Do Not Recommend? Reduction as a Form of Content Moderation

Gillespie, Tarleton

doi:10.1177/20563051221117552

Cited by 75 publications

(39 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Governance by platforms includes the massive content moderation apparatuses, outsourced in large part to contracting firms, that are responsible for detecting and reviewing content uploaded to social media platforms that violate the platform's community guidelines (Gillespie, 2018; Roberts, 2019). Both academic and industry reports tend to disproportionately frame the outcomes of content moderation as a process of content removal instead of the more common reduction in visibility, wherein content remains available on a platform but must be manually located rather than algorithmically curated for users (Gillespie, 2022; Zeng & Kaye, 2022).…”

Section: Content Moderation On Social Media Platformsmentioning

confidence: 99%

“…Mass reporting can be frustrating for casual users of social media platforms, but for full‐time social media creators, the threat of content takedowns, account suspensions and bans, and reductions in visibility through algorithmic recommendations can have severe and lasting consequences for their professional images, financial livelihoods, and emotional well‐being (Are, 2022; Bishop, 2019; Cotter, 2021; Duffy & Meisner, 2023). Research on content takedowns in the creator economy rightly addresses unfairness in automated content moderation (e.g., algorithmic reduction; see Gillespie, 2022), though considerably less is known about the experiences of creators attempting to manage abuses of the infrastructures for platform governance, such as the role of user‐driven flagging in content reporting and moderation (Crawford & Gillespie, 2016).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The weaponization of platform governance: Mass reporting and algorithmic punishments in the creator economy

Meisner

2023

Policy & Internet

View full text Add to dashboard Cite

Amid wider discussions of online harassment on social media platforms, recent research has turned to the experiences of social media creators whose compulsory visibility renders them vulnerable to frequent attacks, ranging from persistent trolling to much more insidious, offline acts of violence. This study examines a contemporary form of harassment against social media creators known as “mass reporting,” or the targeted, automated flagging of creators' online content to trigger content takedowns and account bans. Mass reporting is particularly challenging for social media creators because of its weaponization of platform infrastructures for community governance, leaving creators with few avenues of support after orchestrated attacks and restricting their access to platform support. Drawing on in‐depth interviews with social media creators who have been subjected to mass reporting on their content, this study builds an understanding of the ways in which tools for platform governance, such as content reporting, can be weaponized to harass and introduce vulnerabilities for creators.

show abstract

Section: Content Moderation On Social Media Platformsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The weaponization of platform governance: Mass reporting and algorithmic punishments in the creator economy

Meisner

2023

Policy & Internet

View full text Add to dashboard Cite

show abstract

“…Virtually all social media networks have established rules of conduct and acceptable content for their services. They have a range of techniques at their disposal to enforce these standards, such as deleting [40,58], downranking [31], quarantining [11,56], labeling [69], or demonetizing [15] undesirable content, or temporarily or permanently banning the accounts of users who repeatedly post violating content ("deplatforming") [19,39,51].…”

Section: Content Moderationmentioning

confidence: 99%

Understanding the (In)Effectiveness of Content Moderation: A Case Study of Facebook in the Context of the U.S. Capitol Riot

Goldstein¹,

Edelson²,

McCoy³

et al. 2023

Preprint

View full text Add to dashboard Cite

Social media networks commonly employ content moderation as a tool to limit the spread of harmful content. However, the efficacy of this strategy in limiting the delivery of harmful content to users is not well understood. In this paper, we create a framework to quantify the efficacy of content moderation and use our metrics to analyze content removal on Facebook within the U.S. news ecosystem. In a data set of over 2 𝑀 posts with 1.6 𝐵 user engagements collected from 2,551 U.S. news sources before and during the Capitol Riot on January 6, 2021, we identify 10,811 removed posts. We find that the active engagement life cycle of Facebook posts is very short, with 90 % of all engagement occurring within the first 30 hours after posting. Thus, even relatively quick intervention allowed significant accrual of engagement before removal, and prevented only 21 % of the predicted engagement potential during a baseline period before the U.S. Capitol attack. Nearly a week after the attack, Facebook began removing older content, but these removals occurred so late in these posts' engagement life cycles that they disrupted less than 1 % of predicted future engagement, highlighting the limited impact of this intervention. Content moderation likely has limits in its ability to prevent engagement, especially in a crisis, and we recommend that other approaches such as slowing down the rate of content diffusion be investigated.

show abstract

“…The 15 most popular social media platforms in the United States each have different definitions of harassment in their community policies (Pater et al, 2016) reflecting not only a technical problem for platform operations but also a challenge to platforms' abilities to recognize and repair harm for the diverse communities they aim to serve (Schoenebeck & Blackwell, 2021). Given that all platforms must moderate the glut of content uploaded to their sites every second (Gillespie, 2018;Roberts, 2019), most major platforms have implemented extensive content moderation apparatuses that remove content and accounts as well as algorithmic recommendation systems that render some content and accounts more or less visible (Gillespie, 2022). Content moderation-especially algorithmic content moderation (Gorwa et al, 2020)-not typically embraced content and cultural expressions deemed non-normative, including from already marginalized groups (Duffy & Meisner, 2022;Thach et al, 2022).…”

Section: Governing Harm and Toxicity On Social Mediamentioning

confidence: 99%

Networked Responses to Networked Harassment? Creators’ Coordinated Management of “Hate Raids” on Twitch

Meisner

2023

Social Media + Society

View full text Add to dashboard Cite

This study investigates how social media creators navigate and respond to severe cases of ongoing, networked harassment. Drawing on 19 in-depth interviews with Twitch streamers who experienced a form of networked harassment known as “hate raids” on the platform, including three creators who built and shared tools to combat these attacks, this analysis pays particular attention to the nature and coordination of responses to networked harassment and the extent to which creators’ responses are also networked. This study’s findings suggest that, in the absence of communication and technical support from Twitch, creators started ad hoc networks for sharing technical tools, offering strategies for managing audiences during attacks, and providing emotional support for peer creators. Yet these networks of support were unevenly accessed across the streamers interviewed in this study and often existed only temporarily. This study’s findings furthermore indicate that communities on Twitch form primarily around individual streamers, which fosters supportive connections within that streamer’s audience community but limits opportunities for solidarities to form across streamers as a class of creative workers. I conclude by considering more broadly how platform infrastructures can facilitate or constrain different forms of community building among creators and their audiences.

show abstract

Do Not Recommend? Reduction as a Form of Content Moderation

Cited by 75 publications

References 62 publications

The weaponization of platform governance: Mass reporting and algorithmic punishments in the creator economy

The weaponization of platform governance: Mass reporting and algorithmic punishments in the creator economy

Understanding the (In)Effectiveness of Content Moderation: A Case Study of Facebook in the Context of the U.S. Capitol Riot

Networked Responses to Networked Harassment? Creators’ Coordinated Management of “Hate Raids” on Twitch

Contact Info

Product

Resources

About