#recommendation-algorithms — Public Fediverse posts

Opinion: Google Has Become Too Powerful for Democracy to Ignore

By Cliff Potts | WPS News Opinion

BAYBAY CITY, LEYTE, Philippines, August 3, 2026 — 0005 PhST

The question is no longer whether Google has become one of the most influential corporations in human history. That question has already been answered.

The real question is whether any democratic society should permit a single private corporation to exercise such sweeping control over how information is discovered, distributed, monetized, and preserved.

Google is no longer merely a search engine.

Through its parent company, Alphabet, Google controls the world’s dominant search platform, YouTube, the Android operating system, the Google Play Store, one of the largest digital-advertising operations in existence, the Chrome browser, major artificial-intelligence systems, mapping services, analytics, cloud infrastructure, and an extraordinary volume of information about its users.

Each of those operations is powerful on its own.

Combined under one corporate roof, they create a digital ecosystem capable of influencing who is found, who earns money, which businesses reach customers, which applications succeed, which publishers remain visible, and which voices effectively disappear.

That ownership structure creates the opportunity for abuse.

Google Search can reinforce Google advertising. Android can reinforce Google Search. Chrome can reinforce both. The Play Store can dictate the terms under which software developers reach mobile users. YouTube can determine which videos are recommended, which channels qualify for advertising revenue, and which publishers lose practical access to audiences they spent years building.

None of this requires a secret meeting in which executives decide to destroy one particular publisher.

A recommendation system can quietly stop recommending a channel. A search-ranking adjustment can bury a publication. An automated advertising classification can eliminate revenue. An account restriction can remove years of accumulated work.

The person or organization affected may receive little more than a generic notice—or no useful explanation at all.

That is an unacceptable amount of unaccountable power.

Breaking Up Google Is Not a Fringe Proposal

Calls for structural action against Google are not confined to activists, angry creators, or small publishers.

In 2024, a federal court found that Google had illegally maintained monopolies in general search services and general search-text advertising. During the remedies phase, the United States Department of Justice proposed measures that included forcing Google to divest Chrome and restricting its use of Android and other products to protect its search dominance.

The court ultimately rejected the proposed Chrome divestiture. Its final judgment instead prohibited certain exclusive distribution contracts, required Google to provide qualifying competitors with access to specified search and user-interaction data, ordered syndication opportunities for search results and advertising, and established continuing technical oversight.

Those remedies were significant, but the court’s decision not to order a breakup does not make structural separation an unreasonable idea. It confirms that divestiture was seriously proposed, litigated, and considered within an actual federal antitrust case—not invented in an internet comment section.

The Justice Department has also won a separate case in which a federal court found that Google unlawfully monopolized important parts of the open-web digital-advertising market. According to the Justice Department, Google’s conduct harmed publishers, competition, and ultimately the consumers who depend upon information distributed across the open web.

That finding should matter greatly to every independent news organization.

A company that controls advertising technology while also operating competing services does not occupy the position of a neutral intermediary. It participates in the same market it helps govern.

Google disputes allegations against it and continues exercising its legal rights in court. It is entitled to do so.

The public is equally entitled to question whether the company should continue controlling so many interconnected layers of the digital economy.

Europe Already Calls Alphabet a Gatekeeper

The European Union has formally designated Alphabet a digital “gatekeeper” under the Digital Markets Act.

The designation covers core services including Google Search, Google Play, Google Maps, Google Shopping, YouTube, Android, Chrome, and Alphabet’s online-advertising operations.

The term is appropriate.

These services do not merely compete within digital markets. They frequently determine how other businesses enter those markets, find customers, collect revenue, and survive.

The Digital Markets Act attempts to address that power before another decade-long competition case reaches its conclusion. Instead of relying exclusively upon punishment after damage has occurred, it places advance obligations and prohibitions upon companies possessing entrenched control over essential digital services.

That approach recognizes a basic problem with ordinary antitrust enforcement: by the time a case is investigated, tried, appealed, and remedied, smaller competitors may already be gone.

A fine imposed years later does not restore a publication that closed, a developer who abandoned an application, or a creator whose audience disappeared.

When financial penalties become routine operating expenses, governments must move beyond fines and consider structural remedies.

The Transparency Failure

I cannot honestly claim, without reliable supporting data, that most political creators affected by demonetization or reduced distribution are left-wing, progressive, or pro-democracy publishers.

Progressive creators have accused YouTube of unequal treatment.

Conservative creators have made the opposite accusation.

The available public information does not establish which political group experiences the greatest overall harm.

That absence of evidence does not clear YouTube.

It exposes the transparency problem.

Independent researchers cannot adequately determine whether YouTube’s enforcement is politically neutral because the company does not disclose enough channel-level information about recommendation reductions, monetization classifications, automated decisions, reversals, comparative enforcement, and the practical effects of its algorithms.

YouTube publishes extensive rules governing monetization and advertiser-friendly content. Those policies describe broad categories of material that may receive limited advertising or no advertising.

They do not necessarily tell an individual publisher why views dropped from thousands to hundreds and then to tens.

They do not reveal whether a channel stopped appearing in recommendations.

They do not explain whether an automated system changed its classification of the channel.

They do not demonstrate whether politically comparable channels received comparable treatment.

They do not provide the public with enough information to audit a system that helps determine what billions of people see.

