To some extent we do when we do API rate limiting, but I agrree. In general, this is pretty silly. I feel like software engineering is the one field where open sourcing your work is expected to some extent
We have in the past, and we are again, which is just a little ridiculous, and I think everyone can see that.
Back in the day, IBM trialled it with some version of their mainframe and a particular database program, or tried to.
From what I recall, IBM said it would save us in the long run, and claimed they were able to because we hadn't bought the mainframes, they were under a contract. It didn't fly. (Business hint - don't try and fudge numbers when dealing with an insurance company.)
Today? AWS Lambda, and other "function" cloud services, charging for execution time, which isn't quite the same, but I have to feel we aren't far away from someone building their product around Lambda... And then directly passing costs to customers with the same pricing model.
For sales, royalties make sense for songs if we believe that more valuable songs (as measured by how the public values them) should earn their creators more than less valuable songs earn their creators. That's because it is almost impossible to tell ahead of time which songs are more valuable and which are less valuable. You can only tell that based on what songs the public actually buys. Furthermore, this changes over time...a song can start off not valuable, and then years or even decades later become very popular.
What you really want for setting a fair price to the creator is the integral of value to the public over the copyright lifetime of the song. Royalties are a way to approximate that integral.
How about streaming? Similar argument to that for sales, just using play rate to approximate current value to the public instead of sales. This is arguably a better approximation.
For programmers, most of us get paid a decent wage up front to write programs as employees. The employer has estimated how valuable having the programs are to them, and made an offer based on that, and we accepted. We don't ask for a royalty instead, because we would rather have a big check upfront instead of tiny checks dribbling in over the life of the program (which would not be the long copyright life, but rather how long the employer uses it).
For programmers who are selling their programs to the public instead of writing them as employees for a specific employer, their cut of the sale is effectively a royalty.
Finally, what about per-execution royalties for programmers? There are a few differences between programs and songs that make it generally less sensible.
First, number of executions is not as good an approximation to value as it is with songs. That's because programs are utilitarian, whereas songs are art. We run programs to accomplish tasks, and the value of the program to us depends greatly on the value of the particular task we are accomplishing with that run.
Second, most programs aren't streamed for each execution. You can't get execution information back to the programmer unless such tracking is specifically built in. It's very hard to do that in a way that is not visible, and potentially disruptive and intrusive, to the user. If you are in a competitive market, that can put you at a disadvantage compared to competing programmers who are not doing per-execution tracking.
Third, program value is not as uncertain as song value. Programs are tools, obtained to accomplish specific tasks. Users often have a good idea of how valuable accomplishing those tasks are, and how much the program will aid that, so can often can figure out up front what the lifetime value of the program is to them. Programmers who know their market can also figure out the value to their customers of the program. Unlike with songs, this means that we can set a fair to both parties up front fixed price for a program. We don't have to resort to a drawn out approximation like they have to do with songs.
> Jerkins’ stake in the song generated $146,000 in performance royalties, while streaming revenue from the same period garnered $278 for 38 million Pandora plays and $218 for 34 million YouTube streams.
In those two scenarios, how much does the band/performer get?
This carried over from radio. It was decided that recording artists benefited sufficiently from the publicity of being played on the radio leading to record purchases and concert ticket sales. Song writers managed to get a compulsory royalty instead. Their compensation is written in law, they can not negotiate for more, but at least they get paid.
I thought royalties were split between writer, composer and performer. This leads to the saying, “change a word, get a third” for performers who try to get co-writing credits on the songs originally written by others.
I'm not going to correct all the replies to this post, I'm just going to suggest that you should read TFA. This is about the songwriter's cut, and if you didn't write the song you performed, this changes nothing about the 0% of the songwriting royalties to which you were entitled.
Labels negotiate their own deals. Lots more than the songwriter, which makes sense to most people since they pay for musicians, recording studios, advertising, and bunch of other stuff for the song plus all the other songs that didn’t turn out to be the popular song.
wow, that feels very unfair. most artists feel like the labels have shafted the artists, among other things making very very unfair contracts to the artists when digital and streaming came along.
I think that's an opinion that only looks fair to the labels. The labels also do things to suppress competition, and their very opaque contracts only benefit themselves. Clearly the millions of streams should result in more money than less than $300 for 38 million streams? And I'm not an artist, just another programmer. :-)
This perfectly captures the problem with Spotify’s business. Netflix has the same issue on the video side and is working around it by producing original content.
However, music is different than video. With music, Spotify needs a full catalog for a great streaming service. With video, Netflix just needs sufficient compelling content to keep the subscriber from churning.
> However, music is different than video and Spotify does not have the same choice.
Not only have they been creating more and more original audio content, but they're about to try video content again (their first effort failed) as well. In their mind, music was just their beachhead offering.
If not kidding: the more skewed the negotiating power, the worse the outcome from an equity perspective. So individual bargaining would favor a fewd make most songwriters worse of. The buyers in the oligopsony (wiki that) will benefit the most.
They are a panel of three judges which sets the statutory compensation rate for music played on streaming services. Their goal is to set it at a market rate of what a willing seller and a willing buyer would agree to.
It is what keeps streaming stations from having to have lawyers negotiate hundreds of deals for the content they play and then still get sued when they screw up and play an unlicensed song.
