Hi HN, I have a question! As you know you can listen Spotify playlists in a "random" order. I've been wondering about this lately and I don't think there's any chance that these randomizers are implemented using regular rand() functions.
Do you have any experience/idea on how these random functions are implemented in companies that have massive user data? What factors they take into account?
Here is related Spotify Engineering blog post from 2014:<p>>How to shuffle songs?<p><a href="https://engineering.atspotify.com/2014/02/how-to-shuffle-songs/" rel="nofollow">https://engineering.atspotify.com/2014/02/how-to-shuffle-son...</a>
It's knows not be true random "It doesn’t like songs that are totally different bumping up against one another" and users wouldn't like if the same song is played twice in a row regardless how logical that would be with rand().<p><a href="https://lifehacker.com/the-reason-spotify-shuffles-aren-t-really-random-and-h-1849756947" rel="nofollow">https://lifehacker.com/the-reason-spotify-shuffles-aren-t-re...</a><p><a href="https://medium.com/immensity/how-spotifys-shuffle-algorithm-works-19e963e75171" rel="nofollow">https://medium.com/immensity/how-spotifys-shuffle-algorithm-...</a>