Knowledge base and research workspace for the Center for Cooperative Media (CCM) at Montclair State University.
Contains CCM staff and program profiles, scraped web archives, annual report references, and a multi-platform video content analysis of NYC Mayor Zohran Mamdani's social media.
mooc/
├── ccm-profiles/
│ ├── staff/ # 20 staff profiles
│ └── projects/ # 22 project/program profiles
├── mamdani-video-analysis/
│ ├── downloads/ # 76 videos from 5 platforms (gitignored)
│ ├── transcripts/ # Whisper transcripts (JSON + text)
│ ├── frames/ # Extracted frames at 3s intervals (gitignored)
│ ├── frame-analysis/ # Vision analysis of on-screen text and visuals
│ ├── analysis/ # Topic, sentiment, and cross-platform analysis
│ ├── scripts/ # Pipeline automation (download, transcribe, extract, analyze)
│ ├── web/ # Interactive dashboard
│ └── metadata.json # Master video index
├── .firecrawl/ # 13 scraped web snapshots (markdown + JSON)
├── reports/ # Annual reports (PDFs not tracked -- see reports/README.md)
├── CLAUDE.md # Claude Code project instructions
├── CONTRIBUTING.md # How to add or edit profiles
└── README.md
The Center for Cooperative Media is a grant-funded program based at Montclair State University's School of Communication and Media. It works to grow and strengthen local media through collaboration, training, shared services, and cooperative projects.
- Director: Stefanie Murray
- Flagship program: NJ News Commons -- a network of 300+ news and information providers across New Jersey
- Website: centerforcooperativemedia.org
Each staff profile follows a standard format:
- Name, title, email, social links
- Background and education
- Areas of focus
- Role at CCM
Browse them in ccm-profiles/staff/.
Each project profile includes:
- Type, scope, website, membership info
- Overview and key features
- History, funders, and staff leads
Browse them in ccm-profiles/projects/.
A content analysis of ~76 videos from NYC Mayor Zohran Mamdani's social media accounts (Twitter/X, TikTok, YouTube, Instagram, Facebook), covering November 2025 through April 2026.
Pipeline: Download videos with yt-dlp, transcribe with Whisper (GPU-accelerated), extract frames at 3-second intervals, analyze on-screen text with AI vision, run keyword-based topic/sentiment analysis, and present results in an interactive dashboard.
Key findings (from transcript analysis):
- Top topics: governance (4,115 mentions), economy (1,925), housing (1,830)
- Dominant tone: celebratory (30/76 videos), followed by persuasive (22/76)
- 307 minutes of video content, 47,448 words transcribed
- Facebook videos are longest (avg 12 min, mostly press conferences), while TikTok/Twitter/Instagram are short-form (avg 1-2 min)
Dashboard: Run cd mamdani-video-analysis && python -m http.server 8888 and open http://localhost:8888/web/index.html.
See CLAUDE.md for pipeline details and known issues.
PDF annual reports from 2020--2024 are available at centerforcooperativemedia.org/annual-reports/. They are not tracked in this repo due to file size. See reports/README.md for details.
The .firecrawl/ directory contains point-in-time snapshots of CCM website pages, scraped as markdown. These serve as reference material and are not automatically updated.
This repository is licensed under CC BY 4.0. You are free to share and adapt the contents with attribution.