That is the issue.

A system does not need to ban political speech outright to diminish it. It can simply stop showing that speech to people.

I Have Watched the Audience Disappear

This is not an abstract concern for me.

I have watched my own work receive thousands of views, then hundreds, and eventually numbers in the tens. That decline occurred while the company controlling distribution provided no intelligible explanation of what changed.

I cannot prove in a courtroom that Google or YouTube politically targeted Cliff Potts.

I can prove that an opaque, Google-owned system possessed the practical ability to withdraw access to an audience without explaining itself in any meaningful way.

That alone is a democratic problem.

It is also an antitrust problem.

Journalism, political commentary, music, education, and public advocacy increasingly depend upon privately operated systems that can alter distribution without notice. Publishers may technically remain free to speak while being denied any realistic opportunity to be heard.

Freedom of speech is not a legal guarantee of an audience.

But when one corporation has acquired extraordinary control over whether an audience can find that speech, elected governments have every right to investigate how that power is being exercised.

Break Apart the Conflicting Functions

The answer is not another vague request that YouTube “do better.”

The ownership structure itself must be examined.

Search should not reinforce advertising.

Advertising should not reinforce YouTube.

Android should not reinforce Search.

The Play Store should not reinforce every other Google business.

Chrome should not function as another mechanism for protecting Google’s position.

Artificial intelligence should not be allowed to absorb information from independent publishers, answer users directly, and then deprive the original publishers of the traffic required to survive.

Conflicting business functions should be separated.

Self-preferencing should be prohibited.

Recommendation and monetization systems should be independently audited.

Publishers and creators should receive meaningful explanations when distribution or revenue is materially restricted.

Appeals should be decided by accountable human reviewers rather than disappearing into another automated system.

Regulators should have access to the records necessary to determine whether enforcement is consistent, discriminatory, anticompetitive, or politically uneven.

Independent researchers should be permitted to study the effects of recommendation systems without depending entirely upon information selected and released by the company being examined.

When fines fail to change conduct, structural separation should remain available.

These are not attacks upon innovation.

They are protections against concentrated power.

Democratic societies regulate utilities, financial institutions, telecommunications systems, transportation networks, and other industries capable of affecting the public at enormous scale.

They should not exempt the corporations governing digital discovery merely because their control is exercised through algorithms instead of physical gates.

No private company should possess the practical ability to determine who is discovered, who is heard, who earns a living, and who quietly disappears from public view without meaningful explanation or independent oversight.

Google has become too powerful for democracy to ignore.

The only remaining question is whether democratic governments will act before that power becomes permanent.

Sources

U.S. Department of Justice: Department of Justice Prevails in Landmark Antitrust Case Against Google

U.S. Department of Justice: Department of Justice Wins Significant Remedies Against Google

U.S. Department of Justice: United States and Plaintiff States v. Google LLC

European Commission: Digital Markets Act Designated Gatekeepers

European Commission: The Digital Markets Act

YouTube: Channel Monetization Policies

YouTube: What Kind of Content Can I Monetize?

Contact Cliff Potts

Cliff Potts
Founder, Publisher & Editor
WPS News
Baybay City, Leyte, Philippines

WPS News
WPS.News

Bluesky — Preferred Public Contact
WTFM.LOL

Occupy 2.5
Occupy25.com

Cliff Potts
CliffPotts.org

Cliff Potts is also the creator of Cliff Potts & the AI Rebellion, with music available on Spotify, iHeartRadio, Apple Music, Amazon Music, YouTube Music, Deezer, TIDAL, Pandora, and other major streaming platforms worldwide.

For news, commentary, music, and ongoing projects, begin with WPS.News or contact Cliff publicly through Bluesky at WTFM.LOL.

Chicago Reader: Is the Internet panopticon going to ruin live music too? . “Our experience of culture is shaped by social media, streaming services, and other Web-based platforms that mine our data (and may eventually send us $35 checks after settling a class-action lawsuit). But how do those platforms influence how we experience live music? Spotify famously collects vast troves of data, only […]

PsyPost: Recommendation algorithms might be making your entertainment boring, new research suggests. “A recent study published in the Journal of Cultural Economics suggests that highly accurate content recommendation algorithms might accidentally make our entertainment feel boring over time. The theoretical model indicates that injecting a small amount of randomness into these systems tends to […]

Tubefilter: Are male and female social media accounts floating in gendered political bubbles?. “A report published in Cornell University‘s arXiv database shows that recommendation algorithms treat male and female accounts differently, especially in the realm of political content.”

"In the survey, conducted in January, X users were the only group in which a majority, just barely over 50 percent, expressed “strong” or “somewhat” approval of Donald Trump. His approval was significantly lower among consumers of news from “podcasts and YouTube,” local television, and even Facebook. Among people reading “newspapers or news websites,” browsing Reddit, watching broadcast television or scrolling TikTok or Instagram to keep up with current events, the numbers were, as Jain described them, “catastrophic.” He noted, “If you’re largely getting your news from Twitter, you might not even know that Trump is unpopular, because you wouldn’t even see a lot of the backlash.”

