I used to work in the music industry professionally, on the ground level doing booking and management. This trend has been happening slowly for nearly a decade but it’s finally here. Rap and Hip Hop figured out a long time before most other genres that rapid small releases was a far better way to keep hype and sales up. Before Spotify was a thing, the shift was happening with YouTube but it wasn’t as predominant. Now it’s basically assumed you’ll be releasing singles every month.
The music isn’t your product, the music is your marketing. The shows, the merch, your influence - that’s your product.
I prefer good lyrics over mumble rap, but music is entirely subjective. You think it's a step backwards, but young people think it's a step forwards.
Mumble rap is essentially just natural selection. Some random rapper started doing it, people liked it, and so it propagated. If people didn't like it, it would have died out quickly. You just aren't one of those people.
> It's samey as hell and stagnant and lacks innovation.
I've said the same thing about metal since the first time I heard it. Metal fans disagree. Who's right? Does it even matter?
I don’t know what mumble rap is, so I’m not commenting on that in particular.
But, something being selected for doesn’t imply that it isn’t a bad thing.
Even if we assume that “how much the typical person likes it” or “how many people like to listen to it” or something along those lines is a good measure of goodness, those are not the only things involved in the selection pressure. Things like cost to produce, discoverability, etc. are all also things that influence the selection pressures. And if there are contributions to the pressures which aren’t entirely aligned with the direction of good, then it seems entirely possible that the selection could make something worse (as in, not as good, not as in, less successful)
The point of my comment wasn't that mumble rap is good, it's that music is subjective. There is no good or bad music. There is only music that is good to some, and bad to others. Mumble rap is good to enough people that it propagated.
An implicit premise here might be that music is good if people think it’s good.
If a genre is taken up through “natural selection,” that means a lot of people like it, and it is therefore good music to those people - regardless of what some critic might think.
A fairly significant amount of young people believe that modern American music is repetitive crap where most songs sound exactly like one another. (Because they are really are)
Yet it is being pushed down their throats.
New Yorkers said the same thing about all southern rappers decades ago and now many of those artists and songs they panned are considered classics. Rap doesn't have to fit some narrowly defined category where all artists are doing the same style during a given period.
You can also like conscious rappers calling for social change and still enjoy a club banger asking women to shake their butts. They aren't diametrically opposed.
>Same beats. Same cadence. Same flow. Voices indistinguishableffrom each other.
This is how people used to talk about Hip Hop in the 90s. People familiar with Trap can easily distinguish between the different rappers and producers - just like people familiar with 90's Hip Hop could distinguish between Nas and Ghostface or DJ Premier and Pete Rock.
And, that's only Trap. There are still popular MC's and producers who operate outside of Trap.
>This is how people used to talk about Hip Hop in the 90s
This isn't true unless you're talking the bling Era of the late 90s.
Also, past criticisms existing doesn't invalidate new ones even if there's similarities in the complaint. So that's a side step.
Tribe, wu, west coast gfunk rappers, mobb deep, fu schnickens, onyx, bone thugs, cypress, busta, we're all wildly different. Even at the height of g-funk, I could have named 50 at least semi popular rappers who deviated from that style.
Now who deviates? Danny brown? He's like 40. RtJ? They're like 40 too. Kendrick tends to always be the modern exception to every hip hop criticism. So I consider him the exception that proves the rule.
>are still popular MC's and producers who operate outside of Trap.
And they're treated the same as Rhymesayers or DefJux rappers, Saul Williams or immortal tech were during the bling and Crunk eras. It's all "white" even if black, nerdy, back packing shit thats not considered part of the "culture". Thats the way most lyrical rappers are treated today. They're talked about like Atmosphere was in 2004.
>This isn't true unless you're talking the bling Era of the late 90s.
You lived in a different 90s than I did then.
>Now who deviates? Danny brown? He's like 40. RtJ? They're like 40 too. Kendrick tends to always be the modern exception to every hip hop criticism. So I consider him the exception that proves the rule.
Who deviates from what?
>And they're treated the same as Rhymesayers or DefJux rappers, Saul Williams or immortal tech were during the bling and Crunk eras. It's all "white" even if black, nerdy, back packing shit thats not considered part of the "culture". Thats the way most lyrical rappers are treated today. They're talked about like Atmosphere was in 2004.
Not my experience. During that time, and a little before it, you had Blackstar, Common, Dead Prez, and a bunch of other underground acts who were respected - to the point Jay Z name dropped Talib Kweli and Common in the early 2000s. Freddie Gibbs and Big Krit both operate primarily outside the scope of Trap, are very talented, and respected. I don't like J Cole, but he's another one who is liked and respected.
>Tribe, wu, west coast gfunk rappers, mobb deep, fu schnickens, onyx, bone thugs, cypress, busta, we're all wildly different.
I would argue that they all sound the same when compared to any southern rappers that came after them. You keep referring to the bling and crunk era like they are sneers but I would argue that if New Yorkers had gotten their way and got to be the gate keepers of hip hop and you were only allowed to rap in a style like Wu Tang and rappers like Outkast, The Hot Boys, and just the dozens of ATL rappers were not allowed on the air ways that the art form would have gone the way of disco and died. Sure its cool when a song has a multilevel meaning and deep metaphors spread through out the song, but its not a necessity to make a good hip hop song.
My hot take is that if you can only appreciate NY State of Mind, I Got a Story To Tell, 1st of The Month, but can't appreciate Back That Azz Up, Crank Dat, and XO TOUR Llif3 then you don't actually appreciate hip-hop. Just a very narrow subset of it.
If anything mumble rap is a devolution. Add auto-tune to this and you are left with a worthless corpus of shitty rap, hip-hop and also pop music.
It is samey, because everybody does the same thing. You can’t really differentiate music by different performers anymore. Each song is the same.
There used to be a cost to producing music; you needed a studio, producer, physical copies of the samples you wanted to use (in case of most rap), etc. Now you can just boot your computer and use one of the many digital studios.
It is a blessing in some ways, but a curse in most I think. Quality is by no means the element by which (popular) music is judged. It is all about presentation, which somehow means the most deplorable people get the most praise.
Note: I am talking about what is mainstream and played on still high quality radio. I don’t listen to rap stations and luckily there are stations that have a good selection of older music and only play some of the new stuff. The new stuff there is already worse than music from the past though.
If you look at the rap and hip-hop scene however, the quality has plummeted even more. It is just one homogeneous soup of crappy, lazy music. In no way like the music of 2Pac, Notorious B.I.G and Outkast but still all the same as the other new crap.
You say they're older as a joke, but it's very much relevant. Old artists dismissing the developments and approaches of their younger colleagues is a tale as old as time.
Concerns about making money are not going to destroy art. Almost all art through history has been made by people trying to make a living, not trying to express for its own sake.
We don't have a thousand years of portraiture of the elite because artists were just absolutely convinced that was the ultimate expression of the medium.
The lack of discoverability and gatekeeping (to some extent) ensures that unpromoted art will not be known. When did the last non-supported musician hit it big? Artist support and development is important as well.
I think the claim is that mumble rap dominates this new ecosystem, that this ecosystem is less diverse, and that this ecosystem does not seem to be changing significantly.
I have no idea about Spotify or mumble rap but that's how they might not contradict.
Even Mozart and Bach made crap music on a bad day.
I think in our time we are very fortunate to have the ability to discover music so easily. Personally I never listen to mainstream music but there is a lot more to discover.
Music generally lacks innovation for the same reason why phones and men's clothing lack innovation. Most of the easy low-effort ways to be appealingly different have been discovered already. Music's creativity has been spurred by a series of technological innovations, the invention of the electric guitar, then the invention of synths, multitrack recording, drum machines, then computers with sampling, loops and DSP. It's been artistically transformed by blues, country, jazz, rock and roll, urban, dance and rap. Each one had its own series of innovations driven by creativity (doo-wop? math rock? new jack swing?) but the genres have settled into their mature phases. Probably some entirely new genre needs to come along and shake things up, but that's hard. Until then we'll continue to get the artistic equivalent of a new coat of paint every few years.
Which is the sad reality. There are definitely talented rappers that work their tails off to produce great lyrics, in some cases their own beats / tracks and are being overshadowed by what I call pop rap (you can throw mumble rap under than umbrella). I can only imagine the same is happening in other genres. The one decent thing about rap was always all the mixtapes on the other hand.
That's just how popular music and pop culture works. Internet and streaming did enable long-tail though. Its way easier to listen to and find those underground rappers compared to 20 years ago.
A friend once explained this to me a decade ago. I would buy albums, but he would torrent everything. In his eyes, music has shifted to the public domain now. He said the best way to support a band now would be to buy t-shirts and go to as many concerts as you can. I still buy all my albums, but finally see his point. I mean now, all artists have their entire collection of music on YouTube. If everyone almost always has an internet connection, is buying music even necessary? Does Taylor Swift make more money off of the albums my wife buys from her or the 8 t-shirts and yearly concert she goes to?
I would say this is definitely true for big-label musicians. For independent or smaller-label musicians, they still have a large upside potential from certain retailers, like Bandcamp [1].
If you really care about them and they're listed on Bandcamp, it's pretty trivial to sign up for email announcements at an artist level, so you don't miss an opportunity to support them financially whenever they release new material.
Many big musicians don't even have bandcamp accounts, which (1) amazes me (2) implies that they make most of their money through other means, e.g. touring, private performances, etc.
[1] From their website: When you buy something on Bandcamp, 80-85% of your money goes to the artist, and we pay out daily. The remainder goes to payment processor fees and Bandcamp’s revenue share, which is 10-15% on digital items, and 10% on physical goods.
Bandcamp is great. If you're with a major, though, you will likely be contractually prohibited from selling through anything but the label's online store (I am).
I think that only applies to very few labels. Bandcamp isn't some kind of exclusively niche underground store. These days all kinds of big, critically acclaimed labels and artists sell through them.
Your friend was absolutely right. I never cared if people pirated my bands music (or later the bands that I was managing) because the real thing I wanted was them showing up at shows, buying merch, posting us on facebook, asking events to bring us on, etc...
At a certain point, Taylor Swift and larger artists _do_ make a good chunk of change from their actual music sales, but it isn't a feasible strategy to rely on that for indie and DIY artists.
> Your friend was absolutely right. I never cared if people pirated my bands music (or later the bands that I was managing) because the real thing I wanted was them showing up at shows, buying merch, posting us on facebook, asking events to bring us on, etc...
This model breaks down for people like me who are neither interested in merch nor in going to concerts. I just like the music.
You always represent value even if you don't pay. You might introduce the music to someone who will pay. It's freemium model applied to music industry.
Can confirm! I've sent tips to musicians who put up music under free/Creative Commons music!
My theory is if you want to see something tomorrow, the best insurance you have it is to make it profitable for the people doing it today, so giving my money to musicians means a better chance there will be more music in the future <3
That's also why I don't give money to homeless (but my country support for homeless is outstanding; when I lived elsewhere I donated to a homeless organisation - which hopefully meant that money didn't go on drugs).
I listen to a lot of music but I’d never go to a live show (can’t think of anything worse) and I’m not going to wear a branded t-shirt. How do they capture revenue from people like me in your model?
Now I think about it I did once buy a video log of the production of an album. But that’s one artist from the thousands I listen to.
To be honest - they don't. You're not the target demographic for them, and that's fine! But for this approach is optimized for the audience that will wear those shirts, go to those shows, buy the posters, follow them on instagram, etc...
I think these people may still provide some value in the form of Retweet, Gossip, Views, Media, Hype, Clicks that helps marketing. In the world of Social Media just having your name known is better than totally unknown. And will funnel back to other places.
>How do they capture revenue from people like me in your model?
It's not like bands stopped selling CD's/BluRays's, collectible box-sets, stickers, pins, access to behind-the-scenes videos and mailing lists. But the most someone like you can do for an average band/performer is promote their art through any channels available. The most valuable resource you have is your attention.
This is crazy but we used to have this scheme where you actually just paid them for access to the recording. Weird, right?
Then we Spotifyd and that was the end of that and people got a new hobby coming up with justifications for why it's good that we replaced a revenue source for musicians with ... maybe giving them money? Or maybe having them run a T-shirt business?
Maybe they can work for a living. Like go on tour and perform live. Selling tickets and such. If they are any good, people will want to go see them live. Look at the Grateful Dead, they allowed people to copy their music. There are massive sites dedicated to archiving all of their music recordings and you are free to download and listen to it. Yet they made a fortune on touring and performing. Not playing the song one time and then trying to sell copies of it.
Saying "maybe they can work for a living" is saying that people don't think that has value, in which case it's weird that anyone uses Spotify at all.
"Playing the song one time" is minimizing the work/time that goes into learning to play instruments, writing a song, coordinating the musicians for recording, and working on the production.
And it's a lot like saying software developers shouldn't be able to write a piece of code one time but get paid for people buying a copy or using a service running it on a server somewhere.
I have been saying this again and again, but not every artists is a rock band. There are countless releases that are best enjoyed in a home-listening environment and attract a mature audience that couldn't care less about t-shirts and merch. Thankfully the album still seems to be the primary product in those circles, but obviously even those artists will be affected by the expectation of having their discography available on Spotify.
> Maybe they can work for a living.
> Not playing the song one time and then trying to sell copies of it.
As in writing a few lines of code and selling copies of it? What a condescending and out of touch take on the issue. Those albums and songs aren't made of thin air - it's actual work that goes into them.
Yeah, it was weird. I don't want to have to keep track of a particular object, whether it's a physical recording or a downloaded file.
Make sure the music is available when I want it. Send a bit of money to artists I want to support. But don't couple them together with a bunch of transactions I've got to manage individually.
Retailed-recording doesn't mean you have to manage the file. That nut's been cracked for at least a decade. Every digital purchase I've made through outlets like Bandcamp or Amazon (and some of Apple, IIRC) is kept track of by the service and has a cloud player I can use. And the artist actually sees revenue from each track that's a fraction of a dollar rather than a fraction of a penny.
But you can actually download/manage the file if you want to. Which is nice, because it's the most practical distinction between owning and a long term lease, given the fact that few things are forever.
> Retailed-recording doesn't mean you have to manage the file. That nut's been cracked for at least a decade. Every digital purchase I've made through outlets like Bandcamp or Amazon (and some of Apple, IIRC) is kept track of by the service and has a cloud player I can use. And the artist actually sees revenue from each track that's a fraction of a dollar rather than a fraction of a penny.
You still have to keep track of what you've bought when and where. You have to answer questions like which release of an album you want (extended edition or not? Explicit or clean?). It's not a huge amount of effort, but it's extra faff compared to just listening to what you want when you want, and sending money to who you want when you want.
I gave up on buying after I moved countries and found I couldn't access my old cloud player and my new one at the same time. Yes, I have mp3 files of everything I bought in my old country, but it's such a fiddle to actually listen to them that I don't bother - particularly when I can find almost all of them in the service that I'm paying monthly for in my new country.
> It's not a huge amount of effort, but it's extra faff compared to just listening to what you want when you want, and sending money to who you want when you want.
While we're talking about extra faff, how do you figure out how to send money to "who you want when you want"? Doesn't seem to be just one service everyone's taking OR sending money through any more than there's just one service anyone might buy recordings through. Or do we just say "If they're not on Kicktreonattrpaymo when it's on my mind I guess they don't want my money"?
Doesn't sound easier than keeping track of a few places you buy music. I've certainly never had any trouble at all figuring out whether I bought a recording on Amazon or Bandcamp.
And I know that my ledger's clear and doesn't depend on when I decide to get around to kicking a donation over.
It's interesting how the tenuously more convenient value proposition of cloud recording buffets features the convenience of pushing the hop through the hoop of actual economic support out into the undetermined and perhaps entirely optional future.
If this is all about convenience, Spotify could probably largely solve the economic problems by bumping up what they charge 5-10x and directing additional listener revenue by specific listener choices.
> Or do we just say "If they're not on Kicktreonattrpaymo when it's on my mind I guess they don't want my money"?
Yes. I don't owe them a business model, they're the professional here, it's their job to be where I want to pay them.
> And I know that my ledger's clear and doesn't depend on when I decide to get around to kicking a donation over.
There's no ledger here. If you want to pay them, pay them, if you don't, don't. If that wasn't an arrangement they were content with, well, revealed preferences.
> If this is all about convenience, Spotify could probably largely solve the economic problems by bumping up what they charge 5-10x and directing additional listener revenue by specific listener choices.
What economic problems? There's no shortage of content being created, so evidently the deal for creators isn't actually that bad.
The simple truth is that success in any creative field is based more on chance and mass appeal than any creative merit. For every JK Rowling there are hundreds of published novelists who barely make a living, and thousands who a publisher wouldn't even take a chance on.
If you're in it for the passion, you'll do it whether you make money or not. Big successes are an anomaly, regardless of which mechanism the industry chooses to split up the loot. (Always in the favor of the publisher/label/whatever.)
If we want to encourage creativity, UBI is a better deal than worrying overmuch about trying to shoehorn creativity into revenue models. Because then at least the extra money, however much, would be extra on top instead of wholly inadequate.
Both Youtube and Spotify allow you to listen for free with ads.
But to humor you. Youtube is for video content primarily. Youtube Music is a premium service that costs as much as Spotify does and has a whole lot less content.
As someone who listens to a lot of music. Spotify offers me a way better UI/UX for consuming music than Youtube Music does.
Buying the albums mostly supports the music industry when you do the accounting and does little to support the actual band directly. The middlemen mostly exist to build a machine to allow people to profit from music... Which is an arguable indirect benefit to the content producers. Yet it does lead to the counterintuitive situation where a merchandise buying pirate is actually directly contributing to a bands coffers more than a law abiding citizen may be.
Correct! I’m not sure why comments here equate music on YouTube with giving it away - while it devalues it by making it free to listen, artists still get royalties from advertising.
This is very true. One of the first artists I noticed doing this was a rapper, Curren$y. All the mixtapes were free on Datpiff.com (major throwback) and the tours / Merch were the $$$ makers.
This also allowed artists to go fully independent and keep all their profits.
Added into the mix now are independent vinyl / cd / tape makers who the artists partner with on Physical music releases. Yes, physicals are back in a major way thanks to platforms like Bandcamp.
Even without physical media, I'd rather pay for a copy of the music I can "own" (I.e. like a physical CD) from Bandcamp as the revenue the artist gets is _significantly_ higher than a dozen streams of the same album on Spotify.
At the end of the day, I want the artist to continue producing music, and prefer to find a channel by which they get the most revenue.
IMHO Bandcamp is one of the best things to happen to the music industry in the last decade. Love it!
yeah, Datpiff, worldstar, HHWW, etc... these sites were all at the forefront of this paradigm shift for backpack hip hop and rap. Hardcore and punk genres had similar approaches with BandCamp, PureVolume, and even MySpace, and if you actually look at punk and rap they have really interesting overlaps in these areas. Hip hop going independent was a major shift from the established late 90's/early 2000's tradition of signing with a label.
The Dave Matthews Band had a very permissive bootleg policy (bootlegs are fine, we'll even help you place microphones for the best quality, you just can't sell the bootlegs) which was important to the early popularity of the band.
Services with recommendation algorithms can help. Mixcloud can be good, because it features longer mixes that often contain similar music, collected by someone who knows their stuff. You can also check the associated label's other artists, smaller labels often organize around a similar mood or style. I check the people my favorites collaborate with.
Korean pop music has been doing this for a while too. This group has pretty rapid releases, 5 singles a year, and 10 EPs and 3 albums since 2015. https://en.wikipedia.org/wiki/Twice_discography
While I enjoy Kpop, I don't feel it's necessarily a great model for artistic expression. It's an industry where fake is the status quo. Kpop is not selling people music per se. It's selling people this idea that you could be an idol or be friends with an idol or be a partner of an idol. It's an emotional hack like a carrot being dragged in front of you.
I personally like a lot of Kpop for the strange music production that comes out of it with bands like Red Velvet, Mamamoo, Stellar, Lim Kim, Dalsooobin, Dal Shabet, etc., but that is really more of a side effect of Kpop's main product.
To be quite clear I'm not endorsing it. It just seems to be a natural evolution of music industry in the digital age. South Korea was, if not is, one of the countries farthest on the digital curve.
In addition, Kpop groups put a huge amount of effort into their physical albums. Sales of physical albums are extremely high in Korea (Japan I think too) compared to other countries (just my impression, but if anyone has sources I'd be interested).
Albums are usually sold as boxes with a photo book inside and collectible cards, they are almost more like magazines. Some albums come in different editions with different idols featured - it's common for a fan to buy multiple copies of the same album in hopes of getting the edition or cards of their favourite idol (bias). The fans see album purchases and 'streaming' (playing their youtube singles on repeat) as ways to support their groups and increase their reputation.
Think of buying albums more like a kickstarter fundraising for fans to support their idols. The photo book or the chance to get a ticket for a fan meeting is arguably a bigger draw than that CD itself.
In fact, what I know, a lot of sales, especially oversee ones, those albums are not even physically sent back to the country where they originated at all. It is just a gesture to help their idols rank better on the music shows.
Rap & Hip Hop are different because historically there's much less of a touring component than there is in Rock, Country, Soul/R&B, etc.
I'm not saying there's no overlap or that nobody can learn from anybody else, but to lump them all together ignores a lot of actual differences between marketing among genres. Heck, I bet TV talent shows are for country music right now what radio has been for hip-hop/R&B since the late 90s, which makes sense because country has (ironically) been appropriating just about all they can find in radio rap for the past many years (way before Lil Nas X).
Since independent venues have been gradually disappearing in the US in the 21st century, and ticketing is locked down by Live Nation, it sure seems that the only revenue stream American musicians can control is their merchandise.
Makes sense that everything else is just "marketing" these days.
I'm sure this has an impact on who decides to make music professionally. Why stick with it, only to sell your own t-shirts? (Note: I can't figure out how to get trend data for from the Bureau of Labor and Statistics, to see if that question has merit.)
The only album that comes to mind is Exile on Main Street, but I'm not particularly into back stories. Were a lot of people doing this in the 70s?
