Varied Listening

Steve Smith, on the Log Journal, recently posted a nice meditation on a day in the life of a record addict.

And at some point it hits you: The amount of excellent music out there that you just don’t know about yet is quite likely infinite.

There’s so much music, period, being made and recorded nowadays that this isn’t at all unlikely. Or if not infinite, it’s effectively so: more than any human can expect to hear in their life. Smith also commented in an earlier post that when Simon Cummings, author of the excellent blog 5:4, posted his “Best of 2016” list, it “included any number of fascinating recordings I hadn’t even heard of.”

The other day, I posted a quote from Brian Eno about the ephemerality of music. Here it is again:

I really think that for us, who all grew up listening primarily to recorded music, we tend to forget that until about 120 years ago ephemeral experience was the only one people had. I remember reading about a huge fan of Beethoven who lived to the age of 86 [in the era before recordings], and the great triumph of his life was that he’d managed to hear the Fifth Symphony six times. That’s pretty amazing. They would have been spread over many years, so there would have been no way of reliably comparing those performances.

All of our musical experience is based on the the possibility of repetition, and of portability, so you can move music around to where you want to be, and scrutiny, because repetition allows scrutiny. You can go into something and hear it again and again. That’s really produced quite a different attitude to what is allowable in music. I always say that modern jazz wouldn’t have existed without recording, because to make improvisations sound sensible, you need to hear them again and again, so that all those little details that sound a bit random at first start to fit. You anticipate them and they seem right after a while.

The promise and the curse of being a music fan in the early 21st Century, I think, is a new kind of ephemerality. There’s simply so much great music out there, and it’s so readily available thanks to the Internet that, regardless of how good it is, it’s going to be harder and harder to listen to anything more than a couple of times. Every album you listen to a second time is a new album you can’t hear.

I’m the sort of music fan who will listen to the same album twenty times on repeat, when I like it. That means that my natural tendency is to miss an awful lot of great music. Over the past half-year or so, particularly since I’ve had a kid on the way, I’ve been trying to vary my listening more, to avoid the trap of only hearing (or playing for her) stuff I’m familiar with.

I try to be systematic, because imposing systems on myself is the best way I’ve found to avoid being lazy. I thought about the types of music that I tend to listen to most often (six months ago, mainly orchestral and piano music written by white men in the 18th and 20th centuries, and rock music from the 20th), and came up with six categories of work that defy those types:

  • Something written since the year 2000
  • Something I haven’t heard before
  • Something more than ten minutes long
  • Something by a composer I usually don’t like
  • Something not written for solo piano or for orchestra
  • Something written by a woman and/or a member of a minority ethnicity

I try to tick each of those boxes at least twice a day. And I try to take it easy on myself too: one work can tick multiple boxes (which is useful on heavy-workload days, when bedtime comes and I’ve barely listened to anything). There’s a handy iOS and Apple Watch app I use to encourage me, called Streaks. It tracks the number of consecutive days you’ve succeeded in doing something. (I was somewhere in the 40s for all of mine, before a couple of rough days last week broke everything.)

The other side of the coin that’s made this possible is Bandcamp. I’ve made my opinion of streaming in the context of classical and other niche-interest music clear in the past (more than once), but I see Bandcamp as fundamentally different: it gives far more leeway to musicians to make and sell what they want, to make money from their sales, and to control what can get streamed. The music on the service is also often far more interesting.

I’ve found some great music through this system: Caroline Polachek’s (free!) electronic album Drawing the Target Around the Arrow, Alarm Will Sound’s Modernists, Gabriel Kahane’s Craigslistlieder. There’s been plenty I don’t like too, but that’s not the point of this effort. The point is to find my boundaries, and push them.

Zeitung Citing

I was delighted to learn last week that my article in VAN about streaming and piracy, in its German translation, was cited by the major German newspaper, Süddeutsche Zeitung. I had a friend translate the relevant paragraph for me last night, and unfortunately the author seems to have spectacularly missed the point:

[The fact that most people use Spotify for free] affects classical musicians even worse than their colleagues in pop, as the specialist online magazine for classical music, VAN, has pointed out in the past few days: not only are classical “songs” more seldom listened to than non-classical; they are also often considerably longer. Because of this, VAN recommends that classical musicians and composers should publish their recordings and compositions for free online. Many are doing this, for example on the platform SoundCloud, which is delighted to receive the work of composers of contemporary and famous or popular “e-music”.

Just to be clear, in the original article (and in the translation), I said the opposite:

Professional musicians who know that their music has limited appeal should think very carefully about whether their music belongs on a streaming service at all. Small record labels should do the same. Few people are in music, least of all classical music, for wealth or power, but giving music away for next to nothing is a surefire way to never make a living from it. Instead, musicians can continue doing what they’ve been doing in one form or another for centuries: selling their music.