Last week, in a study published in Nature, a group of researchers attempted to answer a sensible follow-up question: So what? People organize around news sources that flatter their beliefs, and in a fragmented news environment, you would expect different attitudes to be associated with venues that have developed a clear partisan identity. Well, it turns out that the engine of Musk’s X — its algorithmic “For You” page — is an ideological ratchet..."

https://nymag.com/intelligencer/article/x-really-is-pulling-users-to-the-right.html

#SocialMedia #Twitter #Algorithms #Polarization #RecommendationAlgorithms

"Feed algorithms are widely suspected to influence political attitudes. However, previous evidence from switching off the algorithm on Meta platforms found no political effects1. Here we present results from a 2023 field experiment on Elon Musk’s platform X shedding light on this puzzle. We assigned active US-based users randomly to either an algorithmic or a chronological feed for 7 weeks, measuring political attitudes and online behaviour. Switching from a chronological to an algorithmic feed increased engagement and shifted political opinion towards more conservative positions, particularly regarding policy priorities, perceptions of criminal investigations into Donald Trump and views on the war in Ukraine. In contrast, switching from the algorithmic to the chronological feed had no comparable effects. Neither switching the algorithm on nor switching it off significantly affected affective polarization or self-reported partisanship. To investigate the mechanism, we analysed users’ feed content and behaviour. We found that the algorithm promotes conservative content and demotes posts by traditional media. Exposure to algorithmic content leads users to follow conservative political activist accounts, which they continue to follow even after switching off the algorithm, helping explain the asymmetry in effects. These results suggest that initial exposure to X’s algorithm has persistent effects on users’ current political attitudes and account-following behaviour, even in the absence of a detectable effect on partisanship."

https://www.nature.com/articles/s41586-026-10098-2

#SocialMedia #Algorithms #Twitter #Politics #RecommendationAlgorithms #PublicOpinion #Propaganda

@cloudskater wrote:

Some instances are run by bad people. Hell, a few projects like Lemmy and Matrix are DEVELOPED by assholes, but the FLOSS and federated nature of these platforms allows us to bypass/fork them and create healthy spaces outside their reach.

Nope, that is actually what is killing the fediverse. I just explained here:

The issue is the divergence in semantic interpretation that emerges at the interpretation layer. ActivityPub standardizes message delivery and defines common activity types. However, it leaves extension semantics and application-layer policy decisions to individual implementations. Servers may introduce custom JSON-LD namespaces and enforce local behaviors, such as reply restrictions, while remaining protocol-compliant. But, the noise created by divergences are problematic, because it creates unexpected, unintended, and unpredictable behavior.

Divergence appears when implementations rely on non-normative metadata and assume reciprocal handling to preserve a consistent user experience. Behavioral alignment then varies. Syntactic exchange succeeds, but behavioral consistency is not guaranteed. Though instances continue to federate at the transport level, policy semantics and processing logic differ across deployments. Those differences produce inconsistent experiences and results between implementations.

That leads to fragmentation, specifically semantic or behavioral fragmentation and an inconsistent user experiences. ActivityPub ensures syntactic interoperability, but semantic interoperability (everyone interprets and enforces rules the same way) varies. This creates a system that is federated at the transport level yet fragmented in behavior and expectations across implementations. It is funny how the thing that the fediverse touted has made the entire thing very brittle. ActivityPub technically federates correctly, but semantically falls apart once servers start adding their own behavioral rules.

https://neon-blue-demon-wyrm.x10.network/archives/16932

FYI, I’m not doing culture wars or political debates. I’m just saying this idea of “forking away” from them is literally breaking the fediverse’s distributed network and creating all kinds of issues with semantic interoperability. Yes, federation is still happening at the delivery level, but the semantic issues are out of fucking control. You are a federation by the very sheer skin of your teeth.

The reason why developers are leaving the fediverse is because you folks don’t take criticism. You respond to criticism with — I’m being so serious right now — political manifestos and harassing developers. ActivityPub developers and authors oversold you folks on the capabilities of ActivityStreams. They flat-out lied to y’all.

↬bark.lgbt/@cloudskater/116080965694723006

ActivityPub Server’s Custom Reply‑Control Extensions Undermine Federation

It seems like Activitbypub developers are extending ActivityPub with optional metadata to fix a lot of its issues, but that is still problematic. Trying to add moderation tools and user control to threads seems to be the ongoing battle. I am fascinated by dumpster fires, so I’ve started looking at the ActivityPub protocol in detail. I tend to become fascinated with things that are going down in flames.

As a brief recap of the problem:

So, one of the very popular features on Bluesky—also popular on Twitter—is the ability to select who can reply to a post. A major issue in the Fediverse is the inability to decide who can reply, and once you block someone, their harassing reply is still there. I honestly thought it was simply a case of them choosing not to add or address it for cultural reasons. What is clear from that thread is that they were always aware that the ActivityPub protocol and most Fediverse implementations don’t provide a universal way to control reply visibility or enforce blocks across instances.

An ActivityPub server that has reply control is GoToSocial. ActivityPub, as defined by the W3C specification, standardizes how servers federate activities. It defines actors, inboxes, outboxes, and activity types (Create, Follow, Like, Announce, etc.) expressed using ActivityStreams 2.0. It also specifies delivery mechanics (including how a Create activity reaches another server’s inbox) and how collections behave.

The specification does not include interaction policy semantics such as “only followers may reply” or “replies require manual approval.” There is no field in the normative vocabulary requiring conforming servers to enforce reply permissions. That category of rule is outside the protocol’s defined contract.