I'm D'Angelo is as good as it gets. Voodoo and Black Messiah are two of my absolute favorite albums, and they're the only two albums he's put out in the last 20 years. Other than it being a long process, I don't know a lot about the Black Messiah recording process, but Voodoo was multiple years of jamming with super-talented musicians then condensing that material into an album. They'd listen to classics like There's a Riot Going On then use that inspiration to guide the jam sessions. I think this approach is pretty similar to what you're referring to. On the other hand, look at how quickly acts like The Beatles, Bob Dylan, and the Stones put out music in the 60s. The Beatles released two albums in '67 then a double-album in '68; Dylan released two albums in '65 then a double-album in '66 - then recorded all the Basement Tapes material in '67 and released another album that same year.
I never worked in the music industry, but hasn't it always been this way?
I have a lot of my grandparents old records, and when I first went through the collection I was surprised at the output of some artists both big and small. The obvious example would be Elvis. Just counting his studio albums, the guy put out over 20 albums over his career 24 year career, and that isn't counting singles, soundtracks, or whatever else he put out. If you factored in all of that, it's easily in the range of 150-200+ releases.
When vinyls were the predominant media form, it was prohibitively expensive for most artists to press a master, cut vinyls, package them, market it, and distribute it. I can do all of that in 10 minutes with Spotify now.
Tapes made this a little bit faster. And then CDs made it a little bit faster from there. And then MP3 players made it quite a bit faster. And then download speeds and prevalence of internet made streaming possible, and it made it a LOT faster. And then here we are - about 10 years after the emergence of Pandora and similar services.
To use your example: Elvis had a TON of funding and was able to maintain that rapid release cycle while touring because of that. Smaller DIY artists don't have funding, they're all working jobs, etc...
Oddly enough, in the decades before the phonograph, there was a sheet music industry with almost every feature of the recording industry. There were superstar composers, such as Scott Joplin, shady deals, composers getting rich and then getting screwed, the whole nine yards. People bought sheet music to play on their parlor pianos.
There was plentiful work for pianists and organists in movie theaters before the talkie. And so forth. Musicians have faced a constant back and forth relationship with technology practically from the git go.
Fair. But doesn't that stifle genres that focus more on experimentation, deep philosophical lyrics etc.? I don't mean to imply that Rap is not that, just that some types of music, (including Rap & Hip-Hop in many cases am sure), require more time and experimentation to 'feel right' than others, they require research, studying history etc.
Not sure these are fit for a single/month type of release. Even on a 12 song album that used to come out every 3-4 years, you only have enough material for a single year now, at that pace.
Quantity is often said to come at the expense of quality and for a good reason.
I really don't think it does. If you spend 3 months straight on a single (which isn't out of the ordinary for my friends who are still in the scene) you're approaching the same level of time input per song, but you're still on a tighter release schedule.
See, but that's the thing. Many artists don't work in this way where they'd have a major project that is this single to get our in 3 months and that's like me having a sprint deliverable and that's that.
They'll write songs when there's inspiration, but maybe nothing for six months after. Then some don't feel right so they get thrown into the bin etc.
An album release is very much "when it's ready". Forcing it on a schedule only works when there's already a formula, which is mostly true only for very mainstream music.
As an example, look at the distance between these 2 albums[1], both great in my opinion, but technically it took 14 years after the last one to get it out, (abet not being worked on continually of course).
Spotify isn't saying that if an artist doesn't put out material every 3 months they'll be de-ranked. That's absolutely not the idea here. What they're saying is that in the current climate, you have to fight for attention, and rapid, small releases achieve that.
A quick check of Dawn of Solace's Spotify page and tour schedule shows me that they're active but probably not making full-time jobs out of this. And that's absolutely fine. They're doing their thing and probably enjoying it.
The article is saying that to be profitable and relevant, you're going to have to put out more frequent, smaller, rapidly-digestible bits of music. This was a model I applied regularly to death metal and hardcore bands that I worked with during my time in the industry and I can tell you right now- _it works_.
The CEO of Spotify was saying that we have a glut of music and customers don't care much about which music they get, and Spotify has won control of the distribution channel, so there's no money for artists but they need to crank out content to stay in the rat race and lower Spotify's costs even more, or somehow form a strong union and market themselves as premium content.
But he put an obnoxious self-serving spin on it because that's what rent-seeking billionaires do.
Yeah frankly I was shocked how obnoxious this argument he's making is. "No it's not that we don't pay artists enough, they just need to be working for us twice as hard to make money these days. ¯\_(ツ)_/¯ " At least have grace to act like you can't pay artists more, instead of just saying "well you'd better work harder then, haha."
That's fair, my issue is that these sorts of schedules only work for somewhat mainstream, full-time musicians.
> Spotify isn't saying that if an artist doesn't put out material every 3 months they'll be de-ranked.
Not explicitly, but if someone releases 5 times a year, they're bound to be on the frontpage much more frequently. This of course makes sense and increases band awareness and engagement, but possibly puts pressure on smaller bands to release quantity over quality if they want to make it.
Saying more frequent releases == more press time is so obvious that I don't really get the point of saying that. This is true for software too, btw.
But as a software developer, I can tell you that while a minor release can generate almost the same amount of press as a major release for me, a minor release is in no way significant. It usually doesn't contain novel ideas, merely bug fixes and security patches, it doesn't push the software "forward" in any way, whereas a major release usually does.
I think DJs have a thing where they release a track and then slowly drip in various guest remixes of the same track.
As for metal, am sure it works to increase engagement and that's fine. But hearing from my musician friends, they tend to like to take their time to get things right and many fans there like big album releases.
Of course everybody enjoys teasers, but if we're talking full tracks here, you'd still want that 10-14 track album where many of the songs are surprises.
It's probably different if one goes to it commercially from the get go, with the express intent of making it a full time thing, rather than sort of failing into the full-time thing as you pick up steam.
Note that this is different from live shows, which I do believe need to be super frequent for these types of bands. But I am not as sure about Spotify digital releases.
It is worth noting that he gives an example of Taylor Swift as someone doing it right. That's Taylor Swift - one of the most mainstream artists right now, with an army of composers, producers etc. behind her. That's exactly my problem with his statements.
Do you have links on articles on how many books are needed to make living and how much of a living it might be expected at median and high ends? I'm interested in the subject and your offhand comment makes me hope you have more insight into it than I do.
If you’re interested in kindle self pub, there’s a pretty vibrant community on the kboards (kboards.com) forum. Tons of info on producing volume as a marketing strategy but people are definitely making a living. It’s not my approach but I’ve certainly considered it.
Also, didn't the hip hop crowd just give away their music for free and make all their money touring? Until signed anyway. And mix tapes were more than a single. So to that end, music definitely was their marketing. To put it another way, they let the music market itself. That's all they needed, and was all they got. I would put influence under marketing though, unless we can settle that music and influence are both both.
You're basically just repeating what I said, but yes. I would say influence is a little different from marketing. Marketing is you trying to reach new fans, influence is you reaching established fans.
I absolutely hate this development. Admittedly within the genres and niches I listen to the album and the music itself is still the product, but I worry that this will change.
Concerning the single format: Going through and listening to my Bandcamp new release notifications is a weekly habit I enjoy. There are a lot of of those notifications and I gladly take the time to check them out, but I won't even listen to single track releases. I'm an album person. If an artist doesn't have enough quality material to compile an actual cohesive selection of his work and rather rushes from short-lived single to single, I'm not interested.
I'm even less interested in t-shirts, merchandise and live shows, though I get the appeal of the latter.
Id go even deeper than hiphop. Jamaican music that was essentially a big influence to rap, was “singles” market from the start and album stuff was more about pleasing the international crowds.
lol, all shit in the world is soon turning to an "agile" model. Soon enough there will be scrums and daily standup for musicians. Time to create a Jira like app for them and cash in on the gold rush.
Quality of the art is of course completely absent from any such discussions.
I think it's a little sad that you can't write some great software that solves a problem for people and get paid for it anymore. Or at least not as easily as in the past.
The changing business model has a lot of negative externalities, too. Like the temptation to monetize user data instead of monetizing the value of the software directly, and the temptation to push the line on privacy.
> I think it's a little sad that you can't write some great software that solves a problem for people and get paid for it anymore. Or at least not as easily as in the past.
I think that you're ignoring the fact that you were actually getting paid for solving their problem, not for great software per-se.
Just because the constraints on different parts of the value chain have changed, altering where the convenient bottlenecks are for extracting payments, doesn't make the past environment any better overall, just different.
But I think that's because open-source software development companies are the exception rather than the norm. No one expects you to open-source your software and then hope someone hires you for your services or buys one of your t-shirts. It's still the norm in our industry to write something and sell copies of it.
Plus, software pays more. Software developers generally don't get squeezed. Musicians, on the other hand, never seem to catch a break.
EDIT: Or, even better, it's the norm to sell a monthly right to use what we wrote.
Also, freemium games have business models very similar to music. Yeah, there are some successful freemium games out there, but in general it's pretty brutal and depressing.
Basically, in software, you can choose to open-source your software as marketing if you want, but proprietary is still a sustainable business model. The claim here (which I am in no position to evaluate) is that for music, you no longer really have any choice: selling music itself is no longer a sustainable business model; you can only give it away in hopes that you can sell something else as a result.
> Software developers generally don't get squeezed
Where I live, they do...go live in one of the countries where software gets outsourced. People are paid < $200 monthly here.
Software developers have skills that are easily transferable than musicians. If software developers stop learning, they will quickly be devalued with legacy maintenance as an exception. We used to value musicians more because of supply and distribution problems. It was an artificially constrained labor market. Now it is not. That's why they aren't getting paid the same. Similar to how writing html and css would get you a good paying job, not now though. Musicians need to change and adapt. They need to pick up more transferable skills.
* infinitely reproducible for near zero marginal cost
* subject to changing tastes and style
* easily obtainable through illegal means
This is especially true for games. Many software companies seem to have solved this dilemma by becoming surveillance or advertising companies that happen to make software. Games have solved it by peddling an increasing quantity of paid, high-margin add-on content and integrating gambling mechanics into their products. I hope the solution musicians come up with isn't as slimy as these.
Edit: I also find that most people treat most music as a commodity much of the time. Why else would random feeds of related music be so popular?
If you stop playing the huge AAA games and start playing smaller ones, you see another solution: lower prices for games that can be finished in an amount of time that fits into the busy life of an adult, and very often a grant from a national art fund.
Not for American games, of course, we basically destroyed the National Endowment for the Arts, thanks a lot guys.
And the King, having vanquished his foe at great cost, surveyed the empty field and muttered: "It matters not, for in the end the claw chooses who will go and who will stay".
I think you will find many here who would argue that code can be art, too.
The thing that people don't realize is that it is in fact quite comparable: the binary that one works with to hear a song is not intrinsically worth any more than the same bytes that produce a software application. I'm a musician who now primarily makes music with code. [0] I've decided therefore to always make my music available for free, without exception. We all agree software can and often should be free - music, to me, is exactly the same way.
Musician and software developer here, and I'm calling BS on that. Here's a transcript of a talk I gave last year on "code as art" (apologies if this sounds like self-promo). You can skip the first half and scroll down to where it says "Which leads me back to code as an artform". There are a lot of ways to demonstrate that code IS art, and this was just a quick 5-minute overview of the subject. Cheers! https://www.johnluxford.com/blog/code-as-art
As an acquaintance of mine put it the other day on this exact topic (who happens to be a writer/graphic artist/programmer):
"First STEM says that the humanities are worthless. Then they say how, really, STEM should get all that glory too because, really, it's the same thing -- but better."
Or this quote:
"Great paintings, for example, get you laid in a way that great computer programs never do. Even not-so-great paintings - in fact, any slapdash attempt at splashing paint onto a surface - will get you laid more than writing software, especially if you have the slightest hint of being a tortured, brooding soul about you. For evidence of this I would point to my college classmate Henning, who was a Swedish double art/theatre major and on most days could barely walk."
I forgot about Hackers and Painters! Pretty sure there's a copy on a bookshelf here somewhere...
Interesting quotes. Reminded me of a conversation I had with some local artists a while back who were staunchly in favour of their kids going into STEM because the humanities don't have a direct application to industry any more. They didn't want their kids wasting money on exploratory education, or in pursuit of the arts which they saw as a dead-end. I found that really sad, but it stemmed from an acute awareness of the direction industries like music are going.
Historically, the sciences emerged as areas of philosophy, but they've since become detached from those roots. Those roots have a grounding effect, a sense of the greater context of human life, that's missing in the culture around STEM today. I see echoes of that in the rationalizations made by Spotify's CEO: Who cares if the album is dead? Get with the new program, whatever the cost; it was inevitable anyway.
It is funny, all of those descriptors for STEM and detatchment seem to apply to the dying institutions which call themselves humanities far more than STEM. Now the fundamentals still exist strongly, there is plenty of phislophizing and art creation. But not through them.
Philosophy as a grounding context of human life? Talk about historical revisionism. Philosophy has long been about getting beyond the dreary context of human life into grand abstract universals. Even the ones which had pretenses otherwise like Marx-Engle's descendants or primitivists still engaged in the same thing. Revolutions were lead by college students instead of factory workers and farmers based upon theories that find validating observations not just inapplicable but rude and offensive to even propose. That they define themselves as fundamentally not scientists is the biggest tell to the ailment.
The Non-empiricists got left behind long ago, unchanging and then insist that the world has lost touch and not them. The split from philosophy and natural philosophy and from natural philosophy to science was from an inability to accept that they must acknowledge the world as it is, even if they wish to change it.
By doing so the old institutions and concepts refused to change and are "undead" - not truly gone but not growing or adapting to their environment like a living thing. To paraphrase from the Sixth Sense they're dead and don't know it. Now like Bruce Willis thet are wondering why their spouse silently refuses to speak to them at the dinner table and never listens to a word they say.
Harshly yes, you should get with the program because not doing so in some way (even if it is defiantly crafting their own path which may or may not fail) is denying reality.
Even the grouping of STEM together is a bad-taste-in-your-mouth industrial crime against human culture. The real acronym should just be M. (S, T, and E are just bastardized ways to get paid fat bucks for doing simpler versions of M.)
That essay was hilarious. The bit about how painters have figured out the most efficient and lucrative way to sit around staring at naked woman all day is a perfect example of the self-deprecating braggadocio that every man who writes a blog seems to be trying to achieve.
I'm often amazed by the negative attitude towards humanities and artists seen in this community. It it not that people hate art or culture, they just really hate the humanities department, art students and artists! (But at the same time they want to be recognized as great artists for writing a neat one-liner in Perl.)
Good response. It's simultaneously true that coding (and engineering and science and maths) is more artistic than generally accepted and that "art" in the conventional sense has more in common with those disciplines generally considered to be it's polar opposite.
There is an art to building software, no doubt, in terms of it being a craft. But I think you've misspoken about functional things being artistically beautiful. What you really should mean is that functional things can be aesthetically beautiful, which is true and different.
Art is primarily defined by expression and exploration, particularly revolving around human emotions. Software products, in the end, do not fit this category. There is a vast difference.
Edit: I defined it elsewhere a bit smoother. I define art as the exploration of emotional expression. Art products are the end result of that exploration. Live acts or art or performances could be viewed as a sort of merger between the two, either as a reenactment of the exploration or as a new exploration happening in the moment.
Software can be art in itself. Consider quines [1]. (Programs that, when run, produce themselves) They serve no real world purpose, and I consider them beautiful. Kind of like the restrictive format of a haiku. (Edit: Or perhaps tesselations, e.g. Escher's work [3])
I particularly enjoy what this artist(?) has done with their 128 language uroborus quine. [2]
Edit 2: I expect the objection will be that this is not functional, or a "product", however the uroboros quine does have practical uses. E.g. as a system stress test. If it were marketed as a stress test, its artistic value would still be apparent to a programmer examining its structure.
The statement of saying someone else's view of art is objectively incorrect or silly is in itself silly. I like your proposed definition of art, but as you know everyone is free to have their own conceptions.
I can agree with that to a degree. Of course, the gut reaction to any statement is typically the more emotional one. :)
However, I find that they didn't really make a statement about how they view art, although elaboration would help. I think people are confusing things being artistic versus being aesthetic. When people say "code absolutely can be art", I find it's too generic that it does indeed become a silly statement. It dilutes what art is and is overly biased to code, especially since it's likely to be coming from a software developer.
What I really think someone means by "code can be art" is that "code can be aesthetically pleasing", the latter of which I agree with. There's also a difference between "<some thing> as an art" as in doing the thing can be an art or craft and also calling the thing art.
The 80s underground scene had many zines that were programs.. the art would be part content part reader/application. The best part were the crazy menus.
The problem now is an application must look like all other applications or you would see more creativity. Making parts of the app 60% visible so the background came through had a cool visual effect.
Can you elaborate on what you're referring to or talking about?
Of course there are loads of software that people use in art. But in the end, those are still solving a problem and are not art. Examples are TouchDesigner, Processing, Logic, Studio One, Pure Data, vvvv, and much more. These software products are not art themselves but are rather solutions to problems to allow artists to create art easier. They are not art products, in my opinion, since they aren't the result of an exploration of expression.
Where the line is blurred for me is in live coding artistic performances. The software there is considered part of the art, as part of the performance. But even then, the software is a tool. If a person uses a hammer in a live art piece, is the hammer now suddenly art? No, I wouldn't think so. It's a tool serving as a utility or maybe medium. It isn't art itself. The art is the performance.
Of course there are always exceptions which cannot be captured in every argument. Something that blurs the line is something like DIN . But see, in such a software product, we begin to see expression and not just a thing solving a problem or being a tool. Similarly with Orca (although I wish they would change the name so as to not interfere with actual orcas which some populations are in danger of going extinct).
A tremendously important example is videogames which as well as being a multi-billion dollar industry akin to music are also a major medium of individual self expression.
In the broader set you have interactive fiction and a whole slew of art that shares a lot in common with time-based art. Demoscene is a good example of the latter.
Games certainly can be art! But I don't see how that conflicts with what I said or serves as a counterexample. Games are not just software products. They are the culmination of work by artists, designers, writers, and engineers. The end product is often an artistic piece, and a rather interesting one. Games being art doesn't suddenly mean software products are art. The software itself helps drive the art but is not the art itself. I feel a good way to think about it is that software and code running the game is not art in the same way that the paint in a painting is not art.
My favorite game designer is Fumito Ueda. His games of Ico, Shadow of the Colossus, and The Last Guardian are prime examples of games can be art.
Videogames are software products though so are a prime example of software products that are art. To quote your own recent definition of software:
"I have always viewed hard, firm, and soft as a description of the ware's malleability. Hardware is not easily changed or updateable, if at all. Firmware is typically code deployed to stay in hardware like an EEPROM, flash, or FPGA, where it can be changed but not necessarily dynamically or easily. Software is able to be easily molded and changed."
And whilst videogames do require traditional artists so do the construction of other mundane software products. Also invoking other disciplines rather ignores a lot of the work that goes into the gameplay itself which is entirely software driven.
Not to mention the other examples you studiously avoided.
> Videogames are software products though so are a prime example of software products that are art.
Did you think about or read what I described? Because I specifically addressed how I think video games being art does not make software products and code art. You aren't elaborating and then just restating an opinion I've already discussed.
If video games being art means that software products and/or code is art, then you need to consider that the material (paint, metal, etc.) used in paintings and sculptures is art as well and all the other repercussions of such a conclusion. No one would argue that. The paint in paintings is part of the medium and brushes and other such things are part of the tools of the process. In the case of video games, the software is a kind of both tool and medium of how that art is created and experienced. No one is saying software can't be a medium for art, but it isn't the art itself. And again, like I've already said, I don't think video games can simply be defined as software products. Other things called software products, like mobile apps, also have artists and designers, but there the situation is even more removed from art because the app is not expressing anything but is rather solving a problem, providing a service, or giving functionality.
> To quote your own recent definition of software
That quote has nothing to do with this discussion, so I'm not really sure why you're pulling it here. It was from a completely different post, and a non-native English speaker had remarked about how they had thought that the firm in firmware meant something else. I wasn't defining software but rather discussing how hard, firm, and soft in hardware, firmware, and software could be thought of.
> Not to mention the other examples you studiously avoided.
I didn't avoid anything. I didn't feel I had anything to say that adds to the discussion without repeating what I already said. Yes, they're examples. And?
I didn't say that. And just repeatedly stating something doesn't make it true, especially if others disagree and you refuse to elaborate, discuss, or explain.
Plus the implication that music might not be a tool for a specific purpose (which it is also absolutely in some contexts).
The thing is that "music", is not "one thing". It all depends on the purpose, if there is one, and on the intention of the composer and of the players and of the audience.
Here, Spotify's CEO adresses a very specific market in the music industry, not "musicians" per se.
This is a tired argument that lacks much nuance.
Ill believe this when people regularly gather in the thousands to see a brilliant coder at work or view amazing code in a gallery.
Are there not thousands of conferences each year? People giving talks at meet-ups, the yearly Apple developer conference draws thousands of people. All events that people sign up for voluntarily to listen to someone else talk about code.
As for viewing it; there are also many places where people participate in coding competitions or code golfs, where others judge and appreciate the code.
Is that really so different from going to a concert or an art gallery?
Code can be describe with aesthetic qualities like clean, beautiful, shitty, complex, delicate, etc. Just like music. Art is just a representation of human creativeness and imagination which is what software is.
There is no art anymore, only products. Once you optimize for monetary compensation via popular consumption, all creative endeavours are seen as and become products.
Your sentence, while it reasonate with me, is seen today as naive and non realistic.
My immediate question to that CEO would be, "OK, so what is it you're marketing?"
In the case of open-source software, they're probably marketing consulting, books, training. (I think conferences would likely be part of the marketing.)
And that, folks, is how we end up with those crapulacious open-source software tools (libraries, whatever) that are deliberately incomprehensible to anybody outside the producer bubble. e.g. Bouncy Castle: Utterly shite API to the point of unusable, despite the undoubtedly well done implementation underneath; completely absent documentation in any useful/usable form. Everything around it is just, "Buy our books, come on our courses if you want to use this library." The code itself is obfuscatory. Not a design document or user-guide in sight.
(Other examples abound. I just pick on the one that burned me the worst.)
So, back to music. How is this a good result for the "market"?
