Incident: Twitter's Fleets Feature Plagued with Glitches and Moderation Issues

Published Date: 2020-11-20

Postmortem Analysis
Timeline 1. The software failure incident with Twitter's Fleets feature happened in November 2020 as per the article [107906].
System The software failure incident reported in the article is related to Twitter's rollout of the Fleets feature. The systems that failed in this incident include: 1. Fleets feature rollout on Twitter - The Fleets feature, which is Twitter's version of disappearing posts, was plagued with glitches and stability problems, leading to the pause in the rollout of the feature [107906].
Responsible Organization 1. Twitter [107906]
Impacted Organization 1. Twitter [107906]
Software Causes 1. Glitches in the Fleets feature causing performance and stability problems [107906] 2. Loophole allowing users to tag others in Fleets even if they have been blocked, leading to potential harassment [107906]
Non-software Causes 1. Timing of the launch coinciding with a contentious election in the US, leading to increased scrutiny and pressure on Twitter [107906]. 2. Challenges related to moderating misinformation, harassment, and abuse on ephemeral content like Fleets [107906]. 3. Concerns about potential loopholes in the product, such as the ability for someone to tag a user in a Fleet even if they have been blocked, leading to potential harassment [107906].
Impacts 1. The software failure incident with Twitter's Fleets feature led to glitches and performance issues, prompting the company to pause the rollout of the feature to address these problems [107906]. 2. The glitch-ridden rollout of Fleets raised concerns about the readiness of the product, particularly in terms of effectively moderating misinformation, harassment, and abuse on content that disappears after 24 hours [107906]. 3. Users reported that some misinformation and hateful content shared in Fleets were not consistently flagged or moderated, indicating a potential impact on the spread of false information on the platform [107906]. 4. Concerns were raised about a loophole in Fleets that allowed users to tag individuals who had blocked them, potentially leading to harassment issues [107906]. 5. The software failure incident highlighted challenges for Twitter in enforcing its rules and policies effectively in the context of ephemeral content like Fleets, where moderation actions need to be taken swiftly before the content disappears [107906].
Preventions 1. Conducting more thorough testing and quality assurance before rolling out the Fleets feature to identify and address glitches and loopholes [107906]. 2. Implementing stricter moderation policies and automated systems to effectively manage misinformation and abuse on Fleets [107906]. 3. Enhancing user reporting mechanisms to quickly flag and address abusive or harmful content on Fleets [107906]. 4. Improving consistency in applying warning labels to misinformation and conspiracy theories shared on Fleets to maintain credibility and combat harmful content [107906].
Fixes 1. Implement stricter moderation policies and enforcement mechanisms for Fleets to tackle misinformation and abuse effectively [107906]. 2. Enhance automated systems to detect and flag harmful content in Fleets more accurately and consistently [107906]. 3. Address loopholes such as the ability to tag users in Fleets even if they are blocked to prevent potential harassment [107906]. 4. Improve performance and stability of the Fleets feature to ensure a smoother user experience [107906].
References 1. Twitter spokesperson Aly Pavela 2. Extremism researcher Marc-Andre Argentino 3. Christine Su, senior product manager at Twitter [107906]