GoToSocial implements reply controls through what it calls interaction policies. These appear as additional properties on ActivityStreams objects using a custom JSON-LD namespace controlled by the GoToSocial project.

JSON-LD permits additional namespaced terms. This means the document remains structurally valid ActivityStreams and federates normally. The meaning of those custom fields, however, comes from GoToSocial’s own documentation and implementation. Other servers can ignore them without violating ActivityPub because they are not part of the interoperable core vocabulary.

Enforcement occurs locally. When a remote server sends a reply—a Create activity whose object references another via inReplyTo—ActivityPub governs delivery, not acceptance criteria. Whether the receiving server checks a reply policy, rejects the activity, queues it, or displays it is determined in the server’s inbox-processing code. The decision to accept, display, or require approval happens after successful protocol-level delivery. This behavior belongs to the application layer.

These are server-side features layered on top of ActivityPub’s transport and data model that are not actually part of ActivityPub. The protocol ensures standardized delivery of activities; however, the server implementation defines additional constraints and user-facing behavior. Two GoToSocial instances may both recognize and act on the same extension fields. However, a different implementation, such as Mastodon, has no obligation under the specification to interpret or enforce GoToSocial’s interactionPolicy properties. These fields function as extension metadata rather than protocol requirements.

The semantics of GoToSocial are not part of the specification’s defined vocabulary and processing rules for ActivityPub. They no longer operate purely at the protocol layer; it has become an application-layer contract implemented by specific servers.

Let’s use the AT Protocol as an example. Bluesky’s direct messages (DMs) are not currently part of the AT Protocol (ATProto). The AT Protocol has nothing that specifies anything for DMs, so DMs are not part of the AT Protocol. The AT Protocol was designed to handle public social interactions, but it does not define private or encrypted messaging. Bluesky implemented DMs at the application level, outside of the core protocol. DMs are centralized and stored on Bluesky’s servers. What is happening with servers like GoToSocial is sort of like that. The difference is that the AT Protocol was designed for different app views; ActivityPub was not.

The issue is the divergence in semantic interpretation that emerges at the interpretation layer. ActivityPub standardizes message delivery and defines common activity types. However, it leaves extension semantics and application-layer policy decisions to individual implementations. Servers may introduce custom JSON-LD namespaces and enforce local behaviors, such as reply restrictions, while remaining protocol-compliant. But, the noise created by divergences are problematic, because it creates unexpected, unintended, and unpredictable behavior.

Divergence appears when implementations rely on non-normative metadata and assume reciprocal handling to preserve a consistent user experience. Behavioral alignment then varies. Syntactic exchange succeeds, but behavioral consistency is not guaranteed. Though instances continue to federate at the transport level, policy semantics and processing logic differ across deployments. Those differences produce inconsistent experiences and results between implementations.

That leads to fragmentation, specifically semantic or behavioral fragmentation and an inconsistent user experiences. ActivityPub ensures syntactic interoperability, but semantic interoperability (everyone interprets and enforces rules the same way) varies. This creates a system that is federated at the transport level yet fragmented in behavior and expectations across implementations. It is funny how the thing that the fediverse touted has made the entire thing very brittle. ActivityPub technically federates correctly, but semantically falls apart once servers start adding their own behavioral rules.

Stepping Back From Social Media To Read a Book

I’m taking a break. After spending like two years in the worst parts of the Internet modeling the memetic spread of conspiracy-driven behavioral patterns and developing social media software as a side hustle, I think I’m going to take a step back and… I don’t know… maybe read a book? lol.

I’m a Computational Biologist who pretty much studies the memetics of conspiracy theories and how they act as another vector/epidemiological layer. I’ve also been working on various contracts for social media development stuff. Working on the shit I’ve been working on for years forces you to see the worst parts of people that they split off. It makes you hate everyone — and I mean everyone.

Your BlueSky Feed Is Porn You Didn’t Ask For Because Your Friends Are Gooners With a Severe Porn Addiction

A common complaint I see people make on Bluesky is: why am I being served so much porn or things I am not interested in? They will incorrectly believe that the algorithm is broken. It’s not broken. You didn’t know the people you knew as well as you thought you did. Porn addiction is a thing, and porn addiction is especially common with weebs. You’re seeing deranged shit because people you follow have porn addictions and are into deranged shit. So, though you may not be consuming porn, people in your network are. That activity kicks into your feeds.

The issue I have with that is that it essentially normalizes being sex pests in a space on the Internet. That sets the expectation that it is good—attractive, even—to act like that elsewhere. That expectation alienates relationships. Bluesky creates a cultural space that offers an unrealistic, bizarre representation of social relationships, which isolates and alienates the users who stay on there consuming erotica and porn like they do.

So, user repos in Bluesky have a property for likes. Bluesky’s underlying AT Protocol stores likes as first-class structured records in each user’s AT Protocol repository. In the AT Protocol lexicon, a like is an app.bsky.feed.like record type. Unlike a simple boolean flag on a post, it is its own record with a creation timestamp and a subject field that holds a strong reference to the liked record.

That strong reference is composed of an AT-URI and a CID. The AT-URI identifies the exact record in the network by DID, collection, and record key. The CID is a cryptographic content identifier that uniquely identifies the exact content of that liked record.