We end up with a flume of mediocre-to-crap music (but frequently! as if that's a good thing all by itself) all in the name of marketing,... what precisely? Where's the passion gone? Where's the art? Where's the music I'll listen to over and over again for decades?
I would disagree. I would instead phrase it as "addressing a need", and with that phrasing, I don't see a clear distinction between software and music.
That is really diluting the difference between art and software, so much so that it's almost comical.
I define art as the exploration of emotional expression, and software simply does not fit into that category except in the cases of using software as a tool or mechanism in the exploration.
The "need" in creating art is often described by artists in that the act of creation is fulfilling an internal need. Writers, musicians, etc. often describe their works as pouring out of them, almost by necessity. This is rarely the case for software and engineering products which seek to fulfill an external need. Software, like engineering, seeks to solve problems. Art can but does not in general.
I personally find it mindblowing and a bit disturbing that I see people in this thread equating software to art. It's a bit disrespectful and elitist to claim that they're the same. Artist make pennies and are being forced by market dynamics to change their expression to merely survive. Meanwhile, software engineers make six figures and retire early playing a game of connect the dots. It's flippant to those artists who struggle for years and years in the hopes of even a modicum of support or recognition and live on pennies.
Who retires first debate: the famous musician or famous programmer is still unsettled.
Classically an artist put themselves in difficult situations in order to grow as a person so the art can say more. Don't be angry at others making money because an artist has choosen a different path.
I define art to be the exploration of emotional expression. If you disagree with that definition, that's fine, as the debate as to "what is art?" is certainly a valid one. But please note the discussion context. The original comments I've replied to are "software is equivalent to art", and I disagreed. There are vast differences. My assumption is that it is people involved in software making this claim, and that is where the elitism comes from.
> Software is the product of expression.
What emotional content are you expressing when you write software? When someone writes a book or music or paints a painting or live performs an art piece, they are expressing emotional content, feelings, stories, etc. Please view my other responses, as I don't want to repeat myself here, but art is different than "<thing> as an art", which relates to art as in craft, and it's also different from things being aesthetically pleasing.
If we're going to dilute the term of what art is so much, then it's pointless to just stop at "software is art", as we should then just say "everything is art" and stop using the word.
> They both can evoke emotion while experiencing.
That's a poor definition. If something simply evokes emotion, that doesn't make it art. Literally everything evokes emotions in humans.
> Who retires first debate: the famous musician or famous programmer is still unsettled.
It's usually not best to discuss edge cases and is better to stick with the general population. Your general software developer is going to be many times well off than your general artist.
> Classically an artist put themselves in difficult situations in order to grow as a person so the art can say more. Don't be angry at others making money because an artist has choosen a different path.
You're implying that artists intentionally suffer? Some artists may, but I think it's more the case that most artists are not given a choice, which was basically the original discussion here. The discussion started off as musicians are now having to compromise and shift from creating music as their end product to simply using music as part of the marketing component of a brand, image, lifestyle, etc. And no one said to be mad at people making money. But the people sitting in one of the most well-paid positions in history, i.e. software developers, shouldn't be making loose allusions to what they do is the same as people living in poverty and the lower class struggling for society to recognize as what they do as useful.
You can also phrase brain surgery and movie watching as "modifying the inner thought processes of the subject", and as such there isn't a clear distinction.
One is art, the other is trade.
One can of course make art one's means of substinance, and one can master one's trade artfully.
What if they said, "Our product isn't our software, our product is our marketing"? This makes more sense because there can me many products offering the same service, but the one with the best marketing can win, even if it's not the best software.
> If the CEO of an open-source company said, "Software isn't our product, software is our marketing", would you feel the same way?
Yes, absolutely, 100%! By all means care about image and marketing, but if your actual product isn't good (software, music, or anything else), then who cares?
For software, actual usefulness. A word processor that's trending on Twitter but which makes it hard to actually edit documents is not useful. A graphics driver that actually renders my screen is useful even if nobody's ever heard of it (although I grant that below a certain size maintenance issues might come up).
For music... I'm far less qualified to say, but there's still a notion of (subjective) quality, which is only loosely correlated to marketing. I suppose we could say "good" music is that music which people enjoy, which still is driven by the actual music and not marketing.
My (implicit) point was that software can be optimized for one factor, which is usefulness. By evaluating usefulness, we can evaluate whether a piece of software is "good" or not.
However, use-value makes for a very poor metric for evaluating music (I am ignoring arguments about the "usefulness" of music when it's played on a factory floor, making workers slightly more productive). So what must we use to evaluate its "good"-ness? Anything you propose here will have certain people agree, and others disagree, e.g. one set of people say good music must make you dance, and another set will say good music must make you feel relaxed (the opposite of dance).
When aggregated, these individual differences will cancel out. So the average most marketable music must, therefore, be as bland as water.
I am speaking of the usefulness of software in a much more general sense - does it work?
Adobe Photoshop and GIMP are two competing softwares for a specific task - editing photos. Not everyone needs to edit photos. Those who do are voting with their money, because Adobe Photoshop worked faster and more easily at the same task than GIMP. (This can change, with the recent subscription-model only Adobe is pushing that is irritating its core users, but that's a different topic)
Let's return to dance music. Let's say we create a category on spotify and call it - "does it make you dance"? Well, for a swath of youth, EDM makes them dance. But for a swath of Latin-speaking users, bachata music makes them dance. Let's say they are of equal number in the population. You can see here that even averaging based on the category of "does it make you dance" will produce some rather unfortunate results (a combination of half-EDM-half-bachata that will fail to tug at the heartstrings of anyone at all, but might be vaguely palatable vaguely dance-y playing in the background of a grocery store). However, this kind of music - the one that can vaguely capture both audiences - is exactly the one that will get the most people to listen to it, and based on Spotify's model of quantity over quality, it is the one that will get most rewarded.
But some open source software is marketing, is it not? The big open source frameworks released by tech companies come to mind. These projects shape the industry and advance the company's recruitment goals. Bootstrap. React. TensorFlow. Kubernetes. WebKit.
If that CEO was in charge of Redhat, sure. Or maybe one of those open source SaaS companies, but many of them are realizing that maybe giving away the software was also a product decision that is now harming the bottom line.
Isn't is? open-source company typically don't sell 'software'.
They sell 'support' ans 'service'. And even microsoft, a typical software company do this now
I see your point. But I still think that if the CEO said "software is our marketing- our real product is our influence", I'd still find that depressing.
I agree with you on DLC. I'd rather just pay $100 upfront (or whatever it needs to cost) to get a finished product.
on the other hand, I think F2P monetized with cosmetic items is a pretty good model for a multiplayer game. anyone with a computer can enjoy the game, and the people who care the most (or have the most disposable income) support ongoing development without messing with the balance.
I think it could be depressing if we think of "your product" as "the most important part of what you do". But we could also think of it as "the thing that you need to make money from". If the music itself isn't the thing that you need to make money from, then instead of charging a bunch for it, now you want to spread it as widely as possible as cheaply as possible. That's a good thing!
> When books or pictures in reproduction are thrown on the market cheaply and attain huge sales, this does not affect the nature of the objects in question. But their nature is affected when these objects themselves are changed, rewritten, condensed, digested, reduced to kitsch in reproduction, or in preparation for the movies. This does not mean that culture spreads to the masses, but that culture is being destroyed in order to yield entertainment.
-- Hannah Arendt, "The Crisis in Culture - Its Social and Its Political Significance"
> If the music itself isn't the thing that you need to make money from, then instead of charging a bunch for it, now you want to spread it as widely as possible as cheaply as possible. That's a good thing!
It is, but only if the music gets better in the process. Unlikely to happen if you rush for faster releases. More production doesn't equal to more novelty.
> but only if the music gets better in the process. Unlikely to happen if you rush for faster releases
Actually, if you are constantly working at it, you'll probably get really good at it and the quality will get better and better.
There's a story about a ceramics professor that would divide his class in two at the beginning of the semester. He would grade one group by the quality of their best piece at the end of the semester and the other group by the weight of all the pieces produced during the whole semester. The professor said that invariably, the best pieces were produced by the people in the group that was graded by the weight of their pieces. His hypothesis was that people graded by the weight would produce a lot more pieces, so in the process they would also get really good at making the pieces, whereas the people graded by their best piece would just spend too much time on their pieces trying to make them perfect and wouldn't get enough practice to actually become better at making them.
I really like this anecdote, I find it very prescriptive for life, but I don’t think it applies. Professional musicians aren’t learning how to make music like the students of the ceramics class, they are mostly fully developed artists with their own artistic style and voice. I think that constantly producing art leads an artist to become a slave to his prior work, a sort of creative fatigue. In order to foster development of an artist’s voice, they need space to get away from that style and explore other styles and forms of art.
Not a popular opinion here, but I really don’t think that coding is very comparable to pure art like music or painting, and I think that a lot of these ideas come from that place(that coding is art). Coding is art like electrical work is art—there is certainly a distinction between work crafted by a master and something muddled together by an amateur, but at the end of the day it’s functional. In these trades it’s perfectly okay to have creative fatigue as long as all the parts are good. If you have great variable names, nice modular form, terse functions, excellent descriptive comments, etc, its actually better if you have a monolithic unchanging form. Pure art, that is art meant to be consumed in the form its created, by contrast, brings with it all kinds of aesthetic values that are put by the way side in coding, and novelty and creativity are front and center.
> Actually, if you are constantly working at it, you'll probably get really good at it and the quality will get better and better.
That's true for engineering, but for music it only goes so far. A lot of musicians working this way will, after a while, find a formula that repeatedly sells, but loose a lot of originality in the process.
You don't necessarily have to "rush" for faster releases, you just have to think about releasing in a different way.
Think of it like a software roadmap: Is it better to wait 6 months and release a bucket of features twice a year, or to release features gradually throughout the year and measure their effectiveness? Some might say that releasing features gradually is "rushing things out", while others might say that it's the only way to ship non-broken software that people actually want to use.
That's one of the aspects of the new economy... only a very small minority of artists can literally go dark for 3-4 years to write something that will rise to the top instantly.
Not if you’re coerced into having to produce a piece of music every month/couple of months to keep the machine running.
That’s just producing a commodity.
Some groups have got it together and are able to fund years of quiet work before releases, but no matter how much I enjoy creating I would hate to be so bound as outlined above.
I am reminded of the story about ceramics students being told that half would be graded on quantity and half would be graded on the quality of a single piece of work.
In photography, you can get lucky. When I come back from holidays and review a thousand of random snapshots, a couple of them look great. Take 200 shots of a single thing, and one of them might be pro-level due to pure chance. Yet I'm not sure how well this "randomly great" percentage would translate to me playing the violin, dancing ballet, or writing a song.
After hearing this story for years and finally trying to source it, as far as I can tell it's not entirely clear this experiment was ever actually run -- and in fact is sourced in the link behind your link as a "parable."
I can imagine it's true that churning out quantity is its own education but I can also imagine there's a plateau past which deliberate practice and polish matter quite a bit.
You’re right. It’s not new. Not for pop music anyway.
I’d argue that it’s worse, though. Shorter cycles are even less healthy.
And just because it’s been done before(-ish) and other endeavours run in this timeline doesn’t make it good or beneficial to anything but those who aspire to only make money off of it.
The fact is, good music doesn’t require business to be good music. And the business doesn’t necessarily require the music to be good—it just requires that it directly or indirectly generates revenue. From the perspective of commerce the rest is incidental.
Encouraging the commercial perspective over something more balanced is a net negative for culture and society. At least I think so.
It’s still possible to batch composition and production with this release model. The lead time saved from releasing digital only can be used to polish the earlier tracks in the schedule.
I dunno. I've known a few people who dreamed of being in a band, and they weren't excited about the part where they sit at a desk revising lyrics; they were excited to perform their songs, to get up in front of a crowd and have everyone hear them.
It’s not tho! It means tighter release cycles which means a tighter feedback loop which means bands can find their sound faster and grow their backing faster.
Some of Steve Albini's interviews are really on point with regards to this stuff. Really worth listening to if nothing more than to provide contrast to comments like the one above (which seem pretty reductive and flippant): https://www.youtube.com/watch?v=yRAc3hx5pok
That's not depressing. That's one part of the "music industry". It's a huge sales channel, per se.
Now, being a musician or an artist, you decide what you do with it. If you consider that what matters is your sales and nothing else, Spotify's stance might apply to you. But you might have different point of view as well, still make music, and still live from it, perhaps.
It's a bit disrespectful to reply to someone saying something's depressing by bluntly saying it's not. It's not up to you to judge what they find depressing.
I also find it depressing. It's a societal problem that we do not value art in the same way that we value science and technology. We see it everywhere from the STEM myth to arts educational programs and artists struggling just to survive. Even earning my living in science and technology, I personally have grown to believe art is far more important for human progress and development than science and technology. It's a real problem that we blindly let markets decide what is important because it is in fact humanity's lack of progress emotionally, which is unfortunately tied strongly to our biology, but our insane progress in technology that causes almost all of our problems.
Huh? Someone's saying "that's depressing" is a statement. It's not "I find this depressing" or "I'm depressed by that". It's not disrespect, it's bringing a different perspective that does not see the depression there.
I don't find this depressing. I don't even find this societal at all. What would be depressing would be to take Spotify's view of "what music should be" or "how musicians should work/behave" for a relevant, world-wide truth. When it is only a very specific, skewed and biased take by someone's who's got a direct interest into making people believe that it is the only way to go.
That many people go this way does not mean they are right, neither they are winning.
I think I vaguely agree with much of what you said, but just want to point this one out.
> It's a societal problem that we do not value art in the same way that we value science and technology.
Want to give a couple of examples against this:
- some of the most valuable software out there is free and open source. The authors of said software don’t get compensated anywhere near the softwares value. Coz everyone uses it for free.
- I very much value the games I play on my iPhone, but I have never paid any money for them.
- I will buy Opeth t-shirts, and Akerfeld custom guitars, but I’m never going to give Spotify as much money.
So, I don’t think your claim is so clear cut as you state it.
If you instrumentalize music to maximise your influencer career with the content of the music as an afterthought what you get isn't art but a mediocre commodity.
It's like Ricky Gervais remark at the Globes, the actors aren't actually actors any more, it's a competition for the best looking, most ripped steroid junkie.
And why most is utter dross. People still buy millions of classic albums from the 70's and 80's, in 30 years time very few artists will have that same comfort.
> People still buy millions of classic albums from the 70's and 80's, in 30 years time very few artists will have that same comfort.
We are only listening to the best stuff from the 70s and 80s, and in 30 years people will only listen to the best stuff from today.
Now, I think a major difference will be the "best" music people will be listening to will be very different from each other. 30+ years ago the popular music people listened to on the radio or MTV was much more broadly known and shared than music today, whereas today different demographics have increasingly siloed music tastes.
I hear a lot of good stuff my teenage sons play on Spotify that I would never hear if it wasn't for them.
I agree with this. There's definitely way more available today, and in many more niche genres than there was available in the 70s and 80s. Arguably there is also a higher proportion of trash because of the ease of entry, but because the fanbases can be so segmented people can find stuff they really like that's more geared toward their specific tastes. Plus, there are greater search tools than before, like Spotify.
If you think modern music is bad you should listen to more music.
There are artists that can publish work now, thanks to the internet, that publishers wouldn't even dream of going near let alone giving them deals. I really like microtonal electronic music for example, not exactly radio friendly.
People still listen to Metallica's music, and their music still ends up in movies.
I don't really think all new music is bad, just most of it. The same could be said for back in the day though. Metallica is a bad example for you to use cause they are still widespread because they are really good. There are other bands you haven't heard of from the 70s and 80s.
Unless you're really, really into music, your taste in music essentially stops developing by the time you hit your 30s. If you're listening to Metallica today, it's because you were listening to Metallica when you were 15 (I was, and I still do).
This "all new music is dross" argument is repeated every generation as new music out-ages the outgoing generation.
I like Metallica just as much as the next guy, but they're not relevant anymore. They've been laughed out of the music industry ever since they got upset about their music being pirated - which is actually a great example of how the music industry has changed and left them in the dust. Do they still sell records? Sure. But that's not what I'm talking about. Would I listen to their opinions on modern music industry trends? Absolutely not.
Of course there are bands I haven't heard of from the 70s and 80s. That doesn't mean the ones I have heard of are relevant. It means they're memorable.
> Do they still sell records? Sure. But that's not what I'm talking about. Would I listen to their opinions on modern music industry trends? Absolutely not.
I bet they feel just terrible to only be making money off of people buying their music, but not paying attention to their opinions.
The way in which Spotify (and other streaming services) hurts the development of music isn’t what most people say.
The problem with per-stream plays is that it is based on pop music logic: more streams
equals more money.
Fair!
But many recording artists made a modest living by releasing music that you might not want to listen to over and over and over. This music is often labeled ‘art’ or ‘experimental’ or whatever, and for the enthusiast, paying $15 for a record of this stuff was great.
Streaming services have no answer for this problem. It’s up to something like Patreon, if anything, to fill this gap.
This is all sorta fine, but what grates in me is that it seems none of the streaming providers have even thought about this. They just don’t seem to be that in to music, and just think of types of pop music as representing all of music.
This is where that above quote is just a little irritating.
If I pay $10 and listen to 50 tracks, I want $10 (minus the provides fee) to be split among precisely those 50 tracks.
The difference is if you only listed to 50 tracks your listens are worth more that someone that diluted their listens over 5000.
Throwing everyone's fees into one big vat and dividing that according to the total listens, is bullshit. I want my favorite artist to be paid by me and I don't want to subsidize the worlds most popular artists that I never listen to.
That doesn't take into account the spotify cut. What would you propose? Right now spotify pays a certain amount per song played, and covers this cost by charging $10 of whatever, and hoping the average user doesn't listen to more than song_cost*songs_listened. In your scenario, suddenly spotify's business risk has been transferred over to the artists themselves - if your songs get listened to by people who listen to a lot of music, you're going to get paid less per listen.
>if your songs get listened to by people who listen to a lot of music, you're going to get paid less per listen.
That still would be fairer than the current system and that edge case would be rare.
As it stands today, we're all heavily subsidising the most popular (and well marketed) bands -not just their fans- leaving crumbs for other artists who end up having very little chance to see their lot improve at all.
I'm paying Spotify to listen to the artists I like, hoping to support them so they can make more music I like.
Instead, nearly everything I pay ends up in the pockets of artist I have no affinity for.
There is a gut feeling that indies and up-and-coming artists are losing out with their fans effectively subsidizing pop that gets more ‘disposable’ plays.
I’m sure Spotify will know the numbers for moving to user centric streaming payouts, would be very interesting to know the truth of who the winners and losers would be.
I think the option to be user centric could be a compelling feature for premium customers, to show money is going to artists fans love, and ultimately move more people on to the premium versions of streaming platforms
Personally at this time I’m more concerned that my Spotify subscription is subsidizing podcasts by Joe Rogan than Drake / Taylor Swift.
Streaming services (music, movies, books, etc.) are completely user-centric. It’s one place for you to consume your content. Discovery was never great there.
Spotify is actually pretty good for music discovery. For example, I routinely find new artists and songs from their daily mix and other generated playlists. It's miles ahead of what Apple Music could come up with even with way more listening data to analyze and is an exception amongst services like this.
I prefer the Kanye West model in which every 8 months or so you promise to release an album on a specific date. Then that date comes and you go completely MIA, and pretend you never even announced the album. Then a few weeks later once all the hype has turned to dismay you quietly release the album and the dismay turns back into hype. Maybe that model only works if you're Kanye though.
Kanye West is a singular being. He is a human work of art, this generation’s Warhol. Perhaps even more fascinating than Warhol, if less self-aware. I’m a huge fan of his brilliant public trajectory, yet I despise his music.
I used to be on your side of the camp, but now I'm both sides.
The guy just struggles with some mental health issues and isn't that great at talking to the public outside of his music. I disagree with him on some things, and he definitely has his issues, but I think he means well for the most part.
The problem is that if you're a nobody, you can say stupid things but mean well. If you're a global celebrity, when you say stupid things, there are consequences.
I struggle to find anything that Kanye has ever done, in or outside of the studio, that will be held as artistically relevant or meaningful in any way in twenty years. His music isn’t groundbreaking or innovative, and anyone can be an asshole to people.
At least Warhol had interesting and relevant things to say.
Agreed. I really don't understand what people find on Kanye. All I see is someone who is clearly only focused on selling as much as possible. I am yet to find a Kanye song that felt really unique in some way. Comparing him to Warhol feels almost offensive.
Kanye West was the first big artist I know to really utilize The new release model to break out of the old formula of releasing an album and then it’s done.
It was fascinating when Life of Pablo was released on Spotify and over the next few months was given continuous updates and patches, like it had been software.
Really proves the old proverb that great art never is finished, it’s abandoned...
I wonder if being able to update albums has been a good thing. It may encourage people to release "MVP"s because they can always improve them later. It doesn't quite fit because reviewers will review the first version.
For me personally, it has been a negative, because I really loved Tame Impala's Borderline, which was switcheroo'd with a song I don't feel for when the album came out. They even took out the single version from Spotify and YouTube (the official channel), so we're stuck with the album version. A comment from unofficial YouTube version sums it up:
"I understand him wanting to make a version with the vision he originally intended for the song but the fact that he removed the other version when it already was presented and existed to other is what is infuriating and upsetting. The single version clearly held a special place in people's hearts and to lose that is hard to swallow."
There is a very scary thing about digital art where platforms at any time can remove things or silently release a new version. It makes me uncomfortable in the same way as when newspapers online make updates to their articles with no transparent revision history of what was changed and when.
The technology just makes it easier though. The Foundations have been trying to erase the original Colin Young version of Build me up Buttercup that I love in favor of Clem Curtis rerecordings since long before Internet was a thing, as an example.
So just aggressively back up versions you like as they are released. If you don't save it for yourself, you can't guarantee it won't eventually disappear.
Obviously the guy knows nothing about creativity and art.
It takes time to grow as a person and cultivate new ideas. It may not be possible to do that if you don't get in and out of whatever it is you're doing.
Imagine telling Hemingway he has to release a new blog post every week instead of a book every couple of years.
"What's he going to record a song about?"
"Nothing."