Rather than all-you-can-stream, charge a low rate per song play to users. Assume that existing services charges $15/mo, and heavy users might play 16 hours a day, and act song length is 3 minutes, that comes out to about $0.0015 per play to be the same value proposition. But guessing average is much lower so we could up the rate 2x or 4x and customers come out the same. Everyone gets the same cut of revenue per song play, so popular artists get the same cut of streams as indie artists.
Per song is a terrible way to go, because then each 'skip' costs a whole song. Let's pay per time interval (also, that helps fairly compensate epic length songmakers)
The distribution might not be what you expect because heavy users are people who use a stream as background music and let it play it all day. It's not like reading books or watching movies where heavy users are the most obsessed users.
It might make some sense to charge for the number of different songs played each month, though this would tend to encourage channels that repeat songs more often (to save money), and perhaps for artists to record more variations on basically the same song.
Incentives are tricky. When making rules like this, you have to think about how it might be gamed.
I wonder what this will do to negotiated rates, will the Spotify's of the world succeed in arguing that they're paying ~20% more for the music, so the base negotiated rate should go down to compensate for this, or will labels manage to hold the current prices?
I.e. whose pocket will this actually come out of in the long run.
33 comments
[ 5.6 ms ] story [ 82.1 ms ] threadBack in the day, IBM trialled it with some version of their mainframe and a particular database program, or tried to.
From what I recall, IBM said it would save us in the long run, and claimed they were able to because we hadn't bought the mainframes, they were under a contract. It didn't fly. (Business hint - don't try and fudge numbers when dealing with an insurance company.)
Today? AWS Lambda, and other "function" cloud services, charging for execution time, which isn't quite the same, but I have to feel we aren't far away from someone building their product around Lambda... And then directly passing costs to customers with the same pricing model.
For sales, royalties make sense for songs if we believe that more valuable songs (as measured by how the public values them) should earn their creators more than less valuable songs earn their creators. That's because it is almost impossible to tell ahead of time which songs are more valuable and which are less valuable. You can only tell that based on what songs the public actually buys. Furthermore, this changes over time...a song can start off not valuable, and then years or even decades later become very popular.
What you really want for setting a fair price to the creator is the integral of value to the public over the copyright lifetime of the song. Royalties are a way to approximate that integral.
How about streaming? Similar argument to that for sales, just using play rate to approximate current value to the public instead of sales. This is arguably a better approximation.
For programmers, most of us get paid a decent wage up front to write programs as employees. The employer has estimated how valuable having the programs are to them, and made an offer based on that, and we accepted. We don't ask for a royalty instead, because we would rather have a big check upfront instead of tiny checks dribbling in over the life of the program (which would not be the long copyright life, but rather how long the employer uses it).
For programmers who are selling their programs to the public instead of writing them as employees for a specific employer, their cut of the sale is effectively a royalty.
Finally, what about per-execution royalties for programmers? There are a few differences between programs and songs that make it generally less sensible.
First, number of executions is not as good an approximation to value as it is with songs. That's because programs are utilitarian, whereas songs are art. We run programs to accomplish tasks, and the value of the program to us depends greatly on the value of the particular task we are accomplishing with that run.
Second, most programs aren't streamed for each execution. You can't get execution information back to the programmer unless such tracking is specifically built in. It's very hard to do that in a way that is not visible, and potentially disruptive and intrusive, to the user. If you are in a competitive market, that can put you at a disadvantage compared to competing programmers who are not doing per-execution tracking.
Third, program value is not as uncertain as song value. Programs are tools, obtained to accomplish specific tasks. Users often have a good idea of how valuable accomplishing those tasks are, and how much the program will aid that, so can often can figure out up front what the lifetime value of the program is to them. Programmers who know their market can also figure out the value to their customers of the program. Unlike with songs, this means that we can set a fair to both parties up front fixed price for a program. We don't have to resort to a drawn out approximation like they have to do with songs.
In those two scenarios, how much does the band/performer get?
I think that's an opinion that only looks fair to the labels. The labels also do things to suppress competition, and their very opaque contracts only benefit themselves. Clearly the millions of streams should result in more money than less than $300 for 38 million streams? And I'm not an artist, just another programmer. :-)
However, music is different than video. With music, Spotify needs a full catalog for a great streaming service. With video, Netflix just needs sufficient compelling content to keep the subscriber from churning.
Not only have they been creating more and more original audio content, but they're about to try video content again (their first effort failed) as well. In their mind, music was just their beachhead offering.
http://variety.com/2017/digital/news/spotify-lures-courtney-...
It is what keeps streaming stations from having to have lawyers negotiate hundreds of deals for the content they play and then still get sued when they screw up and play an unlicensed song.
Rather than all-you-can-stream, charge a low rate per song play to users. Assume that existing services charges $15/mo, and heavy users might play 16 hours a day, and act song length is 3 minutes, that comes out to about $0.0015 per play to be the same value proposition. But guessing average is much lower so we could up the rate 2x or 4x and customers come out the same. Everyone gets the same cut of revenue per song play, so popular artists get the same cut of streams as indie artists.
Am I missing something?
It might make some sense to charge for the number of different songs played each month, though this would tend to encourage channels that repeat songs more often (to save money), and perhaps for artists to record more variations on basically the same song.
Incentives are tricky. When making rules like this, you have to think about how it might be gamed.
I.e. whose pocket will this actually come out of in the long run.