Publishing their music online for free is what musicians are already doing on streaming services, and it’s not working.

What I did say, and perhaps what the author of the article mistook as being the whole of the argument, is that putting music online for free in a limited or inconvenient way, such as a couple of songs from an album, or streaming only through video, can be a promotional tool to generate more sales. Barriers are powerful. Many people will pay out a little money to get something they want a little more conveniently. Video game makers know this, and and use it in a disappointingly gross way; but it doesn’t have to be as manipulative as that. Netflix and Audible—and hell, even Spotify—use free trials with limits (in the former two cases, limited time; in the latter, limited convenience) to encourage users to subscribe to their product. The point of giving something away when you’re trying to make a living on it—especially something as ephemeral as a digital copy of a piece of music—is to make people interested enough to pay you.

I don’t think anyone trying to make a living making any sort of specialist product should give it away.

Addendum: Three Analogies

Speaking of VAN, a few weeks ago I wrote an addendum to my previous piece on streaming and piracy, covering some things I had to trim at the time for the sake of space and pace.

Then I got married, and forgot to post it. For the sake of completion, it’s below:


I’ve written before1 that it’s in the interests of the big record labels to have their entire catalogues, or as much of them as possible, on the major streaming services. By sharing their vast libraries, the labels are more or less guaranteed to make their money back. All they need is a couple of hit artists, and they can afford to lose the bets they’ve made on everybody else (or at least break even). In that way, the big labels are like investment bankers: invest in as wide a variety of stocks as possible, and enough of them will provide a return.

It’d be nice if the labels instead saw their musicians as people, but that’s not likely to happen in a hurry.

This is why small labels are unlikely to receive the same benefits from streaming as big ones: not because they’ll never have a hit, but because the big labels they’re in competition with are almost guaranteed to have a constant stream of them.


Netflix is probably the business that seems closest to the streaming music model in many people’s eyes. In some ways it seems apt. Both services provide access to a large catalogue in exchange for a fee, and pay the majority of their revenue to content providers. But they’re different in a couple of important ways.2

Firstly, as I’ve covered before, Spotify (to take for example the most prominent streaming music service) pays a per-play fee for each item on its catalogue, so a song that gets ten times more plays than another makes ten times the revenue. Netflix, like traditional TV channels, pays for its content upfront. While the result is similar, with bigger earners getting bigger payouts, the fact that Netflix’s payments are made upfront means that all of the content creators get paid, and get to decide whether the payment is worthwhile. If you sell your movie to Netflix, and nobody watches it, Netflix lost the bet. They may not buy your movie again, but at least some of the effort that went into making it is recompensed.

On Spotify, if nobody listens to your music, you get nothing. And if one person listens to it instead of buying it, you’re out the sale of a CD.

Secondly, and as a consequence of this, Netflix’s catalogue relative to the number of films and TV shows released is quite small. Spotify is aiming to have every musician on Spotify, whereas Netflix can only afford a limited (admittedly high) number of shows and movies. This means that among shows and movies featured on Netflix, there is less competition for viewers. On Netflix, you’re up against everything Netflix can afford; on Spotify, you’re up against everything.

The main reason for the distinction, I think, is that Netflix is in direct competition with its providers. Streaming services work with the support of labels—especially the big labels for the reasons outlined above. But the big TV and movie companies have their own networks, and the more time people spend on Netflix, the less they spend on the Disney channel. How long this can last is an open question, since the more subscribers Netflix gets, the more content it can afford, and the more content it has, the more subscribers it can get. But I don’t think Netflix will ever have everything the way that Spotify has everything. There’s too much content being made.

Whatever the reason, the result is clear: Spotify’s business model is a worse deal for artists than Netflix’s, and makes it harder than Netflix does for artists to make a living.


News publishers today are facing a dilemma: give their work away for free, and support it through advertising, or hide it behind a paywall and risk piracy.

Sound familiar?

Only the biggest news sites in the world are able to demand advertising rates sufficient to cover their costs, and more and more those advertisements are going to Facebook instead. On Facebook, people care about the news they get, but they don’t care about who’s published it; they just click the link. But if publishers don’t publish to Facebook, they miss out on enormous potential readership. So big news sites are having to settle for Facebook’s terms or face a massive drop in readers. Elsewhere online, small sites are springing up that tend to fall into one of two revenue categories: paid memberships and sponsored posts. The sites have this in common: they are focussed on a narrow range of topics and interests, and attract people who share those interests.

Readers are willing to pay for the content, firstly because they have a passionate interest in the topic, and secondly because, in a lot of cases, they’re supporting a writer who they like. And piracy, while still a danger, is less of a risk because the audience tends to be both small and loyal. In the case of sponsorship-supported blogs, a similar system is at play: a site focussed on a specific interest will have an audience with particular tastes, and advertisers, particularly if they’re making niche products, are interested in that.

