themselves, and their content will not be visible.

@Jay I think @Brian Vaughan has actually done his best to bring this topic up in a constructive way, and has also synthesized a very complex discussion. We may not all agree with each others positions, but I think things will work best if we assume positive intent.

@jsit It's not disingenuous. Defederate means to de-federate. It means to sever the connections that constitute federation. Retaining federation, as LIMIT does, is not defederating. I was just looking through Social.Coop's announcements through Mastodon, and on one occasion, we had limited mastodon.world and techhub.social, and on another, mastodon.social. These were described as temporary measures to deal with transient problems; I don't recall anyone describing them as defederating.

And these were fundamentally different situations than we're discussing with threads.net.

@Jay sounds like LIMIT is also the way to go for the goal of communicating to people on Threads and recruiting them to leave it and join better servers… anyone who wants to do that could actively seek out places to post anti-Threads messages to Threads users. Right?

@Aaron Wolf It’s also the way to go if users will just plain want to follow Threads users. I know I will.

@jsit You could just make a Threads account.

@Brian Vaughan But I do not want to make a Threads account.

@Jay that's compatible, as long as this happens and you do follow Threads users, please consider regularly making side-note comments to them inviting them to leave Threads or otherwise expressing concerns about Threads.

Consciousness-raising is more important than purism IMO

I had assumed that threads.net would "play nice" at first, and try to build up trust. However, it has become obvious that it has no intention of playing nice at all. On its first day, users are reporting that threads.net is promoting hate groups, including Libs of TikTok, Moms 4 Liberty, and Gays Against Groomers, and the fascist politician Marjorie Taylor Greene. This is already significantly worse behavior than from most Fediverse instances Social.Coop has already defederated. The only qualification here is that threads.net has not yet enabled its ActivityPub bridge.

@Brian Vaughan and also seems to have accessibility issues: https://mashable.com/article/threads-alt-text

@Brian Vaughan

FB are not playing nice with even their existing users:

https://mas.to/@BlackAzizAnansi/110664803472255089

More on Threads and unmoderated bigotry.

https://www.mediamatters.org/facebook/far-right-figures-including-nazi-supporters-anti-gay-extremists-and-white-supremacists-are

In fairness, apparently Threads moderators removed a post by Libs of TikTok for offensive content.

If defederation is implemented, are we talking about limiting or suspending per https://docs.joinmastodon.org/admin/moderation/#server-wide-moderation? I'm much more comfortable with limiting than a full suspension, because it would still allow me to follow reasonable accounts on threads.net.

Email is the mostly important and widely used federated network service on the Internet.

I have hosted my own email server for nearly 25 years because I do not want to trust my personal communication infrastructure entirely to large companies I don't trust. I have helped host email for many other people through organizations like social.coop for many others who feel similarly but don't have the same technical skills.

That does not mean I don't want to ever send or receive emails from people who use Google (GMail) or Microsoft (Outlook, Hotmail, etc). Doing so would make it quite impossible for to communicate with many people I care about and work with, including—I'll bet—most of the people in this thread.

Centralization of email within companies Google and Microsoft has had many negative impacts on email and I think we should learn from that experience. But it seems clear that we stand more to lose more than we gain by simply blocking Threads before anything happens.

@Benjamin Mako Hill This points to the heart of the matter for me, and the analogy with email is accurate. It's not so much that we don't want to communicate with loved ones and creators and organizations on Threads as we want to avoid the threat they represent on two fronts:

1. Embrace and Extend - as we've discussed, the anticompetitive practice of pretending to join an open system in order to dominate or eliminate it.

2. Data gathering - Lest we forget Cambridge Analytica and how they used Meta/Facebook's data-driven microtargeted ads to get Trump elected and Brexit to "yes." What are they really after? Our real names and locations, our likes, social connections, and enough metrics to build a psychological profile on us. The question is, can we allow communication with Threads without letting them profile us?

@Michael Potter Reason #1 is important but I don't see how preemptively defederating is helpful. If they join and extend the protocol in ways we don't like, we can choose to defederate then. Preemptive defederation feels like destroying property to prevent someone from stealing or vandalizing it. Or not making money because you don't to have someone take it from you later.