Software Taxonomy of Faults

Category Option Rationale
Recurring one_organization (a) The software failure incident having happened again at one_organization: - Twitter's new feature, Fleets, faced glitches and issues, leading to the pause of the rollout of the feature [107906]. - Twitter had previously made changes to its platform and created new policies to minimize the harmful impacts of misinformation ahead of the US election [107906]. - Twitter's efforts to tackle misinformation and abuse on Fleets are being questioned, especially regarding the effectiveness of moderation on ephemeral content [107906]. (b) The software failure incident having happened again at multiple_organization: - The article does not provide information about similar incidents happening at other organizations or with their products and services.
Phase (Design/Operation) design The software failure incident related to the development phases mentioned in the articles is primarily associated with the design phase. Twitter's new feature, Fleets, faced glitches and challenges related to misinformation moderation and abuse control, indicating issues introduced during the system development and rollout process [107906]. Additionally, concerns were raised about loopholes in the design, such as the ability for blocked users to tag others in Fleets, potentially leading to harassment [107906]. These issues point towards failures in the design phase of the software development lifecycle.
Boundary (Internal/External) within_system, outside_system (a) within_system: The software failure incident related to Twitter's Fleets feature was primarily due to glitches within the system. The article mentions that Fleets had been plagued with glitches, causing Twitter to pause the rollout of the feature. Additionally, concerns were raised about the effectiveness of Twitter's moderation efforts within Fleets, as evidenced by the experiment conducted by Marc-Andre Argentino where misinformation and hateful content in Fleets were not consistently labeled or moderated [107906]. (b) outside_system: The software failure incident related to Twitter's Fleets feature also involved factors originating from outside the system, such as the contentious political environment surrounding the US election. The timing of the launch of Fleets, amidst a period of heightened misinformation and baseless claims, added external pressure on Twitter to effectively moderate content within the feature. The need to address misinformation and abuse related to political events like the election posed challenges that extended beyond the internal workings of the system [107906].
Nature (Human/Non-human) non-human_actions, human_actions (a) The software failure incident related to non-human actions in the articles is primarily due to glitches and performance/stability problems with Twitter's new feature called Fleets. The rollout of Fleets had to be paused as the feature was plagued with glitches, indicating issues introduced without direct human participation [107906]. (b) On the other hand, the software failure incident related to human actions in the articles is seen in the challenges Twitter faces in effectively moderating misinformation, harassment, and abuse on the Fleets feature. The company relies on user flags and automated systems to manage abuse on Fleets, and there are concerns about the effectiveness of these moderation efforts [107906].
Dimension (Hardware/Software) software (a) The software failure incident related to hardware: - The article does not mention any specific hardware-related issues contributing to the software failure incident. Therefore, it is unknown if hardware played a role in the glitches affecting Twitter's Fleets feature [107906]. (b) The software failure incident related to software: - The software failure incident with Twitter's Fleets feature was primarily due to glitches in the software, causing performance and stability problems. Twitter had to pause the rollout of Fleets to address these issues [107906].
Objective (Malicious/Non-malicious) non-malicious (a) The software failure incident related to Twitter's Fleets feature can be categorized as non-malicious. The glitches and issues with moderation and misinformation on Fleets were not intentional acts to harm the system but rather resulted from challenges in implementing and managing the new feature [107906].
Intent (Poor/Accidental Decisions) poor_decisions [a] The intent of the software failure incident can be attributed to poor decisions made by Twitter in rolling out the Fleets feature. The article mentions that Twitter launched Fleets despite the feature being plagued with glitches, raising questions about whether the product was fully ready for release [107906]. Additionally, concerns were raised about the effectiveness of Twitter's moderation efforts on Fleets, with examples of misinformation and hateful content not being consistently flagged or moderated, indicating potential shortcomings in decision-making regarding content moderation [107906].
Capability (Incompetence/Accidental) development_incompetence (a) The software failure incident related to development incompetence can be seen in the case of Twitter's Fleets feature. The article mentions that Fleets, Twitter's version of disappearing posts, was plagued with glitches, raising questions about whether the product was fully ready for rollout [Article 107906]. Additionally, there were concerns about how effectively Twitter would tackle misinformation or abuse in Fleets, with examples of misinformation not consistently being flagged or moderated properly [Article 107906]. (b) The software failure incident related to accidental factors is evident in the glitches and lag experienced with Twitter's Fleets feature. The article mentions that Twitter had to pause the rollout of Fleets to fix performance and stability problems, indicating that these issues were unintentional and not part of the initial design or plan [Article 107906].
Duration temporary (a) The software failure incident related to Twitter's Fleets feature can be considered temporary. The article mentions that Twitter had to pause the rollout of Fleets due to glitches and performance issues [107906]. Additionally, Twitter mentioned that they were "slowing down" the rollout of Fleets to fix "performance and stability problems" [107906]. These indicate that the failure was temporary and caused by specific circumstances rather than being permanent.
Behaviour crash, omission, value, byzantine, other (a) crash: The article mentions that Twitter had to pause the rollout of its Fleets feature due to glitches, indicating a failure in the system losing its state and not performing its intended functions [107906]. (b) omission: The article highlights concerns about Twitter's ability to effectively moderate misinformation and abuse in Fleets, with examples of misinformation not being flagged or labeled, suggesting a failure of the system to perform its intended functions at instances [107906]. (c) timing: The timing of Twitter's launch of the Fleets feature is questioned in the article, especially given the contentious election in the US and the ongoing issues with misinformation on the platform, indicating a failure related to the system performing its intended functions at a potentially inappropriate time [107906]. (d) value: The article discusses instances where misinformation and conspiracy theories were not consistently flagged or labeled in Fleets, indicating a failure of the system to perform its intended functions correctly in terms of providing accurate information [107906]. (e) byzantine: The article mentions inconsistencies in how Twitter handled different types of misinformation in Fleets, with some content receiving warning labels while others did not, suggesting a failure related to the system behaving erroneously with inconsistent responses and interactions [107906]. (f) other: The article also points out a loophole in the Fleets feature that allows someone to tag a user even if they have been blocked, raising concerns about potential harassment, which could be categorized as a failure due to the system behaving in a way not described in the other options [107906].

IoT System Layer

Layer Option Rationale
Perception None None
Communication None None
Application None None

Other Details

Category Option Rationale
Consequence harm, theoretical_consequence The consequence of the software failure incident discussed in the article is primarily related to potential harm and theoretical consequences: - Harm: The software failure in Twitter's Fleets feature raised concerns about the platform's ability to effectively moderate misinformation, harassment, and abuse, potentially leading to harm to users. Extremism researcher Marc-Andre Argentino experimented with sharing misinformation and hateful content in Fleets, highlighting the potential harm if such content is not appropriately moderated [107906]. - Theoretical Consequence: There were discussions about the potential consequences of misinformation and abuse in Fleets, with concerns raised about the effectiveness of Twitter's moderation efforts and the impact of unmoderated content on the platform. The article mentions that Fleets disappearing quickly could make it harder for Twitter to take action fast enough against harmful content, leading to theoretical consequences of inadequate moderation [107906].
Domain information <Article 107906> Twitter's software failure incident with the Fleets feature is related to the information industry. The Fleets feature on Twitter, which is a form of ephemeral content, was intended to provide users with a way to share posts that disappear after 24 hours, similar to the "stories" feature on other social media platforms. However, the feature was plagued with glitches and challenges related to moderation of misinformation and abuse [107906].

Sources

Back to List