"Spotify'll kill him."
"I guess they will."
"He must have got mixed up in something with the music industry."
"I guess so," said Nick.
"It's a hell of a thing."
"It's an awful thing," Nick said.
They did not say anything. George reached down for his mobile phone and wiped the screen.
"I wonder what he did?" Nick said.
"Failed to write enough songs fast enough to generate user engagement. That's what Spotify will kill them for."
"I'm going to cancel my Spotify subscription," Nick said.
"Yes," said George. "That's a good thing to do."
"I can't stand to think about him waiting in the recording room and knowing he's going to get it. It's too damned awful."
"Well," said George, "you better not think about it."
I don't know if this was intended sarcastically, but novels written in a weekly serialized form were a huge thing [0] and Hemingway wrote at least one novel this way [1].
My point was not that artists need a long or short time to produce something, but rather that they need to pause and reflect on things between projects.
What’s especially funny about your example is Hemingway wrote for periodicals as a day job for a long time, and so he kept that kind of schedule.
Except some of his books came about because he didn’t keep to that schedule, and we don’t usually talk about Hemingway for his articles in the Toronto Star.
Oh! I didn’t mean for my comment to come across as a disagreement of any kind.
I was just making a side comment. I thought it was funny and that your analogy was even more true because we don’t talk so much about Hemingway’s frequent articles in a newspaper but we study at length and hold in high regard his long form works.
Obviously the guy knows nothing about creativity and art.
He's not telling bands how to be successful artistically, he's telling them how to be successful financially in the modern music world.
If the band is only interested in promoting their art, they can do anything they want. If they want to earn a living too, they need to provide the market what it wants.
Generally, the financially successful artist will be touring and playing live shows and between that spend a lot of time in a studio recording many songs at once, not recording a song every month for Spotify pittance.
> Imagine telling Hemingway he has to release a new blog post every week instead of a book every couple of years.
In his prime writing period he was closer to the former than the latter; between 1923 and 1933 he published, it looks like, at least 49 short stories, 10 poems, 3 novels, and two nonfiction books. At least one of the novels was initially released in serialized form over 6 months of magazine issues.
I think he just sees the norms of traditional songwriting as not his job. If I'm the CEO of Chipotle, I'm going to focus on getting my customers good food at reasonable prices, and I'm not gonna take it tremendously seriously if chefs complain they can't express creativity well in my kitchens.
I'm almost sure he knows all of that about creativity and art, and simply does not care. He has instead decided that he is going to try to commodify music, and this is the result. Factory-farmed music benefits the distributors over the artists, after all, because it's reliable.
Spotify is not for putting out art, it's for generating buzz about your art so you can get paid for it later.
If your goal is the best art you can produce, then perhaps Spotify and that roller coaster isn't what you should care about.
If you want to good art while also also getting a lot of buzz so you can make money or be popular, then you need to deal with market forces, just like everyone always has.
Different artists will value to varying degrees generating good art, exposing many people to that art, and getting paid well. All those goals are related, but I doubt any one of them can be maximized without concessions from others.
> Imagine telling Hemingway he has to release a new blog post every week instead of a book every couple of years.
Authors release at their own schedule if they don't rely on the money from the books, they have other income, or they've already made enough to be secure. Hemingway is not really different. Neither is any other already famous artist.
> Spotify is not for putting out art, it's for generating buzz about your art so you can get paid for it later.
If your goal is the best art you can produce, then perhaps Spotify and that roller coaster isn't what you should care about.
And how exactly will people get paid, when Spotify (and other streaming services) have effectively replaced album sales?
This is just the same tired old "exposure" argument.
> And how exactly will people get paid, when Spotify (and other streaming services) have effectively replaced album sales?
Touring and merch sales, which is why the aggressive release schedule on Spotify to keep buzz up is supposed to be good in the first place.
> This is just the same tired old "exposure" argument.
The whole argument about needing to release more often is predicated by that exposure argument. If you don't think it works, then when you release on Spotify doesn't matter at all.
Either you accept that Spotify buzz drives attendance and thus more buzz from more releases more often brings more money, or you don't accept that and then there's no pressure to release more often.
Right, but my point is that if you're absorbed dealing with market forces you won't be able to produce significant work. The more cookie cutter stuff you release, the more you will adopt that mindset and it will consume your creativity.
I guess in the past musicians toured and performed to satisfy that need, and relegated the "important" work to long term writing/composing. With Spotify (and specially now with COVID) it seems this has changed since the only way to generate income is by releasing music.
My guess is that things have only changed for the mega-stars. Household names that are in the public consciousness need to work to stay at that level, and keep people hyped about their work. I think those people are likely already making significant concessions in their work for this, so it's not like it affects them that much.
Does it really matter for artists that aren't being played multiple times a day on iHeartMedia/Clear Channel radio stations already if they follow this model? I doubt it, unless their goal is more about fame than art, in which case the question answers itself.
> Imagine telling Hemingway he has to release a new blog post every week instead of a book every couple of years.
Hemingway's fiction in particular aside, it is worth noting that many of the novel-length works now considered classics were actually created and released as a series of shorter works, often to fairly strict deadlines:
> Imagine telling Hemingway he has to release a new blog post every week instead of a book every couple of years.
It's much easier when there are literally thousands of song writers trying to get a label or famous singer to use their song. The Nashville model is pretty much everywhere now.
It also helps that we've been able to computationally determine which parameters and heuristics are most likely to end up as a hit. There's a reason popular music across genres all more or less sounds the same today (loud, compressed with little dynamic range, auto-tuned, 4 chords, etc), with the difference between them being the style they are in.
It's about releasing, not creating. If you write 12 good songs a year on average, you can release 1/month. It doesn't matter if you actually write one song per month, or write and record them all in one frantic week, and then spend 51 weeks growing as a person and cultivating new ideas.
Agreed, not just the monetization scheme but Spotify and others just push the music listening experience to playlists and singles and have very little room for albums and collecting a library of music you refer to.
I’m not a music aficionado, and I have no interest in albums. I love the singles/playlist format, and consume far more music now that I can easily ignore the 9 out of 10 songs I don’t like in an album (back when CDs were still a thing).
I also find it easy to have a library of singles that I like.
Most of an album being bad as a stereotype I believe was just from actual bad artists when basically anyone could get a deal and possibly go platinum from a single song. Good artists typically had good albums
Possibly, but I think albums are only significant due to the historical medium of music requiring grouping songs together on various disks. Once it’s digital, I don’t see the importance of grouping certain songs together, outside of certain categories like movie soundtracks.
I guess some albums could tell some story about what the musician was feeling at some time, but that probably doesn’t apply for a lot of music if not most. Certainly not the music I listen to, release it one at a time or all together, makes no difference to me.
it really depends on how you like to listen to music. an album is often like a snapshot of the artist's sound at a certain point in time. if you listen to death cab for cutie's first and last album back-to-back, you might not realize it's the same band. if what you want out of a listening session is to hear a bunch of vaguely similar sounds in no particular order, an album doesn't do much for you. if you're in the mood for a more specific sound, a good album can be a lot better than eight solid tracks.
What do you think Bandcamp should be improving? What would a better alternative look like? Genuinely curious as an artist pretty satisfied with Bandcamp.
I don't think there needs to be an alternative, but Bandcamp needs to put more effort into getting more artists on its platform and for those artists to expand their catalogs. I buy a lot of albums on Bandcamp but it's annoying when I see a new release somewhere and the artist isn't on Bandcamp, or when an artist only puts their latest album there.
Bandcamp could also put more effort into recommendations based on a user's purchases, or just when viewing an artist's page. Not "we recommend this new album that is trending" but "this artist is similar to this other one" in the style that the old what.cd site had, with its massive graphed connections from one artist to another based on style (within a genre) of music.
The UI design is just awful for this day and age. It looks like MySpace from 15 years ago.
For example you get to an album page and there's just so much crap. The album and the player should be the main focus but the player is this tiny thing relegated into a corner.
Bandcamp's privacy settings are also pretty awful. There is no mechanism for making my library private; anyone can search my username and see what music I own. I suppose you can "hide" individual albums, but that's not great.
I wish more people, especially tech CEOs, could understand the incredible sacrifice of time, money, and energy it takes to make an album people actually enjoy. Everything you listen to, unless you are willing to listen to raw demos, is a miracle. I've worked both as a sound engineer and and a software developer. It's orders of magnitude easier to make working software than it is to make a hit record. The way tech companies have abused musicians to get rich is one of the most shameful things to ever happen. Furthermore, I'm sick of these guru edicts about how everyone will need to work that much harder in the "new normal". I got into software to pay the bills, but the way geeks talk about the industry and people I love drives me right up the wall. Have some appreciation--even awe--for the art you enjoy. It requires more effort than you could ever imagine.
> It's orders of magnitude easier to make working software than it is to make a hit record.
That's apples-to-oranges. Obviously a "hit" anything is hard in any field, by definition. A hit app is just as hard (even harder, probably, since there are a lot less hit apps than hit songs).
But producing a single track? Seriously, it's not that hard. Yes, it takes a team of people and a lot of creativity and skill, but I don't know what this "incredible sacrifice" or "miracle" that is "more effort than you could ever imagine" is that you're talking about. It's a creative project like any other.
But that's not even the point of the article, which is simply to release tracks piecemeal and regularly rather than in an album only occasionally. There's nothing about everyone working "that much harder".
It sounds like you had a hard time in the biz, and I'm sorry. But I don't have any clue who the "geeks" are who "talk about the industry" that seems to bother you so much, and which doesn't seem to have anything to do with the article. I know a lot of software engineers who are also really into the indie music scene in Brooklyn and I think everyone does understand the work bands put into their music and their touring, and that most of the bands are never going to make it beyond attracting a few dozen or couple hundred audience members at any show, but they do it because they love it.
Not touching the whole heft of the argument, I would like to point out that the original poster didn't compare the difficulty of writing a hit song to a "hit app", he compared a creating a hit song with creating working software.
I can co-sign on this part of his premise. I worked for a music producer who has a number of #1 hits (Beyonce, Fergie, John Legend etc). He wrote >1 songs per day almost every day for over a decade.
His success ratio (hit vs placement vs nothing) was much lower than any mediocre software engineer trying to make working software over the same period of time.
It feels safe to say that in a 365 day period, a software developer can reasonably expect significantly more than one of those days to be used in 'working software' right?
But working software doesn't mean commercially successful software. It just means it runs and does something.
"Working software" would be comparable to "listenable music". I'm sure every song that producer wrote daily was listenable.
But even if you mean "working software" as something that goes up on the app charts, top 100 in a popular category? I mean, then no -- I could easily see a developer spend an entire year building 1, 10, or even 100 apps and have none of them gain any traction at all.
Sorry for the late response, I just mean to suggest that it's pretty clear to me that your average software developer reaps the benefits of his work, day to day, on a higher percentage than your average producer.
There are 4.4 million software engineers in the United States. I would imagine most of the work that is generated by them on a day to day basis moves the needle of their careers forward. I could be wrong, but I bet most of them are employed, and getting paid for it.
At the very least I bet that more than half of them get paid for more than 5% of their workdays, right?
I doubt the same can be said for music producers, at least the ones making hits that the OP was referring to.
The average income indifference between artists and software developers, and how many of those who make up those communities are able to find any sense of financial stability, are vastly different, so even under your logic that a “hit” is a “hit” no matter the industry, is still really misinformed and I’m sure we are all out here just doing what we are doing for the “love of it”.
Speaking as a producer and developer, they are not the same things in any way and I love doing what I do deeply, thank you very much.
You're still off quite a ways in your analogy. I get paid to develop software that someone else owns and brands and makes all decisions for. For nothing more than money. It's nothing like my passion projects, which I do not get paid for at all. Most of the time. Maybe 1 in a 1000 that I will get paid for them in any meaningful way.
This tendency to hyper-analogise everything oh so sinks my boat.
I can’t think of even one example of the use of analogy that has helped me understand a subject better than if it were discussed and explained on its own terms.
It seems all analogies do is enable the analogiser to delude themselves in to thinking similar is equal to same.
Analogies may be less about equivalency and more to do with bridging ideas. They can help the analogiser draw someone closer in thought, like a shout in the dark to help your friend find you
So hard to put into words because the role is so varied but thinking of them as an artistic coach or mentor to the band while recording is perhaps accurate enough. Often hated or loved by the band and or fans of the band for the direction they push the music and the overall sound of the music.
As a part-time musician, even a single track can take tens of hours of effort - depending on the genre and the track.
Sure, you could strum a guitar and sing and have a decent song wrapped up in an afternoon.
But if you're creating anything more complex, you will spend hours finding the right sound, more hours creating the right arrangement, and even more hours fine-tuning the vocals, sounds, and final mix.
Something as simple as getting a snare drum right can take a ton of time. You don't hear it because you don't know what the bad snare sounds like.
Again, it all depends on what kind of music you're trying to make. If you're copying an existing style, it's not particularly hard. I can make a Drake-like beat in an afternoon - a quick search on Splice will even show me the exact drum sounds to use.
But if you're creating your own sound? That takes effort and experimentation.
Creating a song - easy. Creating music that's distinctively your own - tons and tons of effort.
Also, "tons" is amazingly overstated. Very few of those 16 year olds are producing "good" music that appeals beyond their immediate demographic, and that's to say nothing of how low of standards many youth demographics have.
It probably took me about 10 years to get good at programming. Sure, I could create programs pretty quickly---but making something like [1] is another matter altogether.
I guess it depends on how you count, but not really, at least not for the first 6 years or so. The first two-ish were high school, the next four were college, and the next six were in graduate school (so on a grad student's salary). After that I got a "real" job.
For what it's worth, I do other artistic pursuits as well. I did music for about 15 years though I wouldn't say I was ever good. I've been writing for about 10 years. I would say I'm just now barely at the point where I think I can write half-way decently.
To be clear, I am very aware, and very grateful that one of the things I happen to like doing also pays well. I was responding more to the "you can't imagine the level of effort required to do this work" aspect of the ancestors' comments. Yes, I can imagine, thank you very much. While I don't like where we're going with funding models for art, that doesn't mean that understanding the effort is the limiting factor.
I don't think anyone is disputing that ROI is better with software.
That said, programming for 10 years before getting a job in the are is not uncommon. Most of my peers did it for at least five (I learned 15 years before, but I grant that's not that common).
Artistic endeavors always take me a long time, but then I read about people like Kanye who puts down a hit track in 20 minutes (he’s apparently known for working this way). You may not line rap or him, but he has a knack for hit beats.
This is probably similar to the apocryphal story about Picasso:
A woman who approached Picasso in a restaurant, asked him to scribble something on a napkin, and said she would be happy to pay whatever he felt it was worth. Picasso complied and then said, “That will be $10,000.”
“But you did that in thirty seconds,” the astonished woman replied.
“No,” Picasso said. “It has taken me forty years to do that.”
Yeah. Many don’t realize he was making beats for well known rappers before the general public knew who Kanye was. I still think his best talent is as a beat creator/producer.
But if you put out a concept album, do you expect it to pay your bills for three or four years? I don't think that's a reasonable expectation. If you want to make a living off music, then you might need to put out more music, or do other things like playing in a cover band or being a studio musician.
Compare it to actors/actresses who do hollywood movies, but also do indie films. If you want to make a living, you might not be able to do whatever it is you want, and might need to put out some more mainstream stuff.
Only a few people really get to do what they want in the arts and make a living.
Have you ever composed music that others have liked? I strongly doubt so, otherwise you would refrain from comparing developing a hit app to a hit song, to save yourself the shame of looking like a fool. apple-to-oranges, as you said.
Last year the founder of Shazam told his story at a conference. That‘s a hit app. The effort that went into it is mind boggling. Far more than ever could possibly go into a single hit song.
I'm not talking about complexity but about the creative process, which is totally different. You can't apply SCRUM to the composition of a song. You can't set a deadline to it. You can't hire a junior musician and have him write parts for you. It is a much less rational and plannable process than software dev. Hence apple-to-oranges.
A good way to draw that line would be “what level of success is required to make a decent living, house, and typical luxuries of life” compared to your peers. For a software developer, this is not as hard as making a “hit”. A regular 9-5 job at any one of a thousand companies is enough. I don’t think that is the case for music industry though.
What is the relevance of offering insight into the process of creating music?
With or without insight, it's obviously apples-to-oranges to compare "working software" to a "hit record".
Most "working software" is rarely used and makes no money, look at the long tail of apps in the iOS App Store, Google Play Store, etc that didn't become hits.
It's comparing effort to get to the same monetization.
Software is easier to create, doesn't need to be a hit to be used, and you can be paid to write it and to sell it.
Music requires much more creative work, the production is often unpaid, and the work product usually doesn't make much unless it's a hit with millions of plays.
I find that in software, it's much easier to anticipate change. Companies are more open about their future plans and long-term goals. For example, GitHub just publicized their roadmap. This entire website is predicated on the fact that people like to share their latest ideas, inventions, and solutions to problems that everyone faces.
I've never seen Universal Music post their "roadmap" of how they plan to sign new artists, or what new social media marketing ideas they might have up their sleeve. This is because the music industry is, by far, more competitive than the tech industry. Record labels, publishing groups, etc. are not very willing to share their "trade secrets". If the tech industry was like the music industry, there would be absolutely no such thing as open source.
Also...
> I don't know what this "incredible sacrifice" or "miracle" that is "more effort than you could ever imagine" is that you're talking about. It's a creative project like any other.
Your second sentence answers the question in the first sentence. :)
Well,in a way, software industry is very similar: "Aaaa, Britney!!Oh, wait, who's that?Lady Gaga!!!" and then "Java is cool...I mean JavaScript..No,no,I mean Golang... Did I say Rust?"
> It's orders of magnitude easier to make working software than it is to make a hit record.
True, but hardly apples-to-apples; it's orders of magnitude easier to make music than it is to make a viral app. In either case, the barrier to entry is low, the standard for doing it professionally is surprisingly high, and making a hit is a huge undertaking.
If you have an orchestra with 100 musicians who receive something like $10^5 per year, you pay $10^7 per year. You need several years to practice and to form the orchestra into a team with a repertoire. If it takes 10 years, then that is $10^8.
No you don't. Orchestras come together and perform completely new works with, if they're LUCKY, 1 or 2 rehearsals. Years of practice is completely ludicrous.
Your numbers are grossly off here. I would bet that zero orchestra musicians are making anywhere close to $100k just from playing in an orchestra - the absolute top of the top might make that across all of their engagements. Most are making far, far less.
In good orchestras in the US and western Europe, $100k is actually fairly standard, perhaps even on the lower end. You need to be highly skilled, the competition is fierce, and the job is very demanding.
I don't have a good source, but a quick Google search should confirm that $100k is not exactly wild.
As a musician and programmer I find the bar to make money on apps way way lower than on music. Being a good programmer brings you much closer to making money on an app than being a good musician does to get a record out. For one you don't need luck, money and the right contacts to release an app. It isn't even in the same ballpark IMHO. Pro musician is closer to "race driver" than "viral app creator".
There is a musician equivalent of app creator. Where do you think music in TV shows, movies and YouTube comes from. A human musician creates it. Atleast for now...
But I don't need to "release an app" to make money as a programmer. I just need to work on small parts of an app that already exists and I can have a comfortable salary with reasonably good job security.
I don't know if it's reasonable to compare apps--which often address economic problems--with music. Probably a more apt comparison would be between music and entertainment apps (games?).
I highly recommend The Song Machine (Inside the Hit Factory) by John Seabrook. The author describes the modern reincarnation of the Brill Building that pumps out pop hits using methods that we would not call songwriting (Max Martin, Dr. Luke, etc.). The goal is to get you to listen to 30 seconds of a song (that counts as a spin on digital streaming services). They've decomposed the process into mass production of backing tracks (using pro tools) which then get handed to so-called topliners that add the "hook" (the catchy part of the song). They stick the famous artist at the beginning of the song to get you past the 30 second threshold.
In a similar way, back in '88 KLF wrote tongue-in-cheek "manual" for writing a top pop song, a hilarious and cynical view into how a musical sausage is made:
This book is a fun read and goes a long way toward explaining why so much of today’s music is crap. It’s the difference between mass production and craftsmanship - and mass produced music just isn’t very good. Not to me anyway.
I think what the CEO is saying, and it has been an increasing trend in music for a long time now, is moving even further away from large LP releases every few years. The existence of that medium of music doesn't make sense in the current year given how people actually go about consuming music.
The way to keep fans engaged involves releasing EPs and Single on a regular basis, releasing a whole album only really makes sense in the context of trying to produce it for artistic value, not fan engagement value.
LP's or long form collections by artists capture a moment in that artist's creative point. Typically, the cycle was release a record, tour a bunch, write new music, and repeat. This meant by the time that next LP came along, the collection of songs on there could sound fairly different as the artist changed, got better, heard new songs and got inspired etc.
Then there's the recording and engineering challenges around mixing and mastering an album. Albums need to sound cohesive even if the songs aren't necessarily the same genre even. The best albums are those ones that flow between softer or more upbeat music but still feel tied together as a whole.
Forcing artists to pump out strings of singles and EPs diminishes music as an artform. It takes away the format that's allowed some of the best modern music to be created that likely would never have been created if artists were just constantly pumping out individual disconnected songs.
Even today, you can still find some pretty amazing albums that are being made where each song individually would stand as less, but together as an album they come together to make some great art.
I agree with what you said about albums as a cohesive piece of art, I will miss that for sure.
> Forcing artists to pump out strings of singles and EPs
But who is forcing the artists? The market? It's clear people don't want to listen to LPs released every 3-4 years, they want singles and EPs released frequently. If it is indeed a market effect, then artists need to evolve or perish. If you want to make money selling art, you need to make it in a way that people want to buy it. It's that simple.
One of the things that happens when consumers have options they didn't before is previously invisible preferences become expressible. What if the bulk of consumers never cared much about albums as coherent pieces of art? What if they only did so financially because it was the only way to get the separable pieces they did care about?
The advent of fast food has not exterminated fine dining. What's changed is that consumers have more choices. This did not signal the death of fine dining.