Reason #2 is interesting but also doesn't feel like a reason to defederate. I'm completely confident that Meta already has public Fediverse data and will continue to get it. Many people do! They will only get private data like DMs to the extent that individuals chose to communicate with people on Threads. I have written about exactly this situation in relation to email in a pretty wide read essay: https://mako.cc/copyrighteous/google-has-most-of-my-email-because-it-has-all-of-yours

It's a real concern but it's a choice I'd rather make an individual when communicating with my GMail using friends. If something were sensitive, I might not send it to a GMail users. I am absolutely confident that I benefit more from being able to send email to GMail users than Google gains from having copies of those emails. Otherwise I wouldn't do it. Do you disagree?

I too am interested in adding Lemmy/Kbin to social.coop. Now that Reddit is pushing so many out it would be great if we could support an alternative.

Point of process: the ranked choice vote is not something that can constitute a formal decision according to the bylaws. It does not, for instance, clearly state a proposal or allow for blocks. There is a guide to making proposals here: https://wiki.social.coop/wiki/Make_a_proposal

@Nathan Schneider thanks for pointing to the process. As you probably saw it did come up in the discussion leading up to the creation of the ranked choice vote. I guess we need to close it early and restart?

I personally think that the proposal was quite clear, can you say more about how it wasn't? Instructions for blocking were also included, even if there isn't an explicit choice for that.

@ntnsndr I will again emphasize that we had several weeks of discussion, in which I repeatedly asked for advice on process. No one offered advice, and no one objected to my suggestions for how to handle process. I delayed submitting a formal proposal specifically to solicit advice on process, but the fact that threads.net had officially announced its launch meant that we were running out of time. I specifically asked about how to include a "block" option, then suggested it be spelled out, and that we would tally votes manually if necessary to account for blocks, because I couldn't find a way to do it through Loomio.

It doesn't seem appropriate to me to object to the vote halfway through the voting process, when a quarter of the membership has already voted.

@Brian Vaughan sorry that I missed part of the discussion. I did actually offer advice on how to make the proposal bylaws compliant when I shared the wiki page on how to do so, before this proposal was created. I don't want to be a huge stickler for the rules. The structure of this vote makes it a bit hard to evaluate according to the bylaws. But perhaps it will be unambiguous enough.

I don't mean to malign Brian's work here. I know from experience that running a contentious vote in this co-op is tricky!

@Brian Vaughan I think for the sake of clarity and process members should decide on whether or not to make an afirmative direction to defederate (suspend). My preference would include the post-action committee work detailed in #2. The current split of the defederate vote between two options muddies the call for action even though combined they represent almost 60% of the sentiments expressed.

@richjensen3000 Two of the four options entail creation of a committee, so we probably should be thinking about a process for forming that committee, in a manner as compatible with established processes as possible. I'm not sure how we'll assess the results of the other vote, but of course it bears on the interpretation of the results of this vote.

Since we have had at least one vote to block, there's also the question of how to assess that. I'd thought it should be treated as a vote for #4, taking no action, but weighted as nine votes. That's based on my understanding of the way a block vote would normally be assessed.

Loomio allows downloading the CSV of the votes, and it's calculating ranked choice votes by a simple point system. In this case, someone's first choice gets four points, their second gets three, and so on. So I think a block vote should count as 36 points for the option of no action. I won't be able to post my recalculation of the vote until well after the vote concludes, tomorrow evening my time. It would be a good idea for other people to try the math; I'm trying my best to be fair and honest, but I am obviously not disinterested.

The character limit for vote comments is small, so I'm putting my full response here instead:

I don't think any poll which ties a judgment to a narrow course of action can represent an actionable proposal - it is closer to a political push poll. If I had to express my view in the structure of the choices here it would be:

We assume that Threads intends to be a threat to the Fediverse, but federation will actually benefit us more than it benefits them.

