Observing fedi-folk from various marginalized communities snipe at each other over the past week has been devastating and tragic. No conspiracy theory here, but if there were some nefarious plot to weaken the fediverse, provoking a conflict like this one would be an effective way to go about it.
The purpose of this post isn't to further stir the shit. But it's worth taking a look at origins, alternatives and possible consequences in light of the ongoing threat of authoritarian and capitalist recuperation looming over the fedi.
Fact: There's nothing magic about a blocklist. There are numerous of them in use on the fedi. You yourself can make one by popping open LibreOffice and typing a few server names into csv cells. If someone wants to make a blocklist which is transphobic - or for that matter racist - they're free to do it. They would be a piece of shit for doing it, but nothing's stopping them, and nothing stops anyone else from loading it into their personal account or their server config if they're an admin.
The critical issue with The Bad Space isn't the content of the blocklist, or even the nature of its "trusted sources". It's what those who are funding its compilation intend to do with it.
The Bad Space is at the center of a plan called FSEP - the Fediverse Safety Enhancement Project - sponsored by two organizations, The Nivenly Foundation and IFTAS.
The Nivenly Foundation's mission statement cites goals such as "bring[ing] sustainable governance to open source projects and communities" and "building an equitable future for technology communities", laudable goals. Their team is comprised of several tech industry executives who are or have been employed at corporations such as Google, Microsoft and Twilio.
Their endeavors include sustaining the Mastodon instance hachyderm.io and an AI project.
IFTAS - Independent Federated Trust and Safety - is a fediverse-focused organization which "aims to serve as a valued resource for trust and safety in a complex social media landscape". Moderation is their concern. Their team consists of tech industry executives and movers in the non-profit and open-source communities.
FSEP presents The Bad Space as "providing an easy way to keep blocklists up to date from a central location" in a manner which "can be integrated into existing platforms to enhance the overall experience on ActivityPub-enabled applications by automatically limiting the opportunity for bad-faith actors to interact."
It proposes "integration of The Bad Space’s features into backend functionality" by "allowing blocklists to be automatically imported during onboarding" of users and instances. It affirms "For MVP, only one deny list provider needs to be supported in this list (The Bad Space)."
"MVP" is "minimum viable product" - presumably FSEP version 1 - with a later goal of allowing a choice "from one of several vetted providers".
The issue here should be clear enough. FSEP is a plan to centralize control over blocklists.
From here, we could easily dive in to why it might be a Bad Idea to make The Bad Space, specifically, the locus of that centralized control - not least because, though some of its "trusted sources" are solid, a few are in fact broadly untrusted by a majority of the fediverse.
But the point here isn't to join in the dunk on The Bad Space. FSEP would be problematic regardless of where it centralizes its control. The problem is that FSEP centralizes control.
The purpose of this thread is not to suggest malign intentions on the part of anyone involved. We ourselves have advocated for better ways to federate deny- or allow-lists. The concern is the method.
There is a type of social engineering at work here which is unmistakably redolent of big-tech solutionism. These approaches may seem fit for the top-down hierarchies of large corporations, but they won't work on a network whose essential character is defined by its decentralization. The week's debacle is a result of this mismatch.
Notice the comment of @mekkaokereke , "We will make the Fediverse safe for people of all backgrounds." Everyone with a soul will agree with the goal. But it's a different statement than to affirm, for example, "We will all make the fediverse safe together." Who does and does not comprise "we" in a project of this nature?
Compare the FSEP initiative with the independent project Fediseer developed by @db0 , intended initially to federate "endorse" and "censure" functionality on Lemmy. Both are crowdsourced, there is no "blocklist author". In fact, there is no real blocklist in a static sense, but rather a dynamic pool of classifications.
Approvals and disapprovals are affirmed through a human-based "chain of trust". Sets of classifications can be subscribed to in a granular manner. The system also allows for another principle sorely lacking in the blocklist-based approach - grace.
