David Emery Online

Hi there, I’m David. This is my website. I work in music for Apple. You can find out a bit more about me here. On occasion I’ve been known to write a thing or two. Please drop me a line and say hello. Views mine not my employers.

Signup to receive the latest articles from de-online in your inbox:

Music in 2030

16 December 2015

Flying cars, hoverboards, and self-drying jackets — predicting the future is hard.

However, if we’re just to focus on music right now, it’s a fascinating time. Certain things are falling into place, which means that the path is maybe—just maybe—becoming clearer for the minute. At least, that is, in terms of how technology is influencing the way people listen to music.

We are obviously at a point now where legal, on-demand access to almost all music is a reality—whether through streaming services like Spotify or Apple Music, or YouTube (which we might want to count as a streaming service as well, though I don’t think anyone actually uses it in quite the same fashion). In some ways, streaming is still a comparatively niche business, but from where I’m sitting—as someone working in the music industry—at some point in the last 12 months it went from an underground niche to an overground one.

It is now inevitable that streaming music will become the most popular way to listen. It won’t be the only way, of course, but it will certainly account for the majority—in the same way the CD once did.

But what effect will that have? To get an idea, we should first look at the key reason why streaming has taken off: the iPhone.

Pre-iPhone there was a variety of nascent streaming startups all with a similar model to what Spotify launched with: $9.99/month gave you access to the entire music catalogue. Rhapsody was probably the most successful of these, but none managed to get off the ground in a way that was significant.

If you were a customer of one of these early services, it was hardly a mystery why they stumbled. Listening to music on the go was an absolute disaster, involving lots of syncing and terrible, inconsistent DRM. And if you wanted to use one of these services with the most popular MP3 player of the time—the iPod—sorry, you were out of luck.

Then the iPhone came along, and with it the ability to run actual, proper applications on the main device where you listen to music. For the first time, logic and computation sat right alongside music—and the main advancement to come out of this is the rise in streaming services.

So, let’s take these two elements—wide access to all music and logic intertwined with the delivery of music—and extend them out to see where we’ll end up.

First, music will be everywhere. As more and more devices get connected, and music moves toward becoming something stored and available online (rather than content on a disk), decoupling the act of listening from a single device will make music all-encompassing and pervasive.

For example, let’s look at headphones. As battery life improves and the technology gets smaller, they will inevitably become connected to the wider internet (Bluetooth headphones will become Wi-Fi headphones, which will then become headphones with cellular connectivity). However, you won’t want to control your listening activity with your headphones. You’ll want to use your phone, watch, or larger screened device. (Note: I am not bullish on the prospect of mainstream adoption of interfaces that rely on voice recognition, although I may be wrong on this.)

In this scenario, just where will the music play from? Is it the device you press play on, or the thing that streams the music? The answer—other then “it doesn’t matter,” of course—is “all of the above,” as we move to a point where on-demand, personalized music access is truly ubiquitous. Extend this example past headphones to anything you can think of that might have a speaker, and you start to get a fuller picture of the implications that this could have: Your music, everywhere you go.

Cars are an obvious point of discussion, especially as we teeter on the brink of the smart car revolution. Once you can easily listen to everything you’ve ever liked or heard while driving, without having to mess around with Bluetooth, aux cables, and the like, will you still listen to the radio?

Well, yes, but maybe not radio quite like we have it now.

All of this ease of access to your own music collection — whatever the concept of a music collection even means now, let alone in 15 years’ time — doesn’t mean that you’ll always want to choose what you hear at any one time. If you have access to all the music you could ever wish to “own,” recommendation and curation is vital, especially when you’re talking about a mass market that typically isn’t as passionate about music as the music industry would like to imagine. This ties back to the idea that logic is now part of the process of music delivery.

There are some early trends starting to form in the burgeoning streaming sector that may give us some clues here.

There has been an explosion in the use of playlists in the last year or so, and when I say “use of playlists” I don’t mean “adding some tracks to a playlist to listen to,” which still does happen in large quantities, of course. I mean listening to public, curated playlists as a new channel for listening, in the same way that listening to a specific radio station is a channel, or reading a music magazine is one as well. Take Spotify’s New Music Friday playlist, which (at the time of writing) has just over a million followers. A million people now have a new set of 50 tracks automatically appear in their music player every week. If you’re an artist and you get a track added to this playlist, you can expect to get upwards of 60,000 plays in the first 24 hours alone.

