3 ms·

I propose the following paradigm which I'll call "RickRoll UIs." The developer builds a beautiful interface for searching, curating, and organizing media in a

I propose the following paradigm which I'll call "RickRoll UIs." The developer builds a beautiful interface for searching, curating, and organizing media in a given genre. Everything down to the moment of streaming works as you'd expect, at which point some boring public domain content is looped for the given duration of that stream. Users would get the benefit of having the curated data accessible through the GUI-- e.g., you could search all the Tom & Jerry cartoons, get/sort by metadata, etc. Perhaps even view a screenshot (not sure about Copyright law around that). Fringe benefit is that this creates a reference UI against all the streaming services may be judged, and gives new users some idea of what it was like to search and use "completist" services like what.cd, etc. Not to mention how thinking about the potential existence of such services can change the options we think we have for how to spend our time-- compare writing a blog in 2021 complaining about a commercial streaming service's faults vs. "completing" some private tracker's Tom and Jerry offerings with the missing cartoons from your own collection.

paxys 6y ago
Building the interface itself is the trivial part of this exercise. How are you going to source the database of metadata of every media item produced in the last century?
- recuter 6y ago
  what.cd did it just fine for music and that was a tiny community. There are a several longstanding open projects out there to compile metadata databases that are free to use like: https://musicbrainz.org/statistics https://musicbrainz.org/statistics (33 million tracks) Music is less than 100 million items, Movies fewer than 1 million, TV somewhere in between and books are a few hundred million. I'm sure they all have mostly-complete metadata databases available already.
- Solocomplex 6y ago
  TVDB and others have great APIs for this purpose.
- toyg 6y ago
  This is an ironic question, considering how all the best media databases on the web originally started as volunteer efforts by aficionados. Media-metadata gathering is possibly the best case-study of effective crowdsourcing. So the answer is basically “just empower the community to do it”.
chris11 6y ago
That reminds me of trying to stream a tv show off of the channel's website a few years ago. I kept getting an error whenever I tried to stream the actual episode. The worst part was that the ads worked, so to retry I would have to watch a couple of ads just to find out the episode wouldn't play.

paxys 6y ago

Building the interface itself is the trivial part of this exercise. How are you going to source the database of metadata of every media item produced in the last century?

recuter 6y ago
what.cd did it just fine for music and that was a tiny community. There are a several longstanding open projects out there to compile metadata databases that are free to use like: https://musicbrainz.org/statistics https://musicbrainz.org/statistics (33 million tracks) Music is less than 100 million items, Movies fewer than 1 million, TV somewhere in between and books are a few hundred million. I'm sure they all have mostly-complete metadata databases available already.
Solocomplex 6y ago
TVDB and others have great APIs for this purpose.
toyg 6y ago
This is an ironic question, considering how all the best media databases on the web originally started as volunteer efforts by aficionados. Media-metadata gathering is possibly the best case-study of effective crowdsourcing. So the answer is basically “just empower the community to do it”.

chris11 6y ago

That reminds me of trying to stream a tv show off of the channel's website a few years ago. I kept getting an error whenever I tried to stream the actual episode. The worst part was that the ads worked, so to retry I would have to watch a couple of ads just to find out the episode wouldn't play.