Can I ask how it's clear that people don't want to listen to LPs? I think there's a percentage of people that will listen to what's popular in their playlists but there are also plenty of people who listen to LPs and are fine waiting 3-4 years. I believe more records were sold last year than in history.
I completely agree with you from the perspective of being a fan of music for music's sake. Listening to a well put together album is fantastic, but that isn't the be all end all of music.
Music in the current age is equal parts pop culture as it is listening aesthetics. For an artist to thrive financially they have long had to take the first path rather than primarily focus on the later.
The artists all producing the top songs on spotify are doing so through the release of singles here and there and it makes total sense given the cultural context for how music is consumed and discovered in the current year.
You just need to be creative with the release cycle, there is nothing stopping capturing ah moment in time and then releasing a serial, indeed it may allow for a more iterative process wrt to mix and mastering
These full-length albums have long turned into the exception though, which is kind of sad. Zeppelin IV, Sgt. Pepper or Dark Side of the Moon would never become hits today under these artistic limitations.
Only with certain kinds of music. Dance and club music has always kind of been like that. The whole culture of that music is around singles and EPs. I think the thing is, that kind of music is the 'popular money making' music these days.
But even amongst that music, you can still find some albums, being produced as such, with the intention of it being an album. It's harder to find and I doubt it makes money through services like Spotify, but on places like bandcamp, or independent label storefronts you can find plenty of albums in a wide variety of genres.
The best thing we can do is support those artists and keep buying that music if that's what we want. As long as there's a market for it, it can keep existing despite Spotify.
> The existence of that medium of music doesn't make sense in the current year given how people actually go about consuming music.
I'm not convinced. The standard 3-4 minute pop song came about because that's what could fit on one side of a 78. Albums came about because an LP (33 1/2) could fit roughly 40 minutes of music, or 10-12 of those songs. These formats proved to be a popular and effective even though the technologies that created them are long obsolete.
Interesting to note that the play duration of an LP was not seemingly the result of intentional design decisions around sound quality, speed, physical size, ease of manufacture, etc, but rather a bit of a historical accident arising from technology changes, and marketing missteps:
"When initially introduced, 12-inch LPs played for a maximum of about 23 minutes per side, 10-inchers for around 15... Economics and tastes initially determined which kind of music was available on each format. Recording company executives believed upscale classical music fans would be eager to hear a Beethoven symphony or a Mozart concerto without having to flip over multiple, four-minute-per-side 78s, and that pop music fans, who were used to listening to one song at a time, would find the shorter time of the 10-inch LP sufficient. As a result, the 12-inch format was reserved solely for higher-priced classical recordings and Broadway shows. Popular music continued to appear only on 10-inch records.
Their beliefs were wrong. By the mid-1950s, the 10-inch LP, like its similarly sized 78 rpm cousin, would lose the format war and be discontinued."
> A frame of a movie won’t make sense without the whole movie.
Why not? It means exactly what's in the frame and what you can derive from it alone. The movie in its entirety may have a more profound and precise meaning, sometimes entirely different from what disjointed frames will tell you, which is the point of my analogy.
Perhaps songs-scenes make a more fair analogy. Some albums are basically collections of singles because the collection itself doesn't represent any meaningful overarching theme or creative process. I feel the same about movies. I'd bore myself to death watching another Marvel movie, but I'll happily enjoy some disjointed highlights from them on youtube.
Releasing songs individually is a serial process. Releasing an album puts all of the songs published into the same context, and everyone’s first experience of all songs occur at the same time.
A band that makes a successful album produces something greater than the sum of the individual songs inside the album
I meant that there’s no technical reason a musician can’t release multiple songs at one time and call it an album, meaning as long as there is demand for an album, they should exist.
At this time, I would argue that the main reason that the album continues to exist is because it's woven into the fabric of the music industry. As long as there are Billboard album charts and Grammy's are awarded for best album, there will be albums. For many modern artists, they'd be just as happy dropping a continuous drip of singles.
This is sad and feels to me like a dumbing down of everything due to short attention spans.
I still prefer to listen to full coherent albums, and one of the highlights of my week is my Sunday run, which gives me time to listen to an album start to finish with no interruptions.
I hope some of my favorite musicians will continue to conceptualize albums that have a 45+ minute arc, even if market pressures mean that they'll initially need to release the tracks one at a time.
Even before streaming music, I don’t think many people listened to whole albums. I distributed a lot of custom CDs to people in high school because they wanted various singles off of different CDs and I had access to a CD burner and could rip and record them in custom mixes.
I personally find it too tedious to even work out the lyrics of the song because most of the time I can’t figure out what the singer is saying, or whatever cryptic message it is. I just listen for the sounds, mostly, and that scratches my itch sufficiently to not make it worth my while to look more into it. But I can understand if some people do have an interest in it.
I listen to whole albums. I still buy CDs primarily because I don't want piecemeal "hits" when there's gold in the material that will never see wide promotion. You'll never hear that without buying an album. Live albums in particular don't work at all in the streaming era where tracks are discrete units that can't flow into each other. When I rip a CD I know with 100% certainty the gaps will be seamless.
There was once a time in popular music when you could actually understand the words coming out of a singer's mouth, as they weren't digitally over-processed, enhanced with autotune, and mixed into a zero-dynamic-range soup of loudness. Maybe I'm just old and out of touch, but I often can't tell the difference between modern pop song singers--many sound more like the output of a DSP than human.
My impression is that there was actually some benefit to having really long streaming albums, at least if you were already a hit artist. You'd rack up more individual plays and thus more money. This only works, though, if people are coming to you and playing the album rather than being sent to a song via an algorithm. If that's the majority of your usage, a constant drip of new singles makes sense.
I absolutely agree, and that's what I think the CEO is referring to. Slowly dropping 6-12 singles over a two year period is more sensible nowadays than taking two years to write, record, and release an album or EP.
In fact, leaving out the few niche bands that release albums as a cohesive whole, the concept of an album seems terribly anachronistic in this new world of streaming. It's now the era of the single song, whether we like it or not. This is, of course, just considering popular genres of music.
How much of that falls to user experience though? We're not constrained by physical media, but aren't we by Spotify design?
I almost exclusively browse my library by album, but Spotify generally pushes me towards tracks. The release radar, for example, offers no way to filter to full album releases.
I hope artists aren't pressured into releasing tracks one off purely due to design decisions vs artistic choice. Software can easily cater to both.
I blame this on iTunes, not Spotify, when I first got into music because I didn’t want to spend all my cash on an entire album for just one or two tracks I really liked. Up until the last 5 or so years I rarely took the time to listen to a album of a song I heard on the radio or in a mix.
This is controversial for some reason, but I don’t think there’s ever been a significant demand or market for music production. The entire notion that musicians get paid to make music is a false premise.
There was a market for records, not for music. What people were actually paying for was the technology to play music, and the distribution of music. When that became digitized, no significant market formed around music production for reasons that seem pretty obvious to me: 1. It is absolutely impossible for the market to ever reach a state where new music is not produced, regardless of the existence or absence of any monetary factor; and 2. I will be just as happy regardless of what is produced because my brain naturally adjusts that emotion to the scale of whatever I am perceiving.
There was a huge market around distribution and licensing and technology related to music, and a lot of that went away. But that doesn’t mean musicians are entitled to a 50-billion-dollar industry around music production that never existed to begin with.
This isn't "controversial", it's just a kneejerk contrarian position from an armchair expert. It's actually hard to understate the demand for music production. You walk through any major retail store, they're playing music that is generally designed (and proven) to make you purchase more. Advertising, movies, basically any form of audio-visual entertainment would be devoid of life without music. And people demand novelty. Just because you don't personally see the value because of your "automatic brain adjustment" problem doesn't mean it isn't there. These of course, are consumerist arguments, let alone the fact that say, personal expression has been a major feature of popular music for over 500 years, and that people from all walks of life have proven time and again that they are willing to pay for the experience of live entertainment, even to change their entire life to accommodate it (think festivalgoers, Deadheads, other diehards, etc.). I mean you may as well be arguing that the economy doesn't exist—like, yeah it sort of doesn't, because currency is all just an illusion we have to agree upon to grease the wheels, but acting like that invalidates the clear societal value of music... is some willful ignorance. The music industry has grown over the past decade. I implore you to do some reading on this, maybe start with Tin Pan Alley.
I think his comment is insightful and thought-provoking. You might be dismissing it too eagerly. To me, the argument primarily revolves around the notion that music's supply is unquenchable regardless of demand since people naturally want to produce music. They will do it even if there is no financial incentive so there will always be music to listen to.
You should be wary of things that are said confidently by someone with no expertise, for example, this is a flat out lie:
> There was a huge market around distribution and licensing and technology related to music, and a lot of that went away
As I pointed out, distribution and licensing has not gone away, that segment of the industry has grown.
Likewise, you have to ignore everything but the direct-to-consumer parts of the market to have that comment begin to make any sense. Video games, commercials and movies are all being produced with increasing efficiency as well, and thus, sync placements and licensing are more important than ever.
I actually think I disagree with your premise here as well, which is still more cogent and direct than GP, who said "the market doesn't exist". Yet, I think demand has kept up with the supply, I mean, I think the demand is almost infinite. Yes, people naturally want to produce music, much in the same way that some people want to make delicious meals for their friends, and when their dinner parties become popular enough to warrant pop-up level sizes, they might consider commercializing and commodifying their product because there isn't an option to continue doing it at a loss for most people. The financial incentive, in most cases is—how do I keep creating a thing people enjoy without dying? Likewise, people might want to eat at the same 5 restaurants their whole life, or hear the same 5 artists, but even that latter group are going to be enthusiastic when one of those 5 artists releases new content. Think about the fact that whoever happens to own The Beatles catalog can release a remastered version of any of their albums and have consumers ready to buy, what is to the layperson, a nearly identical product. Now, remember your favorite meal you've had out? Have you ever tried to recreate that experience?
I guess what I'm getting at is, you can't think of music in purely economic terms, because it tends to behave in unexpected and unpredictable ways that don't align cleanly with the dry notions "supply and demand".
This is absurd. Why are people who insist on couching everything in economic terms refusing to believe that new music is a commodity that people demand? People don't need cars with new features, and yet the market exists for it. Loads of people throw out their wardrobes every 3 months and buy a whole new set. If that's not demand, what is it? If you can only apply "demand" to basic necessities, and literally no luxury goods, then what is the point of the term? That's not to say the commodification of art is a good thing for artists or listeners, in my estimation, in fact I'd say the artificially induced demand of mainstream music is feeding into the current crisis. Marketers push a homogenized sound and then the force of the industry moves to capture that segment of the market. In fact I'd argue there's unmet demand that exists for quality alternative music that won't be satiated by these models.
It's market valuing things based on scarcity. Music is in abundance right now due to lower barrier to entry for distributing and producing it than ever before. You have to provide more value than someone producing music as a hobby for free on youtube.
Huh, this is an interesting take to me. I'd say that my using Spotify has definitely increased the amount of "bad" music I listen to, but it has _drastically_ increased the amount of amazing, life-changing music I find. There are _so_ many artists I enjoy now because Spotify threw them in my Discover Weekly. By increasing the sheer amount of music I listen to, thanks to Spotify, I've increased the number of bands I love that I stumble upon, and bands I hate. It's a net win though, because I forget the shit stuff, but really, really cherish the gold I find.
I'm just surprised that you feel it's _harder_ to find good music because of Spotify.
I really miss the days when Youtube's "related videos" were seemingly based on a graph data model -- populated like: other people that watched this, then clicked on that.
Honestly, it was far more accurate for new, niche music. My browser tabs would grow exponentially.
Now it's all sorts of weird shit like US news coverage of sentencing serial killers. Which I didn't ever watch before...
I disagree.
Music is subjective. In fact, the lower barrier to entry and discovery has enabled me to find my favorite music that 5 years ago I had no idea even existed. You would probably hate the music that changed my life, and I would probably not enjoy yours.
I do agree that bands are getting screwed though. That's why I use Spotify for discovery and buy lots of merch when a band really lands with me (and go to live shows - pre COIVD of course).
I don’t buy that. Much of my favorite music is video game fan-made music; and, while that is hard to make and their skill is great, there’s a lot of that music.
I love music (and am a musician), but I disagree that good music is scarce. Part of the issue here (among many others) is that music is fairly "evergreen", and certainly much more so than apps. Zillions of people are still listening to the Beatles and Bob Dylan (two examples of "good music") more than half a century later. I worry that the ever increasing catalog of recorded music is making it ever harder to gain mindshare as a musician.
On the flip side SoundCloud, Spotify, Apple Music and websites like http://everynoise.com have made it even easier to find more obscure music, of which some is electrifying, change-your-life music.
It doesn't follow that "more people making music" means "same amount of good music"
Music is life-changing when you hear something that your mind cannot conceive at first, but finds particularly striking. The memory of hearing that song gets embedded in your memory. It's likely that you'll remember where you were when you first heard the song. It becomes part of your history.
Functional music, like what you describe, is easy to listen to. It serves a purpose: reduce boredom, be a companion in sadness, lift the awkwardness of sitting next to strangers in silence. It fits in the genres you are used to listening.
This is, of course, just my experience. But it's clear that for many people, music becomes their life.
The kind of music that I paid money for in the past hasn't been released much recently. So in my opinion, there's a lot of potential for growth by making good music.
> It's orders of magnitude easier to make working software than it is to make a hit record.
I think the same could be said of any (good) art. Art and engineering are fundamentally very different things that just can't be compared, even if there may be some mutual overlapping. (Audio/mixing/mastering engineering definitely crosses the streams more, but for the sake of argument let's just consider the entire end-to-end process of making an album, which I think is almost entirely a matter of art.)
There is some degree of art involved in developing software, but if it's just getting some business app working for BigCorp, it really can't be compared to writing amazing fiction, painting a beautiful painting, directing a fantastic film, making a great album, or any other artistic endeavor. Anyone who scoffs at artists because they're not doing [technical thing] just shouldn't be paid any heed.
> It's entirely possible for somebody relatively lacking in technical ability and/or any great insight into their craft to produce something wonderful.
100% that. Thom Yorke (Radiohead singer) has composed dozens of godly songs without even being able to read the notes (in fact, his band members are calling him the idiot savant and are asking him NOT to learn the notes at this point, as it might spoil the magic). Tricky composed some good songs without being able to read notes OR play any instrument.
Most pop/rock musicians historically have played (and composed) by ear without notation or formal music theory.
It's only in the past 20 years or so that large numbers of pop/rock musicians went to music school.
Even so, virtually all pop/rock is composed by ear today, even for music school graduates. Typically composers start by humming or singing a melody, "transcribe" that to an instrument by ear, and then embellish that further if they know music theory.
You can see from above why most pop music consists of the same 4 chords, with the addition of audience/gatekeeper familiarity (don't mess with a formula that works.)
Yeah the CEO’s take reminds me of Jobs’s “you’re holding it wrong”. So it’s not that Spotify is not paying artists enough, it’s that artists aren’t working hard enough. And even though artists have been complaining about the money they make from streaming for years, “in private” some of them are happy. Sure.
I don't think he was saying that artists aren't working hard enough, I think he was saying:
1. Albums don't maximize artist income - musicians should be steadily releasing new tracks. (Ironically, this is how things used to be up until the 70's, when a new technology, the LP, changed the dominant form of artistic expression.)
2. Nobody reads rolling stone anymore, mtv doesn't play music anymore, if you want a successful career you have to actively market yourself.
Even so, he's wrong according to every thing I've read on the subject.
The conventional wisdom seems to be that if you're a musician the primary way you're going to be making money is through touring and selling merch at shows. Not by cutting more tracks for spotify to make a buck off of.
Of course. If people are going to make poor analogies and mock a quote that has nothing to do with the topic at hand, at least get the quote right. It's not asking too much, is it?
The thing is: there is more talent available than there is available attention, and especially _shared attention_, to go around. Success means outcompeting other musicians for a finite amount of attention. It's a gamble - and there's a good chance you won't get the attention. Even if you did the last time.
Nobody is saying that it isn't hard work to create a record. That is a straw man.
But someone did get up on a stage and point out that monetization of your work will require a strategy that is adapted to reality. If you want financial success you have to do different things than before. You can get angry at it and wish this was 1983 again - or you can get cracking.
Well, you can't really blame the tech CEO here... Spotify charges people what they're willing to pay to listen to music. I think it's a minor miracle that they've worked out a way to get people to pay anything for music considering how easy music piracy has been since the internet came along. (Not taking a moral stance, just stating a fact).
Speaking as another sound engineer and software engineer:
I believe the effort and risks involved with piracy are -- on average in well-off places -- simply "worth it" compared to owning physical media and "not worth it" compared to paid streaming subscriptions.
It's not just about dollars, but the value proposition of the total experience. The average pirate doesn't streamline their rig with all sorts of automation to pull things into a slick xbmc derivative heuristically; rather it's a time sink and less on-demand.
On the other hand, the more restrictive back catalog does keep piracy relevant for those who want less common selections.
The arbitary nature of contracts also keeps piracy relevant for people who get exasperated at the "now you hear it, now you don't" problem with playlists.
If you pirate something and make a playlist, it won't be full of holes after the next round of contract negotiations. With Spotify -it very well might.
For me the time sink for finding new music is enjoyable. Listening to something that I've put a minimal amount of effort on acquiring makes the experience worth more. That's in comparison to letting algorithms play one song after another( which most likely will either repeat the same songs or just play top hits which are boring to me).
Absolutely agree. I've been playing music for basically my whole life, and am now, finally working on producing my first record of solo singer-songwriter material.
Every time I thought I've been "close" over the past 18 months it turned out I still had so, so far left to go. My "scope" is creeping, sure, but it turns out there's so, so, so much work in-between "recording & mixing one voice & one guitar" and "having a really well arranged song with just the right musical details" etc.
I've done a bunch of big, original projects in tech over the years, but I've never worked as hard at anything in my life as I have on this music.
And I'm still barely scratching the surface of where I want my music to go long term, and I know that the closer I get to recording the music that's in my head, the more ambitious my future records will be.
Demanding a "deeper, more consistent, and prolonged commitment than in the past" from people whose profession you've already made more difficult is pretty galling.
It may be the case that their careers were going to get more difficult anyway; nevertheless, Ek is in fact one of the main drivers and beneficiaries.
While I don't deny that the Spotify CEO is the best person to send a message like that, I have to ask myself what life was like for musicians before Spotify (or any other low-paying streaming service).
30 Seconds to Mars made a documentary about the album they recorded in 2008 and the lawsuit they had to deal with at their record label (which had been recently purchased by a larger company that had zero music / talent experience). Their contracts seem to put them almost $1 million in debt to the record label for each album, despite the fact that they sell (err sold?) multi millions and they do worldwide tours.
Sure, Spotify's CEO is a strange vessel to make more demands on musical talent, but he's also in a position to know how his customers want to consume their music and what gets them engaged in his product. I would take it for what it is and no more.
Music is a winner-take-all markets. It makes it really hard for new musicians to break in. It also tend to be a sticky market, so those that have several hits remain relevant for a long time. Winner-take-all markets are reinforced by feedback loops.
I'm sorry that people dismiss the work it takes to make good art, that must be frustrating.
It's silly to compare the ease of making "working software" with a "hit record". Just like probably most of the music you worked on as a sound engineer, most "working software" made little money and got few customers, look at the long tail of apps that aren't hits in the iOS App Store, Google Play Store, etc.
Even comparing "financially successful software" to "financially successful music" makes no sense. Far more musicians have made six figures or more from a song that took tens of hours to make, than indie software developers have made six figures from an app that they only spent tens of hours on. Is that because music is easier? Yet far fewer technically competent musicians are able to make a living using those skills than technically competent software developers. Is that because music is harder?
It's almost like two things can require different talents, have different factors that go into their financial success, and still both be hard.
I think what he is really saying is : if you want to make money, you have to be on the clock, produce every month,week, day. Welcome to the newest rat race addition (which works wonders for creativity /s )
Just as we are pivoting away from unsustainable energy, we need to pivot away from unsustainable value structures. We have the tech, we have the brainpower, now we need the will to work on that.
> we need to pivot away from unsustainable value structures
I think it's a category error to expect the CEO of the music distributor to be the one that stands up for the sustainability of music artists. Even if you did get a "benevolent dictator" figure in that position, it would likely crumble to a competitor who is focusing more on optimizing the value+experience for the listener.
Probably, yes. It’s more of a political proposition and societal debate. How much do we value culture? I do agree he is probably on some expectancy rail that he cannot or doesn’t want to forgo.
Would you be more open and accepting to commentary if there was more kindness, more love, more empathy and compassion in discussions of how the old normal is gone? Expressions of admiration and awe coupled with opining that the old multi-year release cycle doesn't work anymore, and legions of hard-working humans making sacrifices for what they love may need to change?
Thank you for sharing. The world is always a better place with more human warmth.
We live in a world of plenty. We should be working less, not more. We have created more music in this century than all of the last combined. We have the ability to listen to thousands of years of music at the touch of a button. A concert that would have taken hours of work for 50 people to entertain 200 now takes nothing - a million people put on their headphones and listen to a recording made 20 years ago.
And that is just music - the same plenitude is available in almost all aspects of our lives. And yet, people insist on ever more growth, on ever more work.
Hopefully in the next decades, as the reality of global warming and worker empowerment gain traction, we'll realize that in fact the work week can be drastically reduced without affecting our material needs, as long we let go of this level of consumption.
> the same plenitude is available in almost all aspects of our lives.
To be fair, there's orders of magnitude less plenitude in certain necessities like food and housing - still more than enough over all, but not by so large a margin that you don't get local shortfalls (especially in housing).
It's there, it's just not evenly distributed- or evenly zoned. As a species, we can probably produce enough to cover everyone's basic needs if we wanted to.
How do you think the world of plenty came about in the first place? A hell of a lot of growth and work to provide for the demands of broad consumer bases instead of the whims of some head honcho who wanted pointless setpiece folly castles or grand monuments.
The world of plenty mostly came about from technological advancement.
In my opinion, we're way past the point where we should have stopped trying to grow, in most fields. In particular in music, we're at least 2 decades past the point where the music production industry is being helpful.
Also in art in general, it's interesting to think about what this world of plenty has cost us. Even if we have incomprehensibly more music available to us than any generation before say 1920, most people have actually lost the pleasure of producing their own music.