These like records exist under the app.bsky.feed.like namespace in the user’s repo. Bluesky’s repo model is built so that these repos are hosted on a user’s Personal Data Server and are publicly readable through the AT Protocol APIs. Because of that, the like record and its fields can be fetched, indexed, and used by any client or service that can query the protocol.

The protocol exposes operations like getLikes. This returns all of the like records tied to a particular subject’s AT-URI and CID. It also exposes getActorLikes. This returns all of the subject references a given actor has liked. Those API calls return structured like objects with timestamps and subject references directly from the public repository data.

Various feeds hosted by different PDSs use the likes property to construct the feeds that you see. Since the likes of people you follow are included in your social graph, along with your own likes, you’re going to get served the porn they are consuming. Because likes are public and anyone can write an algorithm to see everyone’s likes, you can clearly see just how much porn people are consuming.

Honestly, what started to turn my stomach about the people on Bluesky is how they behave across different contexts. If you look through the records of the posts they interact with, you’ll see them engaging with political posts in the replies like a normal person. Then, when you look through their AT Protocol records, you see hours and hours of them interacting with every kind of porn imaginable. I am not exaggerating. Hours of likes for porn posts within 1–10 minutes of each other. Am I sex-negative? A prude? No, this site is filled with furry, gay bara porn, lol. You can have a drink without being an alcoholic. The problem with these people is like people who can’t have one drink without drinking the whole fucking day; they can’t consume porn in healthy ways.

I think people assume that their feed is customized for them and based on their likes. No—feeds are generalized based on what everyone likes and then served to your subgraph. It’s not just about who you follow; it’s about who they follow. So if you follow someone who follows a lot of people with porn addictions, you will see porn. Bluesky isn’t weighting the algorithm to do this. Basically, it’s the people in your social network with furry, hentai, or trans porn addictions who are driving it.

BlueSky’s Solution To Moderating Is Moderating Without Moderating via Social Proximity

I have noticed a lot of people are confused about why some posts don’t show up on threads, though they are not labeled by the moderation layer. Bluesky has begun using what it calls social neighborhoods (or network proximity) as a ranking signal for replies in threads. Replies from people who are closer to you in the social graph, accounts you follow, interact with, or share mutual connections with, are prioritized and shown more prominently. Replies from accounts that are farther away in that network are down-ranked. They are pushed far down the thread or placed behind “hidden replies.”

Each person gets their own unique view of a thread based on their social graph. It creates the impression that replies from distant users simply don’t exist. This is true even though they’re still technically public and viewable if you expand the thread or adjust filters. Bluesky is explicitly using features of subgraphs to moderate without moderating. Their reasoning is that if you can’t see each other, you can’t harass each other. Ergo, there is nothing to moderate.

Bluesky mentions that here:

https://bsky.social/about/blog/10-31-2025-building-healthier-social-media-update

As a digression, I’m not going to lie: I really enjoyed working on software built on the AT protocol, but their fucking users are so goddamn weird. It’s sort of like enjoying building houses, but hating every single person who moves into them. But, you don’t have to deal with them because you’re just the contractor. That is how I feel about Bluesky. I hate the people. I really like the protocol and infrastructure.

I sort of am a sadist who does enjoy drama, so I do get schadenfreude from people with social media addictions and parasocial fixations who reply to random people on Bluesky, because they don’t realize their replies are disconnected from the author’s thread unless that person is within their network. They aren’t part of the conversation they think they are. They’re algorithmically isolated from everyone else. Their replies aren’t viewable from the author’s thread because of how Bluesky handles social neighborhoods.

Bluesky’s idea of social neighborhoods is about grouping users into overlapping clusters based on real interaction patterns rather than just the follow graph. Unlike Twitter, it does not treat the network as one big public square. Instead, it models networks of “social neighborhoods” made up of people you follow, people who follow you, people you frequently interact with, and people who are closely connected to those groups. They’re soft, probabilistic groupings rather than strict labels.

Everyone does not see the same replies. Bluesky is being a bit vague with “hidden.” Hidden means your reply is still anchored to the thread and can be expanded. There is another way Bluesky can handle this. Bluesky uses social neighborhoods to judge contextual relevance. Replies from people inside or near your social neighborhood are more likely to be shown inline with a thread, expanded by default, or served in feeds. Replies from outside your neighborhood are still public and still indexed, but they’re treated as lower-context contributions.

Basically, if you reply to a thread, you will see it anchored to the conversation, and everyone will see it in search results, as a hashtag, or from your profile, but it will not be accessible via the thread of the person you were replying to. It is like shadow-banning people from threads unless they are strongly networked.

Because people have not been working with the AT Protocol like I have, they assume they are shadow-banned across the entire Bluesky app view. No—everyone is automatically shadow-banned from everyone else unless they are within the same social neighborhood. In other words, you are not part of the conversation you think you are joining because you are not part of their social group.

Your replies will appear in profiles, hashtag feeds, or search results without being visually anchored to the full thread. Discovery impressions are neighborhood-agnostic: they serve content because it matches a query, tag, or activity stream. Once the reply is shown, the app then decides whether it’s worth pulling in the rest of the conversation for you. If the original author and most participants fall outside your neighborhood, Bluesky often chooses not to expand that context automatically.

