Live data from Hacker News

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

github.com

41–50 of 259 posts

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

#41

How does yt-dlp work with sponsorblock? Does it download the video can snip out segments? I wish PLEX still had youtube plugin. Right now I have a googlesheet script that adds latest videos of channels into various playlists on my premium account. Keeps things simple bouncing between devices / chromecast.

Just recently I stumbled upon these options of yt-dlp, but haven't had the chance to dig deeper (sorry in advance for the formatting):

  SponsorBlock Options:
    Make chapter entries for, or remove various segments (sponsor, introductions, etc.) from downloaded YouTube videos using the SponsorBlock API (https://sponsor.ajay.app)

    --sponsorblock-mark CATS                                         SponsorBlock categories to create chapters for, separated by commas. Available categories are sponsor, intro, outro,
                                                                     selfpromo, preview, filler, interaction, music_offtopic, poi_highlight, chapter, all and default (=all). You can prefix the
                                                                     category with a "-" to exclude it. See [1] for descriptions of the categories. E.g. --sponsorblock-mark all,-preview [1]
                                                                     https://wiki.sponsor.ajay.app/w/Segment_Categories
    --sponsorblock-remove CATS                                       SponsorBlock categories to be removed from the video file, separated by commas. If a category is present in both mark and
                                                                     remove, remove takes precedence. The syntax and available categories are the same as for --sponsorblock-mark except that
                                                                     "default" refers to "all,-filler" and poi_highlight, chapter are not available
    --sponsorblock-chapter-title TEMPLATE                            An output template for the title of the SponsorBlock chapters created by --sponsorblock-mark. The only available fields are
                                                                     start_time, end_time, category, categories, name, category_names. Defaults to "[SponsorBlock]: %(category_names)l"
    --no-sponsorblock                                                Disable both --sponsorblock-mark and --sponsorblock-remove
    --sponsorblock-api URL                                           SponsorBlock API location, defaults to https://sponsor.ajay.app

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

#42

Hi Chris, do you know how to handle issues with cookies in production? It seems yt-dlp works fine, but once put in a cloud runner, it doesn't work. Coincidentally, I was also working with yt-dlp this week for another reason.

the project currently supports cookies (never use your own though, of your google profile), just place them in cookies.txt in the root of the project. but it didn't seem to work well on my server, on a residential IP it works well

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

#44

This is awesome and it's one of the countermeasures that the book Chokepoint Capitalism proposed against enshittification. Imagine seeing Twitch, Nebula, Youtube, etc all in one aggregator app, then the switching cost of leaving one platform to another goes way down. If a content creator wanted to move from one platform to another to get a better deal, the users would hardly notice. Unfortunately I think DRM + DMCA m…

[dead]

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

#45
post #12

As a user of a Firefox-based browser, YouTube's performance really is hit or miss. Sometimes it's ok, other times it's barely useable. These days I simply queue up videos in mpv. It is much lighter on the resources, and also provides a nice cache that makes seeking through videos a breeze. I can open a link straight in mpv using a very nice system[1]. Once I have an mpv instance open I simply drag links on top of it…

I use h264ify plugin and didn't see performance issues for playback. The UI depends on test group you go into, but only the first load is really terrible.

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

#46
post #30
post #28

A question for the author or anyone else who has experience in similar solutions. Is there any good solution for discovering new content? Much of the time, I want to stick to my subscriptions, but I do enjoy content surfaced by the algorithm at least once weekly, sometimes more often. My concern in taking my viewing off-platform is twofold: 1) going to YouTube will prompt me with all the stuff I've already watched of…

I view Discovery as a social problem where the content you want is almost always clustered between a relatively small number of creators, regions, etc. Technically it then becomes less of an indexing everything problem and more of a find a few cornerstone creators, say Khan academy, and occasionally branching out. So to answer your question I don’t thing the cost/benefit for automating discovery is much better then s…

There’s a lot of truth to this but one of the most powerful elements of a discovery algorithm is finding things you completely did not look for, ie Christopher Columbus and the western continents. Like your cornerstones are iOS and recipe videos but you discover the right dance video and it changes your whole life.

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

#47
post #28

A question for the author or anyone else who has experience in similar solutions. Is there any good solution for discovering new content? Much of the time, I want to stick to my subscriptions, but I do enjoy content surfaced by the algorithm at least once weekly, sometimes more often. My concern in taking my viewing off-platform is twofold: 1) going to YouTube will prompt me with all the stuff I've already watched of…

I've been using Unhook[0] for years that it's almost a jumpscare for me to see a recommended video or the Youtube homepage. Your social circles and natural serendipity should be plenty for finding new creators. And in general, avoiding algorithmic feeds will help with ADHD and mindless scrolling.

[0] https://unhook.app/

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

#48
post #28

A question for the author or anyone else who has experience in similar solutions. Is there any good solution for discovering new content? Much of the time, I want to stick to my subscriptions, but I do enjoy content surfaced by the algorithm at least once weekly, sometimes more often. My concern in taking my viewing off-platform is twofold: 1) going to YouTube will prompt me with all the stuff I've already watched of…

If you were to have something local build you an algorithm, what signal would you want it to consume and how far from the median would you want it to deviate? Would you want it to use signal from online socials?

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

#49
post #33
post #28

A question for the author or anyone else who has experience in similar solutions. Is there any good solution for discovering new content? Much of the time, I want to stick to my subscriptions, but I do enjoy content surfaced by the algorithm at least once weekly, sometimes more often. My concern in taking my viewing off-platform is twofold: 1) going to YouTube will prompt me with all the stuff I've already watched of…

I looked into this as well since I find the YouTube algorithm terrible, but couldn’t find any API for exploration. Which makes sense they want to control what you watch and hence monetize. In a perfect world you could just pick an open source recommendation algorithm from a marketplace and YouTube would just be a wrapper around s3 buckets and some index.

An even more perfect world would not have S3 buckets.

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

#50
post #33

Earlier quoted context omitted.

I looked into this as well since I find the YouTube algorithm terrible, but couldn’t find any API for exploration. Which makes sense they want to control what you watch and hence monetize. In a perfect world you could just pick an open source recommendation algorithm from a marketplace and YouTube would just be a wrapper around s3 buckets and some index.

An even more perfect world would not have S3 buckets.

You have to store bits somewhere, and an S3 compatible target optimizes for flexibility and optionality. It can be local (Minio), it can be remote, the client does not care where it is. Even the Internet Archive's API is S3-ish.
Post reply on HN