In the times before recording and playing back music were common, a much larger proportion of the population would be learning how to sing and play themselves. Many more people would be composing and sharing their own stories - the almost dead world of folklore.
This can also apply to a lot of crafts - we're more and more becoming dependent on industry for simple things that we would have done ourselves in centuries past.
If all of this had been a way to help free us pursue other passions, it would have been good. But it seems to me that it's mostly a way to free us to do more work for other people, and that, for the average person who is not working in a job capable of using their creativity, it has started stunting a lot their creative potential, which would have found an outlet in some of these domestic pursuits in the past.
I am curious what you think the optimal solution for music in today’s world with massive storage and high bandwidth. Spotify is paying 70% of revenue to artists.
You have factor in piracy if the price is to high or the solution is inconvenient. The public has rejected DRM and songs are easily give away to their friends. Lastly, No one wants physical media.
Agreed, it's also not what an artist gets paid, it's extremely hard (read impossible) to get a direct relationship with Spotify. You require distribution.. that alone takes anything from a flat delivery fee per album per year (the worst prey on independent artists who will rarely ever recoup that per release from spotify, apple and Amazon combined) to 25% of that revenue and a dozen or so other business models inbetween, then if they have a label, the label gets 50%, publisher is taking 25% of the mechanicals or worse you aren't registered at a PRO and that money gets left on the table. Those ~£0.005 per stream (US, UK rates) aren't looking great for most people but that's the reality of what your getting.
So after you've pumped out 6 singles a year, spent £100 a pop promoting those said singles, £50 mastering each single.. how much do these people think 98% of artists are making these days and can be sure of what they may take home every 2 months from these singles. At least with an LP you have a fixed cost and a revenue projection that can be worked out by price per copy.
No it's not their freedom to be on Spotify. That's not how the music business works. The record labels own the master recordings. Spotify negotiates rights to stream the records labels catalog with the labels. The artists have no say. The exception to this are artists who have negotiated to retain ownership of their master(very rare) or they have acquired their masters back through a contractual clause called a reversion(also rare.) Taylor Swift, Prince and Metalica are examples of people who own their master recording and can dictate whether streaming services can license their catalogs and at what cost.
We all have the freedom to do whatever; the issue is the consequences.
The question I think is more about if as a society we think it’s important for musicians to be able to make a decent living at their work and if so do we want to use our collective tools of law or government (or if someone gets lucky and innovates a better business model) to help that happen?
And if Spotify’s current business model is good or bad for our culture, if we believe that having musicians being able to make a decent living is important for our culture.
> it’s important for musicians to be able to make a decent living at their work
I think this depends on a great many factors. Do you mean possible for some musicians, everyone who wants to do music in some capacity, or somewhere in between? How do you define a decent living?
I'm reminded of the "No farms, no food" bumper-stickers I would sometimes see. While obviously true as written, it's subtlety different in practice and politics. Food comes from farms. Farms are needed. Yet this may not the same as all farms being needed, important, or significant to keep functioning. The person with that bumper sticker may not agree with the distinction I've drawn.
Music is absolutely critical to our ongoing cultural life. No musicians, no music. Yet... to what extent should a society with limited resources devote them to the promotion and enablement of musicians, bearing in mind that there are other uses for those resources? Even with an abundance mindset and in today's world of plenty, this key question does not go away.
For sure. I agree with you. And I'd say maybe music education - kids learning how to play an instrument, etc - is one area of focus. However I don't think it's supporting the arts that's sticking point when it comes to America managing it's limited resources ; )
We, as a society, have already decided that musicians have to be "all in" and "exceptional" to make a good living... and if the rest want to make a living - then private events is the way.
We cannot support a million Beyonce's, a million Gaga's and millions of other musicians - we don't live in a Communist utopia.
Now as for Spotify - they are not a monopoly, in US they are neck and neck with Apple Music. So... Why should we intervene with heavy handed laws* - when there's still a fierce battle happening in the market?
Also - Let's not have the government decide how culture should progress.
* - laws are always conservative and change slowly
How does that work for a streaming service that charges a flat rate? Do you think public has an appetite for a pay per usage model/micro transaction model or would it drive users back to piracy? Micro transactions in online news media hasn’t managed to get any traction.
It doesn't work for a streaming service which charges a flat rate - which is the point.
Spotify is not the same as Netflix etc, because Netflix etc is commissioning and promoting new work - something Spotify has no interest in.
So... you sell units, not streams. You buy a perpetual license to play a unit as many times as you want. The artist gets a royalty for the unit sale. This frontloads income around unit release, which encourages new creation, but units generate a perpetual royalty stream as part of the artist's back catalog, so artists are less likely to starve.
Preview streams are available for free as tasters. Or perhaps a full taster can be played X times before it has to be bought.
The player is tied to an app, but most music is now consumed on phones or in a web player, so DRM is irrelevant and there is no real loss of convenience.
This model worked (more or less) for decades.
The point is popular creators should be paid for full-time work. That's how you get the best work from artists.
And it's not as if the economy can't support this. Operating in any other way is entirely down to politics, not economics.
iTunes offers purchasable music. And it could be argued a better deal for consumers who only listen to 100 songs or less a year. Market doesn’t want it. It is inconvenient to force people to go through a checkout flow for every song they want to listen to.
People always forget that storage and distribution of music is a relatively new concept and a new revenue channel for musicians. When did records become main stream? And how old is music? Technology gave artists a means to sell their work besides live performances. Technology also makes the market more efficient which leads to the costs being drive down. Physical media has been replaced with digital media. Physical stores replaced with digital marketplaces.
Without Spotify or other streaming services , piracy would be even more rampant with the ability of people to transfer 1000s of songs between other people in a matter of seconds. Only way to fight it is to lock down the devices which public doesn’t want and would require the government intervention.
> It requires more effort than you could ever imagine.
Unfortunately that applies to pretty much everything. Of all the things in my life, I only even know how to provide the raw materials for some of them. And I could probably replicate a lot of the software given time.
Modern life is a series of miracles and all of it requires more effort than people imagine.
> It's orders of magnitude easier to make working software than it is to make a hit record.
It is order of magnitudes more difficult being in the top 0.1% in Y industry. Then, being in the bottom 99.9% in Z industry. Giving that roughly the same number of people work in each industry.
> I got into software to pay the bills, but the way geeks talk about the industry and people I love drives me right up the wall. Have some appreciation--even awe--for the art you enjoy.
I probably agree with you about your broader sentiment of the commodification of music for the principal benefit of tech investors is not in the interest of society as a whole; however, I don't think your bit about software/geeks served to support your point. Some people appreciate software the way you appreciate music, and perhaps they appreciate music as a sort of commodity. People have different preferences, but sometimes that creates an economic environment that isn't optimal for your own preferences.
Couldn't agree more. Of course Spotify would say that. The reality is much more complicated.
One idea:musicians could organize around federated servers, operated at cost-plus by tech helpers. One avenue to liberate themselves from the mass-market vultures.
For a recent, nuanced view of what's happened by music historian Alan Cross, try:'Ongoing History of New Music' podcast 'Trying to be a Superstar in the 21st Century'.
>>> It's orders of magnitude easier to make working software than it is to make a hit record.
That couldn't be further from the truth, it's extremely easy to make a hit, large labels consistently release hits again and again.
It's all about distribution and marketing network, not about the music.
Sames goes with tech, good luck trying to make a product and garner any amount of traction, whereas the same product released from microsoft/google/aws would get millions of eyeballs right away.
Thinking that good has much to do with success is a recipe to be disappointed. Go spend some time promoting your creations instead of trying to make them better.
For electronic music producers the barrier is definitely lower and releasing one EP every other month is not uncommon. I listen to an artist that has released hundreds of songs this year alone (God's Warrior, check his Spotify or SoundCloud if you don't believe me). Every song is of incredibly high standard as well.
> but the way geeks talk about the industry and people I love drives me right up the wall. Have some appreciation--even awe--for the art you enjoy.
It's the same way with software. People will just vomit up anything up there and truely good, and tested, libraries and frameworks are hard to come by.
The Beatles at the height of their fame put out two albums a year, while making a bunch of movies on top of it. That pace is inconceivable in 2020, but I've never understood why.
It's inconceivable because the Beatles were hugely talented with two and a half very prolific song writers all playing off each other, they had enough money to stop touring and to live in the music studio, and they hired outside talent considered the "fifth Beatle". Not to mention that experiment along with the pace did not last.
Source? I think you'd also need to account for how many big acts had song writing factories working for them back then. Lou Reed for example worked as an in house song writer for Pickwick Records before he went on to form the Velvet Underground.
The way people consume music has changed and for the right reason.
Music now is a commodity that can be obtained in secs. This is totally different the old days when people buy tapes and CDs and use dedicated equipments just to listen music. If you are buying a CD for 15+ dollars, ofc it should contain more than a handful song.
Now you can literally release any number of songs at any time, and test the water as frequently as you want.
The market does not necessarily optimize for hardest sacrifice or how much energy it takes to make something. It optimizes utility and in some cases scarcity. In other words, there are many situations where how hard something is to make is not that important, what is important is the utility of the final product.
tl;dr: Spotify performance royalties paid by PROs used to be fairer than other revenue streams, until independent labels + artists got screwed over by the major labels.
------
Story time.
The streaming point values PRS for Music assigned to Spotify used to be higher per play value than many radio stations in the UK. Someone I worked with actually analysed all this and we were really surprised to discover it. Low Performing royalty revenues were not Spotify isn't paying money fairly to artists, rather Spotify knew exactly how many people streamed any given track and the royalty payout model paid based on % of total streams.
So it was actually a fairer model -- if you assume number of streams is an indicator of performance copyright value.
Compared this to the Radio model. A typical track played on BBC Radio could earn something like £80 for one play (depending on the time of day and the station). But that was because radio audience figures were estimated from sample data for the entire nation. And those samples changed something like every 1-2 years. 10 people in the country might actually be listening to the station at the time -- but you'd still get £80 regardless.
Of course the actual total license value then becomes the big point of issue -- was Spotify being charged enough...?
EMI, Sony, Warner et al. all thought no. They went and cut their own deals with Spotify. Not only was their music being treated as equal to everyone else (they didn't like that), they thought they could get a larger license revenue by dealing with Spotify directly.
So the we ended up in this really weird position: PRS for Music wasn't licensing the majority of UK royalty revenue from Spotify (the majors were) but PRS did all the processing for them (eventually GEMA took over). Leaving PRS for Music with the same workload and less bargaining power during license negotiations.
I bet you all £10 in my wallet that the artists Ek referred to as "happy" were on those major labels' rosters.
------
Source: I used to work for PRS doing usage level data analysis.
Note 1: This was correct as of 5 years ago, certain things may have changed recently.
Note 2: I'm only talking about the Performing copyright. I stayed the hell away from the mess that was Mechanical rights as much as possible.
PRO: Performing Rights Organsiation
streaming point values: 1pt = 1 stream for Spotify. 1pt = 1 second of music on Radio.
royalty payout model: total points in a quarter * point value = total available license revenue for the quarter
> Have some appreciation--even awe--for the art you enjoy.
Haha, no.
Artists aren’t some gift from God here to bestow the world with joy from the heavens that we should all marvel at.
They are people who chose to learn a skill capable of outputting something suitable for consumption as entertainment.
As such, it is their job to be putting out these works regularly, just like everyone else.
Appreciation is nice, but doesn’t matter. Everyday people do mundane things that you have no idea how to do and probably go unappreciated, yet they are more necessary than any sort of “art”. Take art down from the pedestal please.
>As such, it is their job to be putting out these works regularly, just like everyone else.
This framing of human creativity purely in terms of "outputting", "consumption", and mere "entertainment" ignores possibilities outside the sphere of the market, and outside production-for-sale. Art may well suffer from the compulsion to "put it out regularly". I think the fact that art lived through patronage for much of human history is the only reason we have so much good art today.
you're conflating technical ability with the rest of the creative process: intuiting narratives in the collective unconscious & channeling inspiration from a much, much greater source to make something wholly outside of the artist.
that's not to say that artists are "geniuses" who should be put on pedestals, but it's a fundamentally different thing than a vocation
I understand you believe that, but art is literally that which flows from a creative expression. Creative action or expression is the definition, the etymology, and the realization of art.
Artistry is that which finds expression in the creative act of expressing art.
That is not the same as all non-functional human expression.
As a visual artist who regularly works on multi-year projects, and happily grabs a new album from bands she's been following for years who only put new work out once in a blue moon, I would like to cordially invite Mr. Ek to fuck himself.
While I am handing out invitations to auto-copulation, I would also like to hand out a large number to every senator and representative who has voted to defund the arts. Making room for people who have learnt creative skills to pursue their craft without having to constantly worry about whether or not this week's piece is gonna make enough money for them to pay their rent is an important part of society.
Finally, I should express my absolute delight that Patreon continues to exist, and that I have enough people willing to spare a few bucks per month that I can pay my rent while continuing to work on my long, slow projects. I keep on worrying that Patreon will fall apart when it becomes clear that their burn rate is higher than their profits, but I will enjoy it while I still can. It's about the closest thing to the NEA I'm likely to experience in my adult life.
So you're engaged with your audience enough for them to sponsor you on Patreon... and you managed to miss that you're literally doing what Ek talks about? You have auto-copulated right in front of everyone...
I am, and I am not, there's a huge difference between "I am releasing one page of my long graphic novel on a regular basis" and "I am regularly producing stuff designed to stand alone as complete objects" IMHO.
That is not the only thing that he said, it was the opposite to make his previous statement more prominent.
Congratulations on being caught by the title and completely missing the whole argument.
Imagine if I came out and said: "I'm an atheist and I don't need God to tell me to treat people well, love freedom, help the needy and be a nice person... no longer can I justify killing, pillaging and raping using the word of God."
And you put all focus on "I justify killing, pillaging and raping using the word of God"
"The artists today that are making it realize that it’s about creating a continuous engagement with their fans. It is about putting the work in, about the storytelling around the album, and about keeping a continuous dialogue with your fans."
Translation: We don't have a marketing department to handle this royal pain-in-the-ass for you, like a record company would've in the old days. Selling stuff still requires marketing though, so the people who do well are the ones who handle that part themselves, along with making the music. If you're not good at wearing all the hats at once (for example you're amazing at music and not particularly at marketing) well I guess you don't belong in the new future landscape.
Corollary: This future landscape tends naturally toward being filled with mediocre musicians who work at it part-time - almost like a moonlighting side gig to their marketing jobs.
> I guess you don't belong in the new future landscape
There's really nothing unusual in bands losing fans and appeal. If you don't want to change and loose fans - that's your choice. This isn't even new. Many artists had to recreate themselves to stay relevant or just retire.
Now - you have much cheaper and much more direct means of engaging with your audience. If you decide to ignore this - then you probably will not make a splash, but there's always a need for wedding singers and backup vocals...
> There's really nothing unusual in bands losing fans and appeal. If you don't want to change and loose fans - that's your choice. This isn't even new. Many artists had to recreate themselves to stay relevant or just retire.
Pretty hard to lose fans you don't have...and it's really REALLY hard to get fans
Ugh, I very much do not want musicians feeling forced to churn out more content on a set schedule. I don't get how anyone could look at the current music landscape and think that what we need is more quantity, at the expense of quality. Pushing for more new content is going to encourage more derivative low effort songs, and less original ideas.
> Ek claimed that a "narrative fallacy" had been created and caused music fans to believe that Spotify doesn't pay musicians enough for streams of their music.
Fallacy? Scores of well-known musicians, from Bette Midler to Taylor Swift, have talked about how little they make through Spotify and other streaming platforms. IIRC Swift only joined the platform after cutting a special deal. Not everyone can do this. Quoting the guitarist for Mastodon, which has a much smaller following (https://www.digitalmusicnews.com/2018/07/05/mastodon-guitari...):
I could live a thousand years, and Spotify plays [our music] all day long, and maybe I’ll just make a couple of thousand dollars.
Ek also said this:
> "Some artists that used to do well in the past may not do well in this future landscape," Ek said, "where you can’t record music once every three to four years and think that’s going to be enough.”
This quote reminded me of Joni Mitchell, who would take years-long breaks earlier in her career to escape from the business, work on songwriting, and recharge. Many other musicians can't, as Ek puts it, "create a continuous engagement with their fans" for reasons related to privacy, family life, finances, or mental health.
It's bad enough Spotify pays artists peanuts. But if Ek is tuning his platform to benefit only those artists who are willing to jump through Spotify's algorithmically generated hoops, and sideline everyone else, then the future of the music industry looks very dim.
How little they make, compared to what? Streaming has basically saved the music industry, finally getting revenues to go up after about 15 years of decline. Spotify pays out 70% of its revenue to labels and artists. I wonder what percentage would satisfy people.
I'm not sure how Spotify is to blame for bad label deals, which are nothing new. Artists also have lots of ways to get onto Spotify and other services without a label if they wish.
Bandcamp takes only a 15% cut, which is better than Spotify's 30%, but the way people talk about it you would think there's a 10x difference or something.
> Bandcamp takes only a 15% cut, which is better than Spotify's 30%, but the way people talk about it you would think there's a 10x difference or something.
It's not about the relative cut of the services, but the absolute value of the payouts. On one platform you earn fractions of a cent per stream while the other allows you to collect whatever price you set for your release directly. I'm sure the latter approach is significantly better for smaller, niche artists, if not all but the most popular ones.
Or put differently: The number of plays on Spotify required to match the payout of a handful of album sales on Bandcamp is probably out of reach for most artists.
It's difficult to square that idea with the fact that streaming has caused both industry revenue to increase as well as independent share of the overall pie to increase.
The problem with Spotify is that they pool all payments, and then portion it out based on how big of a part of the total amount of plays per month that artist gets.
That means basically all of my $10 subscription would go to huge popular artists, instead of to the artists I actually listen to.
By buying on Bandcamp, I make sure my money goes to the artists I actually listen to.
Yes, most of your $10 goes to popular artists. But people that only listen to popular artists, some of their money goes to the artists you listen to as well. Meanwhile, the fact that it's free to the user to play unknown artists on Spotify lowers the bar to discovery. The net effect is that independent share of total revenue is rising.
She wanted a lot of money, just like in the good old record label days. And BTW - her deal screwed indie bands even more...
> Joni Mitchell
Who TF is this person?
> only those artists who are willing to jump through Spotify's algorithmically generated hoops
Since when is general audience and human attention span a result of Spotify's engineering? Do you think that Taylor Swift doesn't have people earning 6 figures to make sure that she stays relevant?
Let's be honest. The amount Spotify/Pandora/et al pay the majority of musicians per stream is far more than they'd make selling CD's. The overwhelming majority of music is unknown, has few fans, and will never "go platinum" nor sell out stadiums - ie, most musicians are not Taylor Swift.
For all those artists... access to millions of potential listeners from around the world is an amazing advancement.
The number of artists I've discovered, personally, through Pandora's recommendation system is staggering. I'd never know about these artists in a pre-streaming world.
This might be true if Spotify's monetization was distributed fairly, but instead it's basically redistributive to the biggest players. In short, instead of splitting a person's premium payment equally among all the artists they listened to, it's split among all artists on the platform. E.g. if Lady Gaga is 2% of all Spotify plays in a month, they get 2% of the money, even if you personally didn't listen to Lady Gaga at all.
It probably gets complicated w/ ads vs premium subs, but if I was 1 subscriber and there were 100 total subs, and I listened exclusively to 1 artist, but 99 other people listened to a different artist... the artist I listened to gets the same amount if they get 100% of my fee or 1% of the total fees, right?
I guess users with idle subscriptions is another edge case here.
I wonder how that is influenced by demographics, i.e. if the people who are on a family plan or a free plan are more likely to listen to certain artists and less likely to listen to others.
It’s actually even MORE weighted in favor of major label artists like Lady Gaga, thanks to special deals between the big majors and Spotify. My numbers aren’t exact here, but this is the principle of the deal - let’s say Lady Gaga gets 2% of streams and random indie artist gets 2% of streams - because of the deals, Lady Gaga’s music gets paid out 2.5% of the pot and the indie gets 1.5% of it.
If a rando artist suddenly gets as many streams as Gaga, then Spotify might make that deal with the "random indie artist".
But let's be real - majority of people go to Spotify BECAUSE there is Gaga, Beyonce, Taylor Swift, Rihanna, Adelle, The Weekend, Shawn Mendes, etc... If you take those names out of the pot - Spotify turns into crappier Soundcloud.
These indie artists literally cottail on the big names.
Just speaking for my own band: we have had tens of thousands of streams on Spotify, yet we have made over 4x more revenue from music sales on BandCamp than we have on all streaming services combined.
That's great! I wish more bands had this level of success.
BandCamp is more-or-less an "indie" place to find music. People have to be "into" it to go there and find music.
Spotify and Pandora are far lower barriers for every-day people. Just put on some station you've created from your favorite band, and eventually it'll start playing new bands you've never heard and might love. It's amazing.
I read somewhere that the artist behind this discovered that the streaming songs he made the most from were songs that were apparently requested by random small child requests so he created a whole bunch of songs for just that purpose and makes the bulk of his income through that. Discovery on the streaming services is awful and the renumeration for indie artists is also awful. It's the worst of all possible worlds.
I used to be able to do reverse phone lookup by typing a number into google. People have turned that into a business and now it, to put it simply, does not work at all.
This is the music equivalent of that. It's also why there are gajillions of crappy cover songs on the streaming services. It's turned into a money extraction thing and is crowding out the, you know, music.
Once upon a time, not that long ago even, a talented musician could make a living from their music. It's depressing to read biographies of entertainers who came up in the first half of the twentieth century who needed a job so they became entertainers. We've traded our culture for a mess of pottage.
Since the beginning of time, certainly modern times, artists have largely struggled to survive off their artwork. Most don't survive.
There is no time in modern history where artists, of any kind, on average made a living from their art. Even our most famous artists very often die poor.
There has always been a gross abundance of art. It's not a bad thing - but it does mean making a living off it is _hard_. Really _hard_. It's not good enough simply to be talented - there's a lot of luck involved too.