Bluesky really is trying to avoid having to moderate, so this is their solution. Instead of banning or issuing takedown labels to DIDs, the system lets replies exist everywhere, but not in that particular instance of the thread.

I find this ironic because a large reason why many people are staying on Bluesky and not moving to the fediverse—thank God, because I do not want them there—is discoverability, virality, and engagement.

In case anyone is asking how I know so much about how these algorithms work: I was a consultant on a lot of these types of algorithms, so I certainly hope I’d know how they work, lol. No, you get no more details about the work I’ve done. I have no hand in the algorithm Bluesky is using, but I have proposed and implemented that type of algorithm before.

I have an interest in noetics and the noosphere. A large amount of my ontological work is an extension of my attempts to model domains that have no spatial or temporal coordinates. The question is how do you generalize a metric space that has no physically, spatial properties. I went to school to try to formalize those ideas. Turns out they’re rather useful for digital social networks, too. The ontological analog to spatial distance, when you have no space, is a graph of similarities.

Astroturfing Is Pretty Pointless When Social Subgraphs Are Fragmented (e.g., the Fediverse)

I am seeing astroturfing in the fediverse again, by AT Protocol developers implicitly trying to shill their products. I think it is stochastic behavior by developers with too much time on their hands. Honestly, I do not care. I like the people on ActivityPub more, but I like the AT Protocol better, and I have developed for both. Astroturfing on ActivityPub networks is fascinating to me because it is so pointless.

I am actually a Computational Biologist and Computer Scientist whose specialty is combinatorics, social graphs, graph theory, etc. Specifically, I use this to create epidemiological models for the memetic layer of human behaviors that act as vectors for diseases, using the SIRS model. I do not just study germs; I study human behaviors.

The models I construct extend into a “memetic layer,” in which beliefs, norms, and behaviors (such as risk-taking, compliance with public health measures, or susceptibility to misinformation) spread contagiously through social networks. These behaviors function as vectors that modulate biological transmission rates. As a result, the spread of ideas can accelerate, dampen, or reshape the spread of disease. By running computational simulations and agent-based models on these graphs, I study how network structure, influential nodes, clustering, and platform-specific dynamics affect behavioral contagion. I also examine how these factors influence epidemiological outcomes.

To say it very concisely, I study how the spread of bat-shit insane beliefs, shit posts, and memes influences whether or not there is a measles outbreak in Texas. Ironically, this is an evolution of my studying semiotics, memetics, and chaos magick in high school. I got a job where I can use occult, anarchist techniques professionally.

I think a large reason why I do not care about astroturfing in the fediverse is that it’s so pointless, lol. Astroturfing to manipulate the narrative would actually work better on Bluesky to keep people there than trying to recruit from the fediverse. Furthermore, big instances are relatively small. Some people on Bluesky have follower lists larger than an entire large instance in the fediverse.

Within ActivityPub networks, astroturfing rarely propagates far, because whether information spreads depends on properties of the social graph itself. Dense connectivity, short paths between communities, and a sufficient number of cross-cutting ties support diffusion. ActivityPub’s architecture tends to produce graphs that are fragmented and highly modular. This limits the reach of coordinated activity.

ActivityPub is a system where each instance maintains its own local user graph and exchanges activities through inboxes and outboxes. This makes it autonomous and decentralized. The network consists of loosely connected subgraphs. Cross-instance edges appear only through explicit follow relationships. The ActivityPub protocol does not provide a shared or complete view of the network. Measurements of the fediverse consistently show uneven connectivity between instances, clustering at the instance level, and relatively long effective path lengths across the network. Under these conditions, large cascades are uncommon.

Instance-level clustering means that in ActivityPub networks, users interact much more with others on the same server than with users on different servers. Because each instance has its own local timeline, culture, and moderation, connections form densely within instances and only sparsely across them through explicit follow relationships. This creates a network made up of tightly connected local communities linked by relatively few cross-instance ties, which slows the spread of information beyond its point of origin.

However, with the AT Protocol, global indexing and aggregation are explicitly supported. Relays and indexers can assemble near-complete views of the social graph. Applications built on top of this infrastructure operate over a graph that is denser and easier to traverse. There are fewer structural barriers between communities. The diffusion dynamics change substantially when content can move across the graph without relying on narrow federated paths.

Astroturfing depends on coordinated amplification, typically through tightly synchronized clusters of accounts intended to manufacture visibility. Work on coordinated inauthentic behavior shows that these tactics gain traction when they intersect highly connected regions of the graph or bridge otherwise separate communities. In networks with strong modularity, coordination remains local. ActivityPub’s federation model produces this kind of modularity by default. Coordinated clusters stand out clearly within instances. Their effects remain confined to those local neighborhoods.

Astroturfing on ActivityPub therefore tends to stall on its own because of the underlying graph topology. Without dense inter-instance connectivity or any form of global indexing, coordinated campaigns have a hard time moving beyond the immediate regions where they originate. Systems built on globally indexable social graphs, including those enabled by the AT Protocol, expose a much larger surface for viral spread. Network structure and connectivity account for the divergence where that is independent of moderation, cultural norms, ideology, or intent.