The big labels may do well to look at the publishing industry, as they have to compete harder and harder with all sorts of audio content. It may happen that big news sites that produce good work, but not enough of it to keep the lights on, are taken over by other big news sites until only two or three remain. Just imagine if that were to happen with music.


  1. Specifically, I wrote:

    So who wins in the Spotify ecosystem? Well, Spotify do well for themselves, obviously. As do the big record labels, those who have a wide enough variety of popular artists that chances are something they’ve released is being streamed right now.

  2. Kirk McElhearn also wrote on the differences between these services while I was away. 

Music for a New World

Me, writing in VAN magazine:

So if streaming is the future of music, and a musician-led streaming service is doomed to failure, then what’s the solution? Much as with technology companies, there are two ways for a musician to make a living on the Internet: musicians with mass-market appeal, who must appear on streaming services, as it’s in their interests to be easily accessible to as many listeners as possible; and musicians with a small, passionate audience who, to make a living, will have to have dedicated listeners who are prepared to pay for their music. Many artists complain that they are underpaid by Spotify, but Spotify already pays 70 percent of its revenue to musicians—well, to their labels at least—so a significant increase is impractical.

Professional musicians who know that their music has limited appeal should think very carefully about whether their music belongs on a streaming service at all. Small record labels should do the same. Few people are in music, least of all classical music, for wealth or power, but giving music away for next to nothing is a surefire way to never make a living from it. Instead, musicians can continue doing what they’ve been doing in one form or another for centuries: selling their music.

Themes may be familiar to regular readers of this blog—digital economics, streaming and piracy, and the need for musicians to know their audience. VAN is a very interesting new publication, and worth reading. I’ve bought a subscription. This week, I shall be rocking out to their JACK Quartet playlist.

“It’s like you bought a CD and the store told you that you had to listen to it 1,000 times, or they will give your money to Nickelback.”

Speaking of streaming, Sharky Laguana has an excoriating indictment of streaming music economics, specifically where your money goes when you listen to a song:

If you subscribe to a subscription music service such as Spotify or Apple Music you probably pay $10 a month. And if you are like most people, you probably do so believing your money goes to the artists you listen to. Unfortunately, you are wrong.

Read the whole thing, and get angry.

George Grella Reviews Streaming Services

George Grella, on The Big City, has an excellent, detailed run-down of the major streaming services and Tidal. His conclusion: Apple Music wins, with Spotify a close second. And Tidal?

Tidal is so sad and so awful, it may be, in terms of who is behind it and the way its been sold, the worst consumer product ever made. It is the the ne plus ultra of the corporate conglomerate record label, with the inflated costs and lack of taste to prove it. Is this Jay-Z’s design, or has he handed it off to managerial types? I can’t decide which is worse.

The piece is mostly good (though I think Grella’s delineation between “music consumers” and “music lovers” is unfair, condescending, and needlessly dichotomous, but that’s for another day). This caught my eye, though:

At least for the musician’s sake, although it is a pittance, everyone should pay for the music they stream.

You could just buy an album.

Kirk McElhearn, to people selling music

Kirk McElhearn continues to be the best commentator on iTunes and Apple Music. From his blog:

Many record labels and artists who sell music on the iTunes Store direct their fans and customers to Apple’s store through links on their websites. However, since Apple Music went live, these links no longer work correctly.

Instead of sending someone to the iTunes Store, where they can buy an album, these links redirect to the Apple Music section of iTunes, or of the Music app on iOS (if the user has Apple Music turned on).

I don’t see what Apple has to gain from this. They’re as much better off as the artist if they sell an album.


The launch of Apple Music a couple of weeks ago has started another backlash against streaming.

Alex Ross, on the New Yorker website:

[T]he pressure from the margin to the center is strong. Despite “Think Different” maxims redolent of the old Steve Jobs script—“It’s your music. Do what you like with it.”—you’re encouraged to gravitate toward the music that everyone else is listening to. This is what happens all across the corporatized Internet: to quote the old adage of Adorno and Horkheimer, you have the “freedom to choose what is always the same.” The musician, writer, and publisher Damon Krukowski, a longtime critic of the streaming business, calls it the return of the monoculture. “What Apple is doing to music retail,” Krukowski said on Twitter, “is exactly what I saw chains do to books in the nineties: kill indie competition, then eliminate the product.”

Criticism of the “monoculture” has never been less valid. The Internet is an incredibly large place, and within its 3.14 billion users, there’s room for an infinite variety of cultural pockets. While there may be a gravitational pull towards the popular, that pull isn’t strong—certainly not strong enough to change people’s existing tastes. If people can’t find what they want on streaming services, they’ll just go elsewhere. Because the Internet is so huge and so interconnected, it’s never been easier to find people who share your passions, no matter how obscure.