For every Taylor Swift, you'll find dozens of wannabe's with nearly as great, or sometimes as great of songwriting abilities and voice talent. Some even play instruments at the same time! Somehow, Taylor Swift was "picked" and propped up by record labels et al, and now is a household name.
How many artists can you name from the 19th century? Probably a few of the most successful ones. Do you think they were the only ones writing music, or painting? Definitely not.
Hard to say! Spotify definitely isn’t a magic music marketing machine — you need a decent amount of organic traction before you’ll get any algorithmic promotion, and even more for editorial. It took us years of releasing music, playing shows, networking with other bands, pitching to blogs and buying ads before Spotify became an effective way for us to find fans.
It might be that “listen on Spotify before you buy on Bandcamp” is the 2020 version of “download on Bittorrent before you buy the CD”. But it feels like “is this slightly better for musicians than music piracy” is a pretty low bar for a $50 billion company to shoot for.
I don't claim that Spotify is end-all-be-all of music marketing. It's one of the tools, that is part of the arsenal. It feels that middle of the spectrum - above "we have a facebook page" and below "corporate label promotion material".
>is this slightly better for musicians than music piracy
They're a music streaming and recommendation service company, not a "we kill piracy" company. Maybe they should get into more promotional business... but they're not in that business.
I would argue that Spotify is "(insert more successful band in the genre here) is looking for a warm up act for their gig in your town".
If you're in a very middle of the pack mass appeal genre - you're definitely screwed on Spotify, or anywhere, without a breakout act(corporate contract/viral video)
I wouldn't buy "your band's" album, even if I could listen to it right before buying it. I don't have the time to listen to 30 minutes of audio without being hooked onto it first.
One unknown song in my Google Music or Soundcloud auto-playlist may get me hooked... But you need to build a desire to buy the album beforehand.
No way. How can music simultaneously be "unknown, has few fans" but also benefit from "access to millions of potential listeners"? If they get millions of listens they aren't exactly unknown anymore.
One CD sold at a gig for $10 is equal to about 2,000 Spotify plays. A t-shirt sale is double that. A mid-level band can play to 200 people a night and do $1000 daily in merch, easy (and I've personally done it). Sure, if people discover the band via Spotify, that brings them to shows, so there is the discoverability aspect - but the compensation is not remotely comparable.
Spotify profits off of the market value of streaming music having dropped to practically nothing (due to practically infinite supply), and their pay rates are terrible as a result.
You're forgetting all the expense of printing the CD's. It can be thousands of dollars, for a basic bi-fold paper cover. Selling CD's at $10 can, and often is, a loss.
"Mid-level" band playing to 200 people who don't know who they are, and might buy a few T-Shirts isn't making much either. Plus they have to quit their jobs and travel, or settle for local shows once a month (or right now, zero shows for practically the entire year). The bar or "house" might pay the band $250 for the show, split 4 or 5 ways... plus deduct any overpriced alcohol and food the band consumed. They often walk away with barely enough money to put into the gas tank.
Heck, most of the professional, full-time "mid-level" bands can hardly afford their practice space monthly rent.
Very, very few "Mid-level" bands make money. It's a passion project. Very few get lucky enough to make it to the next level and tour with some known bands.
"One CD sold at a gig for $10 is equal to about 2,000 Spotify plays"
Assuming the album is 10 songs, am I then right in thinking that once I've listened to an album over 200 times (2000 individual song-plays) then the artist would have been better off if I'd used spotify rather than buying the cd?
200 plays over a few decades of album ownership sounds like a pretty low threshold for cds to be better for artists than spotify. Perversely, it looks as though the only artists who will have done better from me buying their album are ones I grew tired of quickly (though my reselling the album could mitigate that).
On the other hand I think it's more common to send a friend a link to a music streaming service for them to listen to a single track than it is to send an amazon link for them to order the CD. Streaming lets you capture revenue from people who may end up not liking your song(s), doesn't it?
As a small time musician with about 20 albums in as many years up on the streaming services, I am 100% fine with this trade-off. I get basically nothing, but in return I can share my musically instantly with anyone. It's pretty awesome really. It would suck to have to make a living off music or art.
Honesty has to be matched with clarity of insight to be worth much.
"Access to millions of potential listeners from around the world" was an advance that had arrived with internet itself and monetized with a number of retailed-recording models like iTunes or Amazon.
Or Pandora, if you like, but it's important to note it is a big mistake to think of Spotify and Pandora as the same thing. They're no more in the same category than FM radio is in the same category as a cabinet full of records. Pandora is radio evolved,and it's a discovery and promotional boon, I happily enjoy it and pay for it. Spotify is something else entirely -- it isn't primarily a radio station, it is a cloud replacement for record collections or the practice of owning recordings at all. Even calling it "streaming" almost kneecaps attempts to think about it productively because of how much that term is associated with broadcast.
Spotify's value-ad isn't audience reach. It isn't even algorithmic or social recommendation, both of which were done before it began to gnaw away at the industry. Spotify's contribution to both is marginal.
The big first-order benefit for the consumer is that it replaces retailed-recording revenue with very fractional per-stream revenue, and so it turns out you pay a lot less and the artist gets paid a lot less.
Trying to justify that with "yes, but their potential audience is so much bigger" ignores that artists have to do a LOT of increased volume to make up. 100 plays for every single or album they might have sold is a floor; in a retailed-recording model, someone who nets $.50 a song is already doing better than someone who gets 100 plays on Spotify (also, that's average and there are some rank-takes-all issues with how payments are distributed). And you still have the classic problem artists have always had -- how do you capture fickle and scarce attention?
And if you're relying on attention-volume, then that means niche music isn't going to be as much, so you need to pick a form that has the broadest possible appeal. And like the article says, you need to also focus on volume of material rather than polishing less frequent releases for a long time. And of course, marketing & promotion matters more than ever when a broad audience makes the economics work.
Do we like these incentives? Or do we just like paying less up front and not having to think about them?
As you say, let's be honest and clear-eyed about our answers.
How so? Any person can make a song, and list it on Pandora or Spotify. How many listeners do you think your average garage band has? How many radio spots would they normally get? Albums sold?
Most of us have known or lived with someone in a band at some point in our youth. Some of us still play in bands for fun. How much money did they or do you make from selling CD's today?
It's practically zero, and it's been practically zero the entire time I've been alive. Anecdata, the people in bands I currently know produce and print CD's at a total loss. They literally give them away for free, hoping to get someone to listen to their latest masterpiece.
Pandora and Spotify are a massive boon for these folks.
People also used to steal music too. Having a "free" or minimal-cost streaming option has pretty much removed that desire for most people.
> Revenues to artists from digital distribution is a fraction of what it was during the CD days.
Perhaps for the Taylor Swift's of the world, sure. For everyone else, which makes up 99.99% of musicians, bands and artists... they made zero or negative money before, now they have a chance to get in front of people who would otherwise never know who they are or listen to their music. That's a win for the 99.99%.
So the 'paying' bit happened because of Mp3s, not because of streaming. Artists, were they given the choice, would choose neither, they would chose a kind of DRM.
So yes, you make a good point about the long tail.
However - the barrier was reasonable. It was possible to make an album and get it out without killing yourself - the actual manufacture of a CD/Cassette is not that much.
So that barrier was maybe artificial, but it did definitely make it so that most big music was good. The secondary acts could make demo/mix tapes.
By dropping the barrier you definitely get a bunch of artists we would not otherwise, but the signal to noise ratio is immense. There are 100 waste-of-time-noise artists for every serious one, and it's harder for the good one's to get through.
We see this on the top pop charts. There are still mega acts with decent production qualities like Taylor Swift - where the artistry is getting low (she is no Michael Jackson), and then the rest of it is unlistenable, reductive garbage. There was always quite a lot of that (i.e. a lot of not-very-good music on the charts) - but the signal to noise ratio is a much bigger problem now.
So - the 'decent music' that is not 'really pop' like Taylor Swift never gets through. Decent artists are stuck in a pile of noise and rubble.
In the US, those rates are set by Copyright Royalties Board [0] and then collected by SoundExchange [1]. The current rates for streaming on a subscription service (such as Spotify) are $0.0024 per performance.[2] The rates aren't determined by Spotify, again in the US, so it isn't Spotify's fault if the royalty per stream aren't enough to live on.
Based on the Spotify revenue model my favourite artist might make a few dollars off me over our entire lifetimes. This is not enough to sustain all but the biggest artists. I am happy to contribute more to smaller artists I appreciate. Seeing them on tour, buying CDs, buying merch, patreon type models and even straight up contributions on paypal and venmo. I've been buying a couple hundred bucks worth every bandcamp Friday since the start of covid. It is not just Spotify, none of the streaming fremium music subscription services provide me an effective way to support the artists I like.
Cory Doctorow points the way in Information Doesn’t Want to Be Free talking about future models for artists to make a living. It might superficially seem like the Spotify CEO is saying the same things but this is just cover for the gatekeepers to keep an unconscionable share for themselves.
Yep, he’s telling musicians what he wants them to optimize for his platform, which does not pay viable rates and never will. Which means if you ARE a musician you should run the other direction.
I used to hunt for music and movies. I was an enthusiastic movie and music fan. But these days, I mostly give up. I listen to underground/indie artists and some classical music. I ignore popular music and movies, especially those promoted on big platforms like Spotify.
It's sad that many talented artists are brainwashed by these big platforms. They chase after big money. They don't produce anything original.
I disagree. Musicians can do whatever they want. Not very many musicians spend their time riding a perpetual wave that they have to continually feed. Spotify benefits the most from this churn.
How much money? Enough for personal private jets or enough to make a living? If the bar is to simply pay the rent then I still think that musicians can do what they want.
Really? Do you have any stats to back that up?
I'm not familiar with many musicians paying the rent with their music and it seems to me that you either make it big, or keep your day job.
I don't think this subthread is any use when you're just speaking of "musicians" at large being able to pay their rent.
You can't put Rammstein in the same league as my friend who would like to make a living in music but realizes that it's unlikely he'll ever make pay-the-rent kind of money from that, simply due to the sheer volume of competition and luck involved (while their music is as good as any song I've heard in the genre, it's a combination of reaching the right people and stumbling upon better ideas than your competition, which both have a significant luck factor). The former can do what they want and still get that private jet, the latter has to "spend their time riding a perpetual wave that they have to continually feed" and likely still couldn't pay the rent. That spectrum is way too broad to generalize into "musicians".
Unrelated to the topic at hand, but I've always been curious how much it costs Rammstein to do their shows. Their most recent tour took ~60 hours per stop to set up the stage and equipment brought in by a convoy of 18 wheelers. Then they have a bunch of complicated pyrotechnics, crazy amounts of lights, costumes, props, gigantic screens, and they perform in the largest venues available. The amount of resources and people it requires is staggering to me.
They were the only ones where I was reasonably sure they are popular enough that shows are hard to get tickets for. Other mainstream bands that come to mind, I would have no idea if I know the name because they had 1 popular song five summers ago or if they are really big.
I'm not even that big a fan of Rammstein, just a few of their songs I like :)
This should give some perspective: https://www.vulture.com/2012/09/grizzly-bear-shields.html - you'd think that the members of a band like Grizzly Bear, while not outright rich, should be able to live comfortable without too many financial struggles. You'd be wrong.
And that was 2012 - I'm sure things haven't improved since then.
Without trying to sound too harsh, 50k monthly listeners really isn't all that much. You can argue that spotify's discoverability (or lack thereof) has a large influence on those numbers, but there are no shortage artists with many times that many that aren't all that widely known.
> The artists today that are making it realize that it’s about creating a continuous engagement with their fans. It is about putting the work in, about the storytelling around the album, and about keeping a continuous dialogue with your fans.
Maybe if you want to be mega-successful, but I have to believe there's room for other approaches too. My favorite band that I've discovered since joining Spotify is Darkwater. They have three albums (2007, 2010, and 2019). Right now, this is the only band I would absolutely make sure to see if they tour within a couple 100 miles. I have to believe they're paying the bills without too much trouble.
In the vinyl era, when a single LP was about 40 minutes long, it was pretty normal for artists to release roughly an album a year. (e.g. Led Zeppelin's eight studio albums were released over a span of 10.5 years, but one was a double album and they took a long break when Robert Plant's son died. The first 7 albums were released over 7 years).
It seems like it was only when albums started to be recorded for the longer CD format (75 min or so) that the time between albums seemed to increase, as if artists felt they needed to have enough material to fill a substantial fraction of a CD's capacity before releasing.
Now that we're in the streaming era, there isn't any particular container size to "fill" to make an album, so there isn't any particular reason for bands to wait long periods, except for their own creative cycles. The notion of an "album" isn't really constrained by any physical limits anymore and it's really just "a bunch of songs released together".
I think the Ek's statements just reflect the fact that it's easier to be successful if you're prolific, and the the new model actually pays when people listen (roughly speaking), rather than just when they buy.
The idea behind creating an album, a collection of songs behind a theme or era of your creative life, lets you tell more of a story in a cohesive way. It lets you create more interconnected works, rather than one idea per song.
Also, it lets you take breaks and lets things breathe. It's really hard to take a song from creation to recording to mastering to releasing without giving yourself a break in the meantime.
Also also, batching out writing, recording, and post-prod makes it go much faster as well
In short, it's far far easier to make 12 songs in a year than one song in a month, and you'll get a much better product out of it.
Not all artists look at it from a purely business perspective. Spotify has the power to support artists, to build any feature they want, using data that no distribution platfo of the past has ever had, but instead they're narrowing the path to success; and justifying it with their own profit motive.
I honestly don't think the length of the albums matter. As bands get more successful, they tour more, and that eats up all of their time. Lots of bands will tour for 2 full years after an album release, and only then start to work on another.
Back in the 70's, bands just didn't casually tour the world like now.
Bands toured like crazy in the 70's. Led Zeppelin II was written and recorded on the road. (technically that was the 60s, tho)
I think you have the causality backwards. Because CDs were a longer format, they enabled bands to take longer tours before having to release another album.
To drag that back even further, albums were named that because 78s were generally held in albums (like photo albums).
78s were an A and a B side, they could be released as singles/one offs and generally weren't cohesive. Eventually you had people releasing related works or having longer recording sessions.
It sounds like a scenario of "What is old is new again", not that singles ever faded but all mediums outside of digital required the same amount of effort from the user for 1 track as it did for an album (put on an LP vs 45, cassettes had singles too but it's the same effort, CD changing would be difficult for a single song).
Because the amount of story that the creators want to tell requires more than 2 hours.
Shows with 50 or more episodes in the entire series may take years for the whole series to come out. But they never really try to go more than a year without releasing a batch of ~10 episodes.
I dont know why people listen to artists who can't put out reasonably consistent music. If only 10% of an artists output is not garbage, are they any good at all?
Reviews for metal bands will trash albums that only have 1 or 2 good songs.
No, only vapid pop music follows that one-hit-wonder blueprint. In almost every other genre the expectation is that a great album will be consistently great.
Sorry that you know of only ten decent bands, when there are actually tens of thousands of artists that never add filler to their albums.
Call me young but I didn't really realize until now that having to buy a whole album was a thing. I grew up with Kazaa, Ares, The Pirate Bay, Google Video, Newgrounds, YouTube, Grooveshark and Spotify (in that order). CDs existed but were something my parents used; I didn't like much music in the first place as a teenager and the stuff that I liked I never saw anyone own a CD of. Heck, I don't think I knew anyone that liked metal in the first place at that time.
In hindsight, I should have paid artists fairly, that much is clear to me now (I paid ever since Grooveshark).
Nevertheless, not having experienced the music album phase, it seems like an odd thing to pay for things you know you don't want and have to switch CDs the whole time to get to what you do want. I can't think of a single album I'd listen A-Z. There are a few rare bands where I like most of their songs (because they're very similar), and some bands where I don't like most songs but I like most from one or two of their albums, but I don't think there exists a single album where I like all songs, nor that I like >=50% of the songs in 50% of the albums of which I like >=1 song.
I also grew up during Napster & Co., but it's really not that hard to recognize that albums exist and aren't really tied to a specific medium. You can buy them as digital downloads, on CD or vinyl.
As someone who loves albums and regards EPs let alone singles as a lesser, rushed and not particularly worthwhile format, let me assure you that you don't have to pay for things you don't want. The trick is to buy strong albums that avoid filler content and don't need to rely on a single hit.
I admit that it takes effort to find the gems in a sea of mediocrity, but in absolute numbers there are more great albums than ever in addition to the ones you might have missed in the past. I'd say anyone claiming that there isn't any fantastic, album-worthy music isn't looking hard enough.
I hear this a lot, but I've never understood it. Do most albums have 1 or 2 stinkers? Sure, but I still don't skip tracks. I enjoy music the most when it's in the larger context of an album. I like anticipating my favorite songs and I like the transitions between them, even if it's just silence.
I guess this is one of those cases where the question sounded better in your head :-) You will probably go "duh!" after you read my answer.
First of all, I don't listen to albums, I listen to songs.
Secondly, greatness is hard to achieve and even harder to maintain. Listening just to songs as part of great albums composed of only great songs would mean that I would not listen to a lot of great songs, because in many cases some of the other songs on an album are bad. I would also probably listen to a lot fewer artists I like, for the same reason.
Thirdly, sometimes it's not even about great or not. In many cases my opinion of what constitutes a great song disagrees with what the artist considers a great song. So I might not personally like a song on an album. For example, I hate long guitar solos, I consider them a form of musical bravado which I do not like. Obviously guitarists would disagree with me :-)
And lastly, you committed a fallacy, it's called appeal to incredulity: https://en.wikipedia.org/wiki/Argument_from_incredulity ("someone must be listening to music <<the wrong way>>, their music must be horrible"). I'm free to listen to whatever I want and there are many ways to happiness. You can't objectively prove that someone's path to happiness is wrong as long as they don't harm someone else.
To quote someone I know: "Everybody thinks that all their friends are nice, all the music they listen to is good and all the code they write is good" :-)
Another whole-album-listener here, but I'm the minority among my peers (millennials). I find looking for complete albums where all of the songs are high quality really guides my taste when I'm looking for new music. If an artist's album consists of one amazing song and 11 bad ones, I likely won't listen to that artist again. The artists that have been able to knit 12 songs into a narrative and provide an enjoyable listening experience are the ones I keep coming back to year after year.
But I think you'll find most folks who do are older and/or fans of specific genres. A lot of acts in some genres don't release "albums" like you're thinking of anymore.
I personally salute you and suggest you educate the music lovers around you on necessity of active discovery and focused attention to the music they listen.
That's a strange stance. I assume there are a fair number of listeners that don't care to listen to anything new. Is there some business benefit to Spotify by artificially forcing churn?
It’s not about listening to something new; it’s about the wave of hype and promo you do around each release, it keeps you relevant and gives you something to talk about, keeps you fresh in peoples minds.
I get that side of it. But it seems to dismiss the listeners that don't care for any of that. The jist seems to be that they want to deliberately lower pay outs for older stuff even if it's played a lot.
Found the old person! You must be over 25! Just kidding, I'm double that. And it's true, I don't listen to much new music. Occasionally some rap song catches my fancy. I've branched out a bit after listening to various elaborations on the music and styles in Hamilton to new rap artists. But I still love 70s rock, I'm a dinosaur. I think this article does make sense though, new stuff gets more attention, eps for the young ones.
They've created new revenue streams by allowing the music labels to purchase advertising on spotify to promote new singles/albums. More singles/albums coming out -> More competition for listeners -> Drives up advertising prices. I imagine that's part of the reason.
I'm similar, but I wonder how many people stream often enough minor artists or minor songs from the past. Maybe keeping the engagement up is the only way for non-unforgettable artists to make enough money to make a living?
As my musician friend told me ->
“In music, you don’t get to update your songs. Once it’s out, it’s there for the world to see.” (Or something like that.)
It’s not like software where we can iterate/improve as we learn. In music, you capture a moment, a sensation, an experience - and you had better capture it as intended, even if that means an extra month in the studio.
This same musician told me that the increased ubiquity of singles is intended to speed up iteration/feedback/ship cycles.
It's interesting cause Kanye intentionally or unintentionally experimented with updating Life of Pablo several times. He changed track order, changed the production, changed the lyrics a bit. I believe he released 4 different versions.
It was super interesting at the time and a lot of internet music circles were talking about the potential of this going forward considering it's only possible because of the streaming services. But unfortunately it didn't seem to catch on and I haven't seen any other examples.
Looking at it in terms of one of the "golden ages" of recording music - back in the 1960s, think the Beatles - this would mean only returning to what has already proven possible before. Some artists in those previous eras were able to deliver remarkable music as much as twice a year over consecutive years.
Perhaps it was a less competitive industry at the time, but certainly less efficient (and relatively more expensive) production tools were available compared to today.
There are still plenty of musicians doing that. I haven't kept track lately, but John Zorn used to release albums every 2 months or so. Thee Oh Sees and King Gizzard & The Lizard Wizard are also constantly putting out new albums (King Gizzard had 5 albums in 2017).
I don't know enough about it to know if they are paying enough or if their payment rules need to change, but not everyone needs to be Britney Spears.
In fact, having tremendous wealth and fame come with additional costs. You end up needing a mansion, not for the space per se but for privacy and security. You end up needing bodyguards. Etc.
One way to make services like Spotify make sense for "average joe" musicians and other creatives is to address the other half of the equation: The cost of living for the average joe.
We basically only build upper class housing these days. The average new home in the 1950s was about 1200 square feet and housed about 3.5 people. These days, it's more like 2400 square feet (or more) and houses about 2.5 people
We have also torn down a million SROs and we largely have done away with things like boarding houses, where you rented a room and got breakfast and dinner as part of the rent and you supplied your own lunch elsewhere. Instead, we default to expecting young people to share a house or apartment designed for a nuclear family with a bunch of strangers. Then we make horror movies about it, like Single White Female, and then fail to go "Huh. Maybe that's not such a wonderful thing to insist young, single people do."
Health care is another major issue in the US. It costs way too much and many other developed countries handle that better, so we have many other examples to draw from. It's not like we need to brilliantly design something that's never been done before to make this work. We just mostly need the political will and to take our heads out of our butts in this country.
That sort of discourages deep work, similar to journalists encouraged to tweet often instead of releasing larger , more in-depth articles once a month.
