Thread

JeetBlock - Demographic-Based Social Media Filtering

Analysis of a controversial user-level content curation tool

Today we're diggin' into the source code of JeetBlock, a browser extension that enables users to filter social media content based on cultural and demographic signals. The extension represents an interesting approach to addressing user demand for more granular control over their social media experience.

How It Operates: A Technical Overview

[Scroll to the next section if you don't care about the technicalities]

The control flow follows a straightforward pattern:

  1. Loading Filter Criteria The extension begins by populating its core filtering lists through functions and properties like loadIndianNames(), loadMuslimNames(), ukraineEmojis and lgbtqiaEmojis. This creates comprehensive datasets of names, keywords, and cultural identifiers that users can leverage for content filtering. The system includes user-editable allow-lists and custom blocklists for personal customization.

  2. Real-time Content Monitoring After initialization, the extension deploys HTML element observers across major social platforms including Facebook, Twitter/X, TikTok, Instagram, LinkedIn, YouTube, and Reddit. Using MutationObservers, it monitors for new content across feeds, messages, comments, and search results. When new content appears, it's evaluated against the loaded criteria, with matching content filtered from view in real-time.

Technical Assessment

  • Filtering Methodology: The approach uses cultural and demographic identifiers as primary filtering criteria, providing users with broad control over their social media environment. The system includes demographic-specific toggles and user-level blocklist/allow-list functionality to help refine results. This creates effective demographic-based filtering that inevitably produces false positives by design.

  • Implementation Strategy: The codebase employs platform-specific MutationObserver patterns tailored to each social network's interface. As with many content-filtering tools that rely on DOM observation, the element identification may require maintenance when platforms update their interfaces. Given the extension doesn't auto-update through official stores, this creates a ticking timebomb for breakage.

  • Future Development: The developer (@0biJon) has discussed incorporating X's account origin country data to enhance filtering precision, which could provide additional dimensions for user customization. While this could improve accuracy for X usage, it would require centralized storage and regular updates - essentially building the very infrastructure the extension currently avoids.

Observations

JeetBlock offers a steamroll approach to social media curation: demographic and cultural signal filtering applied at scale. It makes no attempt to hide its targeting methodology, embracing it as a feature rather than a limitation.

However it also represents a meaningful attempt at addressing the clear user interest in demographic-based content filtering. While the current implementation takes a shotgun approach, it acknowledges a reality that has existed in human social behavior for generations: people naturally gravitate toward content and interactions that align with their subjective interests and preferences, and, just as instinctively, will ignore what does not.

What's most notable is how the extension formalizes a behavior that users have long practiced manually: curating their digital environments according to personal preferences. By making this process more systematic and concrete by clearly allowing the option to e.g. "hide all muslim, lgbtq content, no exceptions", it opens up broader conversations about user agency in social media consumption.

Concerns

The claims the website makes versus the reality... It doesn't actually "block" in the traditional sense - it hides content. And while it might catch some "scammers, extremists" by blocking entire demographics, it certainly won't catch all of them. "Demographic-based content local-first curation tool" feels more accurate than the marketing claims.

This was interesting.

For anyone who got to this point, well done!

The first one in the comments to point out a glaring design issue which exists in the extension will get zapped (don't expect a big prize tho). While there's probably more than one flaw, tell me which one you think is on my mind.

Feel free to share your thoughts and findings!

*Disclaimer: This analysis is based on examination of the extension's source code and public developer statements. The extension was not executed live during this assessment.*

*Disclaimer 2: This analysis is presented for informational purposes only. I strongly discourage any form of harassment, organized reporting, or other harmful actions against the creator or their work. Engaging in such behavior is unacceptable and does not have my endorsement.*

*AI Transparency: This piece was drafted with the help of GenAI and was thoroughly reviewed, edited, and approved by a human author. *

Replies (0)

No replies yet. Be the first to leave a comment!