Considering this is a relatively new platform, and that the playlists themselves are pretty nascent in terms of popularity, these numbers already compare favorably to traditional media such as radio. Getting yourself added to New Music Friday gets you heard, right now, by more people than a play on Annie Mac’s Radio 1 evening show, which is seen as the prime show for new music in the United Kingdom.

Playlists as they exist right now have the potential to become a primary source of new music discovery, and they are only going to gain in popularity. What they lack, though, is personality. For all the potential surrounding them, in terms of presentation and voice, all they have is a small square graphic and a few short lines of text. The rest looks like an Excel spreadsheet. There’s no space to explain just why this track follows the other, or that this track is the first from a new album by x, or how amazing the bass line on this one is.

This is something Apple got right when it launched Apple Music, in the form of its Beats 1 radio station, but it hasn’t gone nearly far enough. When Zane Lowe kicked off the start of the first show on Beats 1 it felt like something that had been missing from services up until now: a literal voice guiding you through what was new and interesting on the service, as well as—more importantly, if we’re talking about a mass audience—making the choices for you if you couldn’t make up your mind about what to stream. At its heart, though, Beats 1 really is just a radio station. Sure, you listen to it over the internet, but it’s still the same linear format that, short of a a playlist you can look at after a show’s broadcast, doesn’t interact with the rest of the service.

So, let’s mash up the burgeoning concepts around playlists, with the notion of “live” and the idea that there needs to be context around music for a significant amount of people to engage with it. When you do, you get something that’s a bit like super playlists, or smart radio, depending on the direction you’re coming at it from and your taste in music consumption. Since they’re “smart,” I’m going to call them “bots.”

If you like more chat, you might listen to a bot featuring more DJs and less music. Still, this isn’t a radio station—you’ll never miss the start of a track, you can skip, and it automatically tailors the program to play music that you’re likely to enjoy. It runs news bulletins relevant to your location and interests. Some of the bot is live broadcast, some not, depending on the show.

On the flip side, if you just want music with no talk, you fire up a bot that behaves a lot more like a playlist would today, except far smarter. Maybe it’s a bot that’s dialed in to the health sensors on your watch and plays different tempo tracks depending on your pulse rate. Or maybe it’s just a straight playlist, but you can tap a button and switch to get the curator to tell you why it picked that track, and then maybe even hear an interview with the band. Many people have tried to recreate the idea of liner notes for the digital age, but I think the most natural expression will be to combine them with an on-demand experience that has been more common to radio. Every track you hear is the potential start of a journey, and how much you control and shape that journey is up to you.

This idea of bots also extends the work that Spotify has been doing with its weekly, personalized “Discover” playlists. They are, I think, the first small step on a long road to something different in terms of experiencing music. The concept relies on several mechanisms, which are either new or have been around for decades, and fuses them together for seamless enjoyment of music that uses technology to guide and shape the experience, without losing the human curation and interaction.

Lastly, what does all this mean for musicians? In terms of formats, playlists are already pushing single tracks to the fore, but this, really, is the same as it ever was. Three to four minutes of music seems to be the ideal amount of time for easy consumption and discovery, and that’s as true with playlists as it is for radio. I would say that there’s less room for filler tracks on a record, as people can so easily skip or listen to something else, but people said the same thing when iTunes came out and it didn’t seem to change the way most artists made albums. You can, though, see on Spotify that there’s a huge gulf in plays between the popular and unpopular tracks on an album.

More importantly though, all of this is great for musicians. The more technology makes music easier to consume—more ubiquitous but at the same time more personalized, while also legal and monetized—the more of the incoming money will flow back to the artist. Going back to the New Music Friday versus Annie Mac example I mentioned earlier, all those plays on Spotify will earn artists at least 10x (probably more like 20x, but there’s a lot of variables involved) more then what they’d get from performance royalties. And this is just the reality right now. Imagine the potential the next 15 years could bring.


Originally published on How We Get To Next, who are running a series of pieces on the future of music which you can check out here. Republished under a CC-SA 2.0 licence.