In the event that a temporary issue is resolved, censures can be withdrawn, and those changes can automatically federate. It's a far cry from the quasi-moralistic condemned-for-all-time scarlet letters regularly dispensed by some of The Bad Space's "trusted sources".
Fediseer represents the kind of horizontalist, agency-distributing creative thinking the fediverse needs to solve its blocklist problems. (It should also be noted that @db0 is collaborating with The Nivenly Foundation on another project, and the reference here to their Fediseer work shouldn't be construed as forming any connection between @db0 and the critique of Nivenly in this thread).
And those problems do need to be solved. As we have noted before, Facebook and its associated operatives are coming for this network. They intend to target weaknesses in the blocklist system to claim that the only possibility for sustainable moderation on the fediverse is corporate-centralized algorithmic surveillance, complete with functionality to auto-report subversive elements to the "appropriate authorities".
In a meeting held in early August, the imposition of algorithmic surveillance onto the fediverse was discussed by several participants, including a representative of IFTAS. Their organization was presented as a possible vehicle for the standardization of third-party surveillance-capitalist moderation schemes. It should be noted that there was a mention of possible disagreement within their organization on the propriety of this idea.
The fediverse has real moderation problems, and one way or another, changes are coming to the ways in which they are addressed. Two possibilities present themselves.
We can continue to watch as solidarity and any sense of mutual purpose drain away, while self-appointed central-authorities assume control of the best practices on how to do a fediverse. Or we can proactively affirm and act upon the recognition that democratization, decentralization and the horizontal distribution of agency remain the only hope for preserving the fedi as a safe site of resistance and prefigurative possibility.
I appreciate the feedback, most of which is good, and most of which echoes feedback and concerns from Nivenly members.
FSEP is not an attempt to "centralize control over blocklists." I'm not sure where you got that idea from, but that is very much not the case.
And I don't think the whole "This person used to work at, or currently works at, big tech company" is as useful feedback as "I don't agree with this part of their plan or objectives."
Worth carefully considering what participation means in practice. Including someone in the “we” you state because they’ve provided feedback or made public comment is not the same as involving them in decision-making.
Is there a “we” making decisions here? How is that “we” accountable to the broader “we” of either people explicitly and intentionally “contributing” or the broadest “we” of people using fediverse platforms?
Here's one concrete example:
Members reviewed the FSEP proposal, and asked what the constraint was that's driving TBS being the only list available at MVP. Their feedback is that at a minimum, there should be multiple options.
We'll see if when this launches, if TBS is the only option, or if member feedback was listened to.
TBS put in the work to build The Bad Space, and significantly improve safety on the Fediverse, specifically for Black trans folk, but for everyone. That's not to say that it's perfect.
TBS also lead with the first draft of the FSEP proposal, and with pulling together the coalition of folks that are directly contributing to this work, and providing feedback.
I don't find the conversation of pedantically deconstructing what the word "we" means, as interesting as the conversation around "How would someone provide input into this project?"
Nivenly has provided a number of ways for folks to contribute at this design phase, through GitHub, Docs, Email, or commenting on Fedi.
When we do get to implementation, there will be opportunities to contribute there as well. There will also be opportunities to publish your own lists.
“Pedantically deconstructing” is a curious way to describe reflecting on the implications of different governance and decision-making systems, but a great way to dismiss questions about power and authority.
And “how would someone provide input” is a fine question, if you’ve already made decisions / assumptions about how decisions are made and who gets to make them.
@mekkaokereke
Appreciate you engaging on this thread.
To address one of your points, the reference to centralizing control over blocklists is based on text from the FSEP paper quoted word-for-word in toot 5 of this thread. But it is understood that the project is evolving, and it will be encouraging to see concerns about the concentration or distribution of agency addressed as it moves forward.
@mekkaokereke@ophiocephalic "And I don't think the whole "This person used to work at, or currently works at, big tech company" is as useful feedback as "I don't agree with this part of their plan or objectives.""
Then you don't understand the specter that hangs over us in these endeavors. Unless someone is a confirmed apostate, working at those companies is suspect.
Add comment