These cultural pockets will continue to exist alongside the titans. While it’s possible for companies like Apple, Amazon, and Google to become almost infinitely large, they grow at the expense of middle-sized businesses, not small ones. Because no company will ever be big enough to cater to everything everybody wants, there’s an infinite number of niches to be filled, and the best way to fill these niches is to be extremely small and focussed. Business analyst Ben Thompson has made the analogy to the rainforest: enormous trees taking most of the resources at the top, but incredibly fertile land at the bottom.

Streaming services are best suited to popular tastes, both from the listeners’ and the artists’ perspective. But it’s true that a lot of smaller artists and labels—the types who fit these cultural pockets—are having a rough time on streaming. Their rate of pay is pitiful, and it’s made worse by the loss of album sales.

To address the problems of streaming, though, we first need to think about who’s encouraging artists to be on the services. Through iTunes, Apple is the largest music seller in the world. If they wanted, they could use their clout to push indie artists into a catch–22: join Apple Music or leave the iTunes store—but they don’t. Spotify likewise requires no exclusives from artists. Tidal wants exclusives, but that business is a total disaster anyway. Only Google’s terms of service are onerous and repulsive.

The reason Apple don’t force artists to their streaming service is simple: it’s bad for them too. Think of it this way: if you’re an indie musician, you make a lot more money by selling an album on iTunes than by having a thousand streams of your songs. And so do Apple. Their thirty per-cent cut of an album’s sale is worth a lot more than their nearly-thirty per-cent cut of a couple of thousand streams. So why would they encourage musicians to be in their streaming catalogue? The problem with streaming services is not that they’re a bad model for musicians; it’s that they’re a bad model for some musicians, but at the moment nearly all musicians are on them.

Indie musicians’ complaints about streaming revenue are misdirected. It’s not streaming services that are to blame for the poor payouts to musicians. Even if streaming services could triple or quadruple what they charge listeners, the payouts to musicians per stream would still be vanishingly small. If anyone is to blame, it’s record labels—big ones in particular. It’s no good for smaller musicians to have all of their music on streaming services, but it’s of great benefit to those musicians’ labels. By having a large catalogue of music on a streaming service, big labels have a consistent source of income. A record label doesn’t care if one of their artists gets a thousand plays per month or a million, it’s all revenue to them. So they’ll upload their whole catalogue to Spotify, Apple Music, and all the rest, because they can. It doesn’t matter to the labels if any particular artist is a bad fit for streaming. As long as they have a lot of musicians making them a little money each, they’re sitting happy.

Rather than blame streaming services for not paying indie artists enough, musicians need to take matters into their own hands. They can only do this by knowing their audience. If a musician aspires to be the next Taylor Swift or Adele or Drake or whoever, then the goal is to get everyone listening, and that can only be accomplished by being available everywhere. In that case, being available on streaming services, and being pushed by a big record label, is almost certainly the right call. Those services are, after all, where most people are listening to music these days.

But if a musician wants to be a smaller success, a professional rather than a superstar, then they don’t need to be everywhere. Instead, they need to connect directly with existing and potential fans. That means being online, and it means building a relationship with their listeners. It also means selling, not streaming, their music, and convincing fans that it’s worth buying. I’ve argued before that piracy is a better option than streaming for musicians who want to build a passionate, loyal fanbase, and I stand by the argument I made then. Listeners who pirate music know the artist isn’t getting paid, and those who fall in love with it will often buy it in future.1 People like supporting independent creators, regardless of their field, because they can see that their contribution makes a difference. That’s the stuff on which Kickstarter is made.

Musicians who want to achieve this type of professional success can’t market themselves the same way as pop musicians: that way lies ruin. Instead, they need to develop loyal fans who are willing to pay to support them because they’re unique. The Internet, and social networks in particular, allow that kind of connection. Success as a musician separate from the peloton is still hard, but it’s within reach of more people than it has ever been before. And that’s not monoculture.

  1. It’s a long-established fact that people who pirate music more also spend more on music. It’s also interesting to note that when piracy was more prevalent, it received little of the type of backlash from independent musicians that streaming has.  ↩

Steve Albini on Tidal

Nails it:

In Albini’s view, these plans may prove lofty but misdirected. “The for-pay services are deluding themselves by trying to establish a permanent monetization of something that’s in flux,” he said. “The internet provides access to materials and things. Creating these little streaming fiefdoms, where certain streaming services have certain artists and certain streaming services have other artists, is a crippled use of the internet. If the internet has demonstrated anything over the years, it’s that it has a way of breaking limitations placed on its content.”