Having the walled gardens open up the gates through federation is, for me, the win condition for the Fediverse. It is the goal that drives my involvement here and the creation the Immers Space worker-owned cooperative that creates Fediverse software. For us to collectively build up another wall around that garden would be snatching defeat from the jaws of victory.

Make no mistake - I believe that Facebook's intentions are bad and they will do everything in their power to federate in a way that draws and traps people into their network. However, I also believe this is an historic misstep by a leader prone to recent historic missteps and our best opportunity to bring vast numbers of people out from under Facebook's thumb.

What this really comes down to is a simple question: do you believe that the cooperative model is able to provide a social media experience that better meets the needs of people than the extractive model? If so, then it must follow that federation with threads will bring more people from them to us than vice versa. When they open up federation, they will greatly reduce the social cost of leaving their network as people will still be able to stay in touch with their existing friends, and I believe that cost - Facebook's current practice of holding your online connections hostage - is the primary barrier keeping people from moving en masse to the Fediverse.

Therefore, in order to pursue our vision of collective liberation, we must find a more nuanced approach than defederation that maximizes our ability to connect with the people that use Threads and show them what's possible outside the corporate social world while minimizing the harm to our membership (and there are real, material harms to consider here as Facebook enables fascists that incite vigilante violence against vulnerable groups). Part of the problem here is that the options Mastodon gives us for regulating contact with specific instances are woefully inadequate for this situation, but we must demand and/or build better ones in order to pursue our cooperative's vision.

@Will Murphy Yes indeed, the best possibility is one in which we make the most accessible open door for people to leave Meta, we need to keep that goal in mind in all of this!

@Will Murphy "Facebook enables fascists..." - Meta enables fascists because Meta's business-model depends on the 'controversy' fascists create. Fascism/intolerance/white supremacy/State and Privatized Violence are inherent to their business model. I believe social.coop's value is in manifesting consensual space grounded in an alternative social logic to theirs. While one might make arguments about tone and style, I question whether there is any practical ideological difference between Meta and Gab. I think Social.coop should take a similar stance toward these two entities. Some people seem swayed by 'scale' and popularity and the unfortunate conditions that their friends and family members tolerate. Others are not swayed by these. I appreciate this discussion.

@Rich Jensen

That's a great point. I don't see any fundamental difference between Fb/Meta and Gab in terms of pragmatic expressed value, that is, what they really do (as opposed to what the marketing people say).

Thanks for pointing that out. :)

I appreciate the vibrant discussion that continues on this thread, but I think it is important to note (preferably at the top of this discussion) that FB/Meta/Threads has not implemented the ActivityPub standard to date, and they are incapable of joining the fediverse until they do. Tumblr said the same thing last November, apparently, but I don't see where they've actually followed through on anything. And there's an awful lot of hedging in Meta's announcement, imo, where they address the fediverse:

Soon, we are planning to make Threads compatible with ActivityPub

Source: https://about.fb.com/news/2023/07/introducing-threads-new-app-text-sharing/

I also appreciate the need for concern, but I don't see how this pact does anything proactive to protect social.coop usership, given what we currently know. Sure, people who join Threads could harrass people elsewhere in the fediverse, if and when Meta ever implements the protocol, but I don't see how that differs from any other server out there currently.

I can't even find the comment now, with how much this thread has blown up, but someone suggested early on drafting a list of terms or principles, sort of like a shared CoC I guess, that all servers would have to agree to before we federated with them. That's certainly a lot harder, but it's a positive assertion of values and would go a lot further toward protecting this space from all future threats, not just cause du jour of fear, uncertainty and doubt.

@Jamie Gaehring

It is in https://cryptpad.fr/pad/#/2/pad/edit/5xxQZvxvjfAP2eQTYenLuWOZ/

@Eduardo Mercovich oh nice! I really like the way that is shaping up, kudos to all who've been working on this!! 👏

@Eduardo Mercovich I suggested the name "Friendliverse Act" in a comment to the pad :)

@Arnold Schrijver

I love it! Also, that made me think on a game about diverse and fedi... But this issue of a common (p)act is different from this present thread, so let's open a new one.

Thanks a lot for giving it some time and love. :)