Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

digitalmedia.fi

Post reply on HN
Print view View original

61–70 of 88 posts

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#61

Post by Aachen »

maxerickson wrote: post #29
lucb1e wrote: post #28

Earlier quoted context omitted.

Can you make the math work out in a way that my funds are diverted to another artist? Because what you're saying sounds logical but when I try and compute it, as in the final paragraph of my comment above, it doesn't work out that way. But I could be wrong! Please try it and let me/us know what you find!

Just set aside your "everyone plays an equal number of tracks" condition, because that isn't what happens.

Isn't that fair though? If artist A gets played a million times and I don't listen to them but I play artist B twice that month, shouldn't artist A get most of that money? If artist B was so great I'd play them more often and they'd get more money.

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#62

Post by anonymoushn »

Aachen wrote: post #61
maxerickson wrote: post #29

Earlier quoted context omitted.

Just set aside your "everyone plays an equal number of tracks" condition, because that isn't what happens.

Isn't that fair though? If artist A gets played a million times and I don't listen to them but I play artist B twice that month, shouldn't artist A get most of that money? If artist B was so great I'd play them more often and they'd get more money.

No, it's not fair that my subscription money is mostly allocated by other users' listening patterns, not by mine.

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#63

Post by purple-again »

dsco wrote: post #30
INGELRII wrote: post #25

Alfred Marshall, Principles of Economics (London: Macmillan and Co. 8th ed. 1920). https://en.wikipedia.org/wiki/Principles_of_Economics_(Marsh... ---8 The relative fall in the incomes to be earned by moderate ability, however carefully trained, is accentuated by the rise in those that are obtained by many men of extraordinary ability. There never was a time at which moderately good oil paintings sold more cheaply th…

I'd counter that by saying that the sharpest elbows always outperform the sharpest minds financially. Income and genius have rarely had a strong correlation. Instead I'd look at a mixture between greed, conscientiousness and that feeling of never being content as drivers of income. Also of course the field in which you're in as one could argue that each field has its own "total addressable market".

Interesting assertion you have made because every peer review source I have ever seen on the issue finds a clear correlation between wealth and intelligence.

Could you provide any sources for your claim that income and genius rarely have a strong correlation?

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#64

Post by YetAnotherNick »

maxerickson wrote: post #29
lucb1e wrote: post #28

Earlier quoted context omitted.

Can you make the math work out in a way that my funds are diverted to another artist? Because what you're saying sounds logical but when I try and compute it, as in the final paragraph of my comment above, it doesn't work out that way. But I could be wrong! Please try it and let me/us know what you find!

Just set aside your "everyone plays an equal number of tracks" condition, because that isn't what happens.

I think making use of this condition will undercount the payments of niche artists. I think in general, a listener of niche artist will listen to more song, thus if charges of user are grouped then the niche songs will likely be paid less.(No data, just based on what is more intutive to me)

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#65

Post by lucb1e »

anonymoushn wrote: post #62
Aachen wrote: post #61

Earlier quoted context omitted.

Isn't that fair though? If artist A gets played a million times and I don't listen to them but I play artist B twice that month, shouldn't artist A get most of that money? If artist B was so great I'd play them more often and they'd get more money.

No, it's not fair that my subscription money is mostly allocated by other users' listening patterns, not by mine.

It also works the other way, though. If someone else listens to 1 song of artist A once a month and you play artist B more than once a month, a larger chunk of the total will be going to artist B. And in aggregate things will work out, since an average listener will listen to a random song an average number of times. Broken down to an individual level you can claim that a smaller artist gets less but if that artist's listeners actually play more than the average number of songs an average subscriber plays, then they get a larger percentage instead. That seems like random variation to me.

I'm not sure if it's unfair. If you think artist B is more deserving of your money than artist A despite B being listened to fewer times, then yeah this would be unfair. I'm just not sure why you would say that they deserve your money more than an artist that people listened to more often.

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#67

Post by dtech »

dmitriid wrote: post #59
judofyr wrote: post #19

It's quite fascinating to see the comments on this article as it's pretty obvious who clicked the link and who only looked at the title here at Hacker News. The article itself is called "Pro Rata and User Centric Distribution Models: A Comparative Study" and it's a study showing the difference between "pro rata" (what's happening today) and "user centric" (a user gives their money to the artists _they_ listen to). Th…

I'd also wish all these "studies" brought in the Big Four record labels into the equations [1] as they are the ones primarily responsible for this. [1] See "Consolidation", https://en.wikipedia.org/wiki/Music_industry?wprov=sfti1

You can link to a header in a wikipedia article by using #Header name

https://en.wikipedia.org/wiki/Music_industry#Consolidation

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#68

Post by lucb1e »

kgermino wrote: post #48
lucb1e wrote: post #28

Earlier quoted context omitted.

Can you make the math work out in a way that my funds are diverted to another artist? Because what you're saying sounds logical but when I try and compute it, as in the final paragraph of my comment above, it doesn't work out that way. But I could be wrong! Please try it and let me/us know what you find!

As someone else said, it’s tied to how many songs you listen to. Hypothetical: I mostly listen to Spotify during workouts so I only have about 200 streams a month. My coworker has a top 40 playlist running all day so they have over 2000 streams a month. If we each pay $11 their artists will get $20 while mine will only get $2. I.e. even though I never listened to a top 40 song, the artists I listened to will only get…

So should we play based on manual clicks on individual songs or artists, rather than plays from playlists? Or perhaps exclude only popular playlists, or public playlists in general?

I see what you're saying in this scenario, but in general more listens resulting in more money doesn't strike me as unfair.

I'm also not opposed to doing it the way others are proposing (dividing individual subscribers' money among their played artists instead of aggregating it and doing it by number of plays). I guess that would be one way of solving this issue, even if it's not the underlying problem. In your scenario the underlying problem seems to be that someone plays generic garbage as background noise all day and thereby generates plays that a niche artist whose listeners really love their music wouldn't generate. That's a separate problem, though perhaps changing the revenue distribution model would indeed be the best way to solve this issue.

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#69

Post by anonymoushn »

lucb1e wrote: post #65
anonymoushn wrote: post #62

Earlier quoted context omitted.

No, it's not fair that my subscription money is mostly allocated by other users' listening patterns, not by mine.

It also works the other way, though. If someone else listens to 1 song of artist A once a month and you play artist B more than once a month, a larger chunk of the total will be going to artist B. And in aggregate things will work out, since an average listener will listen to a random song an average number of times. Broken down to an individual level you can claim that a smaller artist gets less but if that artist's…

When I buy a Mechina album on Bandcamp, my money does not go to Nickelback instead. Can you explain how this system currently used by Bandcamp is unfair?

If I wanted to get the same behavior from Spotify, I would need to leave devices streaming at all hours, because I am competing for amount of streams against restaurants and gyms that buy a single Spotify subscription and leave it on constantly.

Re: Pro Rata and User Centric Distribution Models: A Comparative Study (2017)

#70

Post by satyrnein »

Most of the discussion here is missing the point. The actual subsidy is not from unpopular artists to popular artists, it's from (artists listened to by) casual listeners to (artists listened to) by heavy listeners.

Under the pro rata model, any heavy listeners will divert "other people's money" toward the songs they listen to. This could be a superfan obsessed with some indie band, or it could be a fan of popular artists. Moving to a user centric model will certainly penalize any artists that overindex on heavy listeners. For example, children's music that gets put on repeat will probably take a hit.

Post reply on HN