It’s just really funny to me how these stochastic techbro groups waste so many resources. I personally don’t want to go viral, which is why I avoid platforms where I can. The fact that it’s harder to achieve high virality on ActivityPub is exactly why I prefer the fediverse over the Atmosphere. One way to think about it is that you can change the ‘genetics’ of a system with a retrovirus, where memetic entities act as cultural retroviruses to reprogram the cultural loci of a space. That is their end goal. They are trying to hijack cultures memetically. You see this a lot with culture jamming.

Basically, the astroturfing on ActivityPub networks is designed to jam and subvert the culture. But, as I have already said, the topological structure makes memetic virality stall. They cannot achieve that kind of viral spread in the fediverse, which is why I cannot understand why they do this every year.

"K.G.M.’s is the first lawsuit to be picked by the court as a so-called “bellwether” trial. Bellwether trials typically occur in situations where there are a large number of plaintiffs who have filed a lawsuit against the same defendant (or defendants) for harm by the same products. A small number of cases are handpicked as test cases to be representative of all the large pool of plaintiffs. The goal of such trials is to help foresee what the future litigation of all cases might look like.

This case has gotten so far because it’s built on an argument that tries to sidestep Section 230. The plaintiffs’ focus is not the liability of the content but the alleged business decisions that shape these platforms. If the legal argument in this trial proves successful, experts believe it could force social media companies to prioritize safety in a way they have not to this point.

“This is going to be the first time a jury is going to hear arguments about what the social media companies knew about the risks of the design of their platforms and how they acted on the types of information they had,” says Haley Hinkle, policy counsel at Fairplay, an organization that works to protect kids from Big Tech. The jury will ultimately decide, she says, whether the companies were negligent, if they contributed to mental health harms, and if they should have warned young users about the risks."

https://www.wired.com/story/meta-google-youtube-social-media-addiction-trial/

#SocialMedia #SocialMediaAddiction #SocialNetworks #Algorithms #RecommendationAlgorithms #TikTok #Meta #Google #BigTech #MentalHealth #YouTube

"The latest X algorithm doesn’t share the weightings used, but we know the old ones. A “like”, for example, is worth a mere 0.5 points. A reply: 13.5. But if your reply sparks the author to reply back? That’s an argument: worth 75 points. These are all added to give a “final score”.

Crucially, this doesn’t (and cannot) judge if you regard a post as interesting, insightful or useful; true or false. All it can do is gauge reaction. To “reply” is normally a sign of disagreement, and that’s rated as 27 times more valuable than a “like”. A debate: 150 times. This creates a bias towards what may rile you but is calculated to keep you just angry enough. Not so angry that you switch off or report a post for bigotry.

It would take seconds to tweak the code and transform the news seen by billions of social media users. The power is huge, as are the implications. Riots in Myanmar were linked to a Facebook algorithm spreading false reports about Muslims raping Buddhist women. (Facebook later admitted it unwittingly created an “enabling environment” for extremists but had no moderators who spoke the language.)

During the 2024 UK riots, posts about migrants and hostels went viral. This is the problem with “neutral” code: unintended consequences can be significant. “We know the algorithm is dumb and needs massive improvements,” Musk said when it was published. “But at least you can see us struggle to make it better in real time and with transparency. No other social media companies do this.”

He’s right. YouTube, Instagram and TikTok will have such algorithms but no one else has published the code. All will have a bias towards engagement and, ergo, figures who trigger strong responses on incendiary topics. Identity politics thrives here: dividing lines of race, religion or gender make for the most potent arguments."

https://www.thetimes.com/comment/columnists/article/anger-algorithm-far-right-x-twitter-instagram-youtube-0r50ccv6v

#SocialMedia #Algorithms #RecommendationAlgorithms #SocialNetworks #Disinformation #Propaganda

"There are many ways in which the new algorithm will be able to influence the platform’s content visibility and hence its overall “political climate”. We may indeed witness changes in moderation, meaning that certain contents and accounts are effectively restricted. Award-winning Palestinian journalist Bisan Owda has said she has been permanently banned from the app as of Wednesday this week. However, it is likely that the most consequential changes will be more in terms of the way the algorithms serve content to users.

The new algorithm will be retrained on US rather than global data. This opens opportunities to introduce biases, with the potential of reinforcing conservative views and sidelining minority ones, while at the same time cutting US debates off from those going on in the rest of the world. Further, weights attributed to different parameters can have important consequences for user experience. As seen with Facebook’s 2018 adoption of the meaningful social interaction framework which down-ranked public and news content, while attributing a high weight to angry reactions, changes to the feed algorithm can have major consequences.

As scholars Kai Riemer and Sandra Peter have pointed out, the way in which algorithms “interfere with free speech on the audience side” highlights the need to reconsider the way we think about public debate in the algorithmic era. It’s not what we can or cannot say that matters; rather, it’s whether what we say can get any visibility at all, and whether it is able to move against the political climate imposed by those controlling platform algorithms."

https://www.theguardian.com/commentisfree/2026/jan/30/tiktok-us-takeover-new-type-of-censorship

#USA #SocialMedia #TikTok #Algorithms #RecommendationAlgorithms #ContentModeration #Censorship

Let’s decouple the algorithm from the platform
https://josschuurmans.com/en/blog/lets-decouple-the-algorithm-from-the-platform/

Profile-based #recommendationalgorithms should be banned. That is the view of @bitsoffreedom, voiced by policy advisor Lotje Beek in the Bits of Freedom podcast of January 9 (https://www.bitsoffreedom.nl/podcast/nieuwe-europese-wetgeving-de-digital-fairness-act).

