Show HN: A personal YouTube frontend based on yt-dlp

Original Article ↗ Hacker News Discussion ↗

DIY YouTube Frontends & Workflows

  • Many commenters have built similar personal tools around yt-dlp: simple web/PHP/Flask frontends, Plex/Jellyfin pipelines, bookmarklets, and mpv-based queues.
  • Common goals: automated downloading, transcoding to device‑friendly formats, audio-only extraction, playlist handling, filesystem organization, and integration with media servers.
  • Several mention this as a fun “small project” space, even for non–web developers.

Motivations: Enshittification, UI, and Control

  • Strong dissatisfaction with YouTube’s UI/UX: clutter, dark patterns, shorts, community posts in the subscriptions feed, and poor performance (especially on some Firefox setups and TV devices).
  • People want chronological, subscription‑only feeds and fewer distractions. Tips include disabling watch history, using subscription URLs directly, and extensions like Unhook, h264ify, SmartTube, Vinegar, FreeTube, NewPipe.
  • Broader theme of “enshittification of the web” and desire for self‑hosted frontends (SearXNG, Matrix, etc.) to regain control and privacy.

Discovery vs. Going Off‑Platform

  • Concern: watching via third‑party frontends breaks recommendation feedback loops and may cause YouTube to repeatedly recommend already‑watched videos.
  • Some argue discovery is mostly a social problem: a small set of “cornerstone” creators and communities (RSS, HN, niche feeds) can provide more than enough content.
  • Others value serendipitous algorithmic discovery and see real loss in leaving the official app.
  • Ideas raised: local/custom recommenders using signals from HN/Reddit, third‑party “algorithm engines,” or using the official app only for discovery then sharing URLs to an ad‑free player.

DRM, Ads, Legality, and yt-dlp Risk

  • Debate over DRM: one side calls it useless and punitive; another argues it’s “working as intended” for non‑paying users.
  • Some fear popularizing frontends will increase pressure to break yt-dlp (citing youtube-dl’s DMCA issues). Others reject this gatekeeping and blame YouTube’s product choices for demand.
  • Use cases beyond “free stuff” are emphasized: archival, fair‑use clipping, public‑domain work, forensic/archival research, and professional creator workflows.
  • Discussion of ad funding vs. privacy: whether users have any obligation to watch ads, and whether ad‑tracking practices justify blocking.
  • Legal nuances: some countries grant a right to private copies even for copyrighted works; monetizing such tools is viewed as especially risky.

Children, Safety, and Monetization

  • Parents report using curated channel whitelists and “safe” apps to shield kids from low‑quality or harmful content.
  • There is interest in network‑wide or offline solutions built around yt-dlp, but also warnings that monetizing such systems could make them legal targets.

Technical Notes & Feature Requests

  • yt-dlp’s SponsorBlock integration, cookie handling, Docker images, and quality selection are discussed.
  • Some want simpler desktop packaging (single executable/Electron) versus self‑hosted servers.
  • LLM-based layers (LM Studio/Ollama, Videocrawl) are highlighted for transcripts, summaries, and code/reference extraction from videos.