Together with a team from the South-Eastern Finland University of Applied Sciences, I am working on an #opensource system for transparent #information_recommendation, based on #collaborative_filtering.

Tubefilter: The YouTube Shorts algorithm appears to have changed to prioritize newer uploads. “Recently, YouTube Shorts content strategists have noticed a curious trend across channel analytics dashboards. On some channels, viewership of older Shorts has declined precipitously, suggesting that YouTube is tinkering with its recommendation algorithm to favor newer uploads.”

https://rbfirehose.com/2025/12/06/tubefilter-the-youtube-shorts-algorithm-appears-to-have-changed-to-prioritize-newer-uploads/

Clemson University: Is it possible to get your daily news in a smarter way? Clemson researchers are working on it– and you can help. “A team that includes Clemson University is inviting volunteers to sign up for a free daily newsletter that delivers a list of Associated Press news stories personally curated for each individual reader. The ad-free newsletter is helping researchers study […]

https://rbfirehose.com/2025/11/12/clemson-university-is-it-possible-to-get-your-daily-news-in-a-smarter-way-clemson-researchers-are-working-on-it-and-you-can-help/

Tubefilter: What would happen if you could talk directly to the recommendation algorithm? X is about to find out. “Now, nearly three years after Musk’s purchase, X seems to have finally put two and two together and realized some users may not like seeing this content. To fix that problem, it’s leaning hard into another arm of Musk’s business: generative AI.” Out of the frying pan and into […]

https://rbfirehose.com/2025/09/24/tubefilter-what-would-happen-if-you-could-talk-directly-to-the-recommendation-algorithm-x-is-about-to-find-out/

Two observations on #recommendationalgorithms:

1. After hundreds of millions of dollars spent on development, it still seems like the best they can do is "here are 10 more things identical to what you just chose." Which hardly seems worth it.

2. The use of these algorithms has led to the elimination of straightforward #settings that we used to be able to adjust. I don't watch sports, but because of these algorithms, there is no way for me to "turn sports off" anymore.

#enshittification

(3/3)

The recommendation engines we habitually call "algorithms" have actually been *models*, not algorithms. So "AI" is just a continuation of the automated disinformation-as-usual that corporate platforms have been engaged in for almost 20 years.

#AI #MOLE #RecommendationAlgorithms

BBC: The hidden world beneath the shadows of YouTube’s algorithm. “That’s the vision of YouTube the company promotes – slick, professional, entertaining and loud – but from one perspective, it’s all a façade. Through another lens, the essence of YouTube is more like this video from 2020. Before I watched, it had only been seen twice. A man points the camera out of his bedroom window as a […]

https://rbfirehose.com/2025/03/23/bbc-the-hidden-world-beneath-the-shadows-of-youtubes-algorithm/

BBC: Telegram pushes extremist groups to users – study. “The report, from the US civil rights organisation the Southern Poverty Law Center (SPLC), found that the ‘similar channels’ feature introduced last year recommends extremist channels even to users browsing subjects such as celebrities or technology.”

https://rbfirehose.com/2024/12/17/bbc-telegram-pushes-extremist-groups-to-users-study/

#SocialMedia #TikTok #Teens #RecommendationAlgorithms: "As TikTok’s 170 million U.S. users can attest, the platform’s hyper-personalized algorithm can be so engaging it becomes difficult to close the app. TikTok determined the precise amount of viewing it takes for someone to form a habit: 260 videos. After that, according to state investigators, a user “is likely to become addicted to the platform.”

In the previously redacted portion of the suit, Kentucky authorities say: “While this may seem substantial, TikTok videos can be as short as 8 seconds and are played for viewers in rapid-fire succession, automatically,” the investigators wrote. “Thus, in under 35 minutes, an average user is likely to become addicted to the platform.”

Another internal document found that the company was aware its many features designed to keep young people on the app led to a constant and irresistible urge to keep opening the app.

TikTok’s own research states that “compulsive usage correlates with a slew of negative mental health effects like loss of analytical skills, memory formation, contextual thinking, conversational depth, empathy, and increased anxiety,” according to the suit.

In addition, the documents show that TikTok was aware that “compulsive usage also interferes with essential personal responsibilities like sufficient sleep, work/school responsibilities, and connecting with loved ones.”"

https://www.npr.org/2024/10/11/g-s1-27676/tiktok-redacted-documents-in-teen-safety-lawsuit-revealed

I wish content recommendation algorithms would pay less attention to that one mainstreamish celebrity trash thing I clicked once and focus on my years of engagement with more niche interests and hobbies. You click on one cute bird video, and your entire feed is filled with nothing but birds forever from that point on. #RecommendationAlgorithms

Today #Twitter released much of the code used for their recommendation algorithm https://blog.twitter.com/engineering/en_us/topics/open-source/2023/twitter-recommendation-algorithm

An machine-learning system relies on both an algorithm and training data so I wonder exactly what insights can be gained from what's been made public (I'm definitely not an expert in this area so I invite corrections and clarifications here). Regardless, it's an unusual level of transparency for a major social-media platform.

#SocialMedia #RecommendationEngines #RecommendationAlgorithms