Spotify pays independent artists (that aren't part of large negotiated artist pools from seriously major labels) poorly.
Indie artists are sinking more into getting their songs promoted on playlists than they gain in streaming revenue in return. It's literally "pay to be played"
(Remember "Pay to Play" in Sunset Strip clubs in the 80's? You would literally front the money for renting the club on a Tuesday or Thursday and then sell tickets and hope to pay off your "concert" (Friday/Saturday had music the club actually booked & paid, hence their "reputation" was solid)
Anyway, I severely resent the massive influence these relatively new corporate overlords have had in reducing the value of music significantly.
BANDCAMP is one suggested remedy, as they take their (relatively small, I think its' around 10%?) cut from your sales and allow you to have your bands page and sub-domain "yourband.bandcamp.com" etc... in other words: they treat you respectfully as humans should treat other humans.
This race to the bottom can only end in more Aututuned Pop, lol...
Spotify are effectively parasites. They pay poorly.
Dominate the market. Force everyone to accept their poor treatment through their market domination.
(Go watch some Youtube tutorials on paid spotify placement services)
And thus, this youthful tech CEO is speaking from a position of extreme privilege with an eye towards what is good for HIS business, but bad for musicians.
I hope Spotify goes away, but it probably won't.
The consumer side is pretty good/enticing etc.
The musician side is 2-tiered.
The vast majority of the POOLED streaming revenue goes to large labels who manage global pop stars.
Indie artists can just go die under a rock... ermmm "try harder"
I don't really want to read 20 hackernews' argue about how hard it is/isn't to make a hit album.
Quality music takes more effort, training, and skill to make than low-quality music, and this is not about the production alone.
Expert-level musicians are generally disregarded by Joe Q. Public in favor of Joe Q. Publics "favorite" musicians, training and knowledge and compositional skill notwithstanding. The limiting factor is Joe Q. Public's musical listening skills, which certainly will not improve in a world that is racing to achieve cheaper mass-produced conformist "music" that likely will be replaced by algorithmic generation in the near future.
I am hardly a great composer, but even I could describe formulaic "genre music" in such a manner that a decent music AI could iterate it nonstop.
The subject of people "liking genres" rather than following individual artists of talent across their works as they explore all genres, etc... that's a different topic for a different day.
I will also contend that grooming a public to like simplistic rhythms that are easy to program on 16-note step sequencers is not accidental.
Can someone explain why the revenue sharing model is broken on Spotify? A lot of bluecheck indie bands complain about their meager $28 monthly checks. I even saw a performative tweet from a label the other day: https://twitter.com/DonGiovanniRecs/status/12865225202213847...
Let's say I run a bookstore. 70% of my revenue goes to pay workers. A works 32 hours, B works 5 hours, and C works 3 hours.
Is it not fair to give A 80% of the pay pool, since she worked 80% of the hours?
Finally, terrestrial pop radio, independent college radio, and ham radioesque bedroom experimentalists are all on the same platform, the swim lanes have been erased. Everything is flattened, everything is discoverable, yet the sobering truth that few people are streaming their work elicits attacks on Spotify from bands and marginal artists. How is their contempt justified?
If you're running a brick and mortar book store, payroll would unlikely be 70%. Payroll would be your highest cost but it should be 30% on the high end.
If we want to try and analyze this, we should analyze what music distribution was like prior to digital. My guess is that for most artist who went through traditional publishing and distribution, they only received 10-20% per album sale. That seems high too.
I should have been clearer, I used 70% of revenue because that is what Spotify gives to artists. Of course, software can do that as it scale infinitely, retail cannot. I can't tell if 70% is generous or not -- many independent artists think it isn't.
Sounds like he's trying to redefine the music industry. Seems like most artists have a 5 year window, likely in their 20s where they'll produce something resonating with the current generation(s) and after that the younger clientele will be looking for something to define their own thing as well as the artist scratching around for some inspiring material. Generally of course. I take the position that most artists are in that bracket but exceptionally talented musicians have a longer lifespan.
Of course, a platform wants people to keep strutting out stuff on their own platform.
As for royalties from Spotify, perhaps an issue of the middlemen required and not required. I remember chat about the web being a great evener on this front where independents could strut their stuff and virality would take care of the rest.
Apparently convenience and one platform, as with so many things on the web trumps everything else.
People are railing at Spotify, when in reality this is responding to the desire and behavior of customers. All of the platforms are moving to a $10/month all-you-can-eat model, because customers love it and will pay. What this does is move the industry from an album model, where people buy bundles of songs and own them forever whether they listen or not, to a radio model, where people snack on music in curated playlists. In the album model, artists are incentivized to put out one or two hit tracks and then backfill with lesser material, or spend time curating an hour-long story. In the radio model, artists are incentivized to iterate on and tweak their hits with mash-ups, remixes, collabs, etc. That's why Lil Nas X put out FIVE versions of Old Town Road.
I don't get the fuzz about this. He can claim whatever he wants but fans will continue to listen to whatever they want. Artists can cater to it or not and all of it is outside of this CEO's control. It seems to be meant to discredit Spotify when I really don't think the alternatives are much better; even if alternatives currently pay artists better, they don't charge significantly more, which is the only way to pay artists more in the long run. Someone else said it somewhere in a subthread: charging subscribers enough to pay all those artists a livable wage just means they lose subscribers, go bankrupt (they're already not profitable), and end up paying artists zero in a few years.
So many bands lose their creative spark after the first album or two, and it's exactly because of this mentality. I'd rather have one album of music carefully curated over 4 years than four albums than have 4 albums rushed out over a year. The creative process takes time. If you just rush out the next album you're probably just adding onto the already massive heap of formulaic uninspired music. I'm not saying you can't have quality AND quantity, but the number of bands who can crank out both is limited.
Some famous musician (can’t remember their name) said something along the lines “you take 10 years to write your first album and they expect you to write the second in 6 months”.
This moron's statements piss me off so deeply I don't even know where to start.
The commoditization of Music?
The complete lack of understanding of how creativity works?
Telling artists to "work harder" while his net worth is in the billions, all thanks to the exploitation of said artists?
All this non-sense during a terrible crisis that puts many musicians out of job since nowadays they have to rely on live performances to make a buck, all thanks to....platforms like spotify?
I have never used Spotify and I never will. For me it's the Clear Channel of this decade, more interested in monetizing 'content' and having other people define genres so they can pipe similar music into people's ears for profit. A giant swathe of the 'music industry' has zero interest or understanding of creative processes, they just want product that sells easily. These formulaic eras always bland out and I'm sure Spotify will be no different
> The artists today that are making it realize that it’s about creating a continuous engagement with their fans. It is about putting the work in, about the storytelling around the album, and about keeping a continuous dialogue with your fans
The CEO shows a concerning misunderstanding. Some music is timeless. Some songs remain landmarks even decades after their release.
I abhor it when people force their own music preferences and beliefs on others. As other people have said, Mr. Ek can go fuck himself.
As Spotify CEO, we should think he isn't speaking to musicians, but to all the bigger labels on how to make the most money. Musicians should feel they can do whatever they want, and should, and probably would, but I reckon they would eventually hear it from their manager once the instructions trickle down the chain of command. That is, any musician signed and making money. To everyone else, the industry itself is largely irrelevant to their craft or their income.
I'm curious if anybody here has the experience to compare how much harder it is to make a living in music as an artists vs how hard it is to break into tech.
I'm biased myself, having spent the last 20 years studying software dev and then working in software. I feel that while it is a lot of work, the market is quite hot for any technical talent, and making a decent living is straightforward. Even a shitty software dev likely makes a comfortable living, you don't have to be Jeff Dean. In CS you can graduate from MIT and probably be set for life, but in music you can graduate from Juliard or Berklee and likely still be living on a futon for decades. Or am I off here?
Whereas as a solo producer/composer (or someone in a band) it seems that getting anywhere near ramen profitability is really really hard. It's reminiscent of startups: you have to find your niche, you have to constantly stay relevant, you have to market yourself incessantly on every medium, and most likely you'll be drowned into obscurity by someone else who became a winner-take-all in that space. And while the payoff for founders in small to medium startups is still pretty decent (and can be life-changing), as an artist is seems like the polarization between starving and huge is even greater.
Even in photography you can always "sell out" and shoot weddings in perpetuity while you're doing more artistic work on the side. And that "main gig" pays reasonably well, even though it's exhausting. But with music things like playing gigs will likely not get you anywhere close to sustainability. Maybe teaching is the "weddings" of music?
I have flirted with the music industry for longer than I've been an engineer, and the reason I kept getting pulled back into programming is that it is much easier to make much more money. But I think your question makes a false dichotomy. Building a band/solo act is like founding a startup, and being a programmer is more like being a wedding DJ. I think in both cases, you have better odds in the tech world.
I once had a piano teacher who remarked: "why do musicians make so little money? For the same reason they do so many drugs. Professional musicians are paid in dopamine."
There is nothing like a crowd cheering for you.
But you're right, you can make a living being a wedding DJ, or if you can get gigs at upscale hotels/restaurants playing cello or piano, you can make a pretty decent living as a musician. But it takes a lot more work than learning to hook up a bunch of web forms to a database, and usually, it pays a lot less money. Then there's always teaching; and I believe if you have the ability to kindly teach music to children, you have attained enlightenment.
If you are striking out on your own, the grind is like founding a startup. If you build yet another photo-sharing app (or you're a guitar player who sings about love), you have to be uncommonly talented and lucky to break through. Similarly, startups come in many genres; maybe high tech startups are like jazz - often founded by PhDs, and very often failures because no one can wrap their head around it - even though it is brilliant.
If you can make it work - either startups or music, you get paid in dopamine and money. If you make it out the other side of that gauntlet, I believe you have also achieved enlightenment.
If you want to "make a living", it can be done in music. However, you will generally be more stable and comfortable in the world of tech.
That's a good point, never thought of it that way. As a developer, it's satisfying to get better at your craft and to get small dopamine boosts from elegantly written solutions, but it's not like the rollercoaster of being in front of a bajillion people and sweating bullets and nailing your part, especially when playing in a band and melting away into something bigger than you. You can only get so much transcendence from writing login flow #65 and getting your integration tests to pass.
To be fair, as a startup founded you also don't exactly feel these incredible peaks of dopamine, but there's a longer sense of satisfaction, confidence and ownership of being a big part of something successful that puts food on the table for a bunch of people and solves pain points for even more people.
The way I see it, and I'd be curious to hear what you think, is that trying to get big the hard and old way in music (aka make a "stale" genre album, release album, hope you go viral) is kind of like starting yet another pizza shop in the middle of new york city. You're one of many, you likely don't have anything so revolutionary to suddenly crush your competition. You're very much of a commodity. Seems to me that you'd be much better off being at the beginning of a new fashion or trend in music genres, or using a new music or distribution technology (e.g. drum machines & synths, or maybe streaming your music on Twitch, or maybe TikTok) that's suddenly getting people's attention. Like with a startup, you'd be better off taking advantage of a landscape shift of sorts rather than making photo sharing app #57365 in 2020. Still won't guarantee success, but at least will improve it.
Sharing a concrete example:
4 friends of mine are in a moderately successful band in LA. They have 1M monthly listeners on Spotify, and their top track has 17M plays. Spotify sends them ~$2k/month, which isn’t enough for 4 to live on (they live together, and record with their own equipment in a rented house). To make ends meet, they play live shows and tour (which isn’t an option anymore). Support your favorite bands by buying merch.
This implies that every listener would have bought the CD. In reality, many of these listeners were just letting the playlist roll without thinking about it, and would not have spent a penny out of their pockets for that particular song.
The problem with Spotify is how they allocate royalties.
If I pay my £10 a month and only listen to whatever 5 indie bands I like, I’d expect them to get £2 each from my listening.
But they don’t. Lady Gaga will get most of my money despite me never listening to her music, because all they do is put everyone’s money in a big pile and divide it up depending on who got the most listens.
They have to do that because otherwise Lady Gaga wasn’t put her music on Spotify.
I'm a musician. But I will never make music like that. Also I will never use Daniel Ek's platform, unless I just by pure chance become one of the biggest out there, which I doubt. But then I don't make music for a living, and nor would I want to–especially in this day and age–despite a lot of folks telling me that I probably should.
The question then becomes, when I already know other valuable professions, why would I want to? Then the income would have to rival the other work I can do. I mean, I do get more pleasure from making music, but only because I don't overdo it. Work, on the other hand, is work. If I made music into work, I'm not sure I'd like it anymore. So then whatever I produce is intermittent.
But, if I wanted to make music a secondary activity, that doesn't need to be my main source of income, then yeah, I'd probably keep pushing music to places like Bandcamp. And I guess that's the beauty of it, because now there finally is a niche for us who don't want to be slaves to the art. Or to Ek's platform. So while Ek makes these outrageous claims, I'll be doing something completely different. And when I do decide to make love to my guitar, the music that comes out of it, will be a love child, and not the work of a mindless slave.
This is why I think "working musicians" never have to become slaves of the stupid and oppressive ideas of people like Ek. I just don't buy into that crap. There's a market. And that market has alternatives that also works for us who "just" want to release stuff intermittently. In fact I don't even use his platform to listen to music. Why, when there are so many great alternatives, that are also much more fair to the creators? (I guess an Ek fanboi downvoted. Hey if you've got something to say, say it, coward!)
Many of us in the music industry have been doing more single-based releases than album-based, because there really is no point to albums anymore. If you have a couple of tracks that all fit together, by all means, release them at once. But music isn't album-driven anymore, it's track-driven, and has been for a very long time.
That may be true in your corner of the music industry.
In the corners I'm familiar with (metal, rock, a bit of punk, various niche crossovers), the album is still king. Most artists will drop an album every couple of years and maybe an EP or two in between. Some of the most beloved and admired artists have released maybe 3 or 4 albums over a multi-decade career. They put together concept albums that work as a whole. Some will put out 40+ minute albums as a single track, because it's meant to be listened to in full, and needs you to take the time to do so.
There has been a slow change more towards EPs, especially for the more underground and independent bands. There's no reason to make a full album if you have 3-4 tracks that stand up on their own as a cohesive whole. In the old days, label's would often force artists to pad out those with filler tracks, just to produce a "full album".
These artists make their money through merch sales and touring, and it works because of a dedicated fanbase.
Popular music may be singles-driven and I would say that it has always been so. Thankfully there is a lot more out there than popular music.
For all I care, every single one of the big labels and streaming services could disappear tomorrow.
>But music isn't album-driven anymore, it's track-driven, and has been for a very long time.
This is exactly the problem: the technological advances changed the music consumption and music, unfortunately, follows. So instead of empowering the artists with more creative freedom it forced everyone to follow the same route. It's appalling.
I don’t know. Even back when CDs where still a thing, I never listened/liked the full album of any band. I think the most was like 50%. Maybe the technological advancements only make visible what was there before: only a minority listens to everything by one artist. Most people prefer a song or two and skip the rest, but they aren’t forced to buy the whole album anymore.
On the other hand, technology has always shaped music: each instrument's specific technology shaped what could be done, or not done, with it. And how disks, then tapes, then disks again, before Internet, all shaped how music was "packaged", experienced and distributed outside of live concerts.
Honestly, this seems beneficial to the musicians too. No longer worry about months/years of work going into an album that flops.
Be more iterative, experimental, and user driven by releasing an album's worth of content over the course of a few years. Seems like this reduces the risk for musicians publishing music that doesn't have an audience.
The secret to modern music isn't releasing full albums, but to release singles throughout the year. That is the future. I've even seen some (small-mid size) musicians simply have Patreon subscriptions to get new music, and then release the best on Spotify, etc. (I live in Nashville, so I feel like I'm near the cutting edge)
It's probably true and I find it sad. The best musical experiences I had have been from albums, not single tracks.
No more albums every 3/4 years means no more singles leading to the hype of an album nor more meaningful shows around one theme. No more albums means no more ways to put something on for an hour of development and feeling like you're (re)living something that has a beginning and an end while having consistency.
During this quarantine, After Hours by the Weeknd and The Slow Rush by Tame Impala have been my best partners. Sure I've listened to playlists, but the delusion of The Weeknd and the tempo of Tame Impala's album will stay in my mind as elements of this period, not single tracks. Asking musicians to constantly release is like removing the concept of books to only satisfy ourselves with news articles and tweets. Maybe it makes more money, but what a depressing artistic world.
I really only listen to whole albums. Especially Tame Impala, Father John Misty, Mac DeMarco, Lana Del Ray, Fleer Foxes, St. Vincent, Sufjan Stevens, and on. Those type of artists won’t change because of whatever the Spotify CEO believes. There is still a place for thoughtfully-crafted full albums and they’ll continue to exist.
Great, but 80% of it is crap either way. Does anyone else have difficulty finding good music suggestions?
Recommendations from friends for movies and TV shows are usually very good, but not with music. My opinion of Spotify, Apple Music and Amazon Music recommendations is much worse. I find that I am better off randomly picking music tracks.
There are a lot of good options for finding music. Some take more work than others. Some things I've done with success:
* go to the wikipedia page for an artist or album you like and look for
* reviews and ratings that line up with what you thought: you might find a critic or outlet with similar tastes that can provide new recommendations
* the producer or label: their other works might be similar to what you already like
* (on the artist page) the "associated acts" of the artist: you might like other projects with which the artist is involved
* any influences listed: you might like what the artist liked
* find a forum that aligns with your taste
* reddit has many music sub-reddits https://www.reddit.com/r/Music/wiki/musicsubreddits
* /mu/ is actually a pretty good 4chan board. The essentials wiki is a good starting point that doesn't require filtering through trolling: https://4chanmusic.fandom.com/wiki/Essential_Charts
When I craze the music often enough I go and purchase those tracks in bandcamp. This particular music mixer has a ton of content and clearly goes out of their way to sift through a bunch of music in that space they think is worthy to compile into a collection which they do a good job of.
Their mixes are reposted to soundcloud too I think.
I typically use https://rateyourmusic.com to find new stuff. Not their recommendation system, that’s trash.
Instead, I find albums I already know I like and look for well-written reviews on the album page. Then I click into the reviewer’s profile, see if I like/recognize any of the other stuff they’ve reviewed, and if so, I follow them. Boom, it’s like you’ve found a friend who actually enjoys the same music you do.
"Albums" are an artifact of the LP, cassette, and CD era. It's no longer necessary to release music as a collection, unless the point is to tell a larger story (eg. Kendrick Lamar's DAMN).
I think singles and smaller albums (5 tracks or less) are the sweet spot these days - for consumers and for artists.
Strange interview. As a Spotify customer this doesn’t give me that warm fuzzy feeling about supporting artists. It’s more likely to make me think twice about giving them my money.
It’s almost as if he was trying to annoy his artists and his customers. At best it comes across as pretty arrogant.
I notice an interesting adaption to this in the metalcore scene. Bands release shorter EP's (5-6 song mini albums) more frequently. In some cases 2 separate EP's are re-released after as a combined album plus 1-3 new bonus tracks.
At first I didn't really like this trend b/c I enjoy listening to the full album through on first listen. But I do understand why they are doing this to adapt to the streaming age.
Instead of releasing a 12-song album once every two years, they split that release into 3 every 6 months (5 song EP, 5 song EP, combined album + 2 bonus). As consumers we get the music earlier and more frequently. They get 3 release hype cycles for the same amount of creative work, which is what they need to be discovered in streaming platforms.
That has been going on in the greater metal scene for a long time. Especially for smaller artists, they put out a bunch of demos and EPs before putting together an album. In contrast to what you're mentioning, often the main tracks of some of the EPs will be on the album, but all of the "b-sides" will only be available on the EPs, but I've also seen examples of multiple EPs being re-released together. Not necessarily as an album in the traditional sense, but as a compilation.
I wish as a species we were smart enough not to turn everything good we did or could do into shit in the name of profit by building models that just churn more.
TBH, reading "what is required [by music artists][...] is creating a continuous engagement with their fans" my arms fell off to the ground.
Seriously, now the requirement for a form of art is engaging with fans? Well thanks Spotify for the nails to seal the coffin of good music I guess.
>'"Ek claimed that a "narrative fallacy" had been created and caused music fans to believe that Spotify doesn't pay musicians enough for streams of their music. "Some artists that used to do well in the past may not do well in this future landscape," Ek said, "where you can’t record music once every three to four years and think that’s going to be enough.”'
I'm pretty sure that paying a streaming royalty of $00.0038 per stream is an established fact and not a "narrative fallacy."[1]
Only a small handful of artists in the world have ever been able to afford to take 4 years off between albums. These are the U2 and Cold Plays. These are also the touring acts that make a fortune on the road so it's highly unlikely that these people are complaining. The majority of artists are hard working. The only way they make money now is by going on the road. It's helps to sell tickets if you have a new release you are promoting that their fans are excited about. It takes time to write an album's worth of good material. Once that's done you have to record, mix and master the record. This can easily take months. Once the album is in the can its up to the record company to choose a release date. There's usually some strategy with the date that is chosen. It takes time to book a tour and tour dates have their own strategy so it's not uncommon for a record to be done but not release for another few months. Then the artists goes on the road - they will try to hit the festival circuit in the US in spring and the European festival circuit in summer and then a tour in the Fall. All of this take time. A lot of time. Maybe you could complete this cycle "write, record, tour" cycle every year and certainly band in the 70's did for a time but that's not sustainable. Certainly not for anyone but a young person in their early 20's. Even being able to complete the "write, record, tour" cycle every 18 months is a grind.
Does Daniel Ek really not understand any of the basic realities of being a musical artist in 2020? Is he that clueless or is he simply promoting his own "narrative fallacy" of the lazy complaining artist?
Yup. Musicians are releasing like 20 new tracks per hour to Jango. I know because I'm on a mission to collect all of it. And like 71% is from independent artists who will never make it big and nobody will listen to. Not even me even though I like obscure stuff but even I can't keep up with the amount of new music that gets released all day every day.
Frustrating, but unsurprising. I look at electronic musicians online and you can see that - 15 years ago they'd drop one album every few years and it would be wall-to-wall perfection. Now they drop multiple albums per year and it's mostly... okay.
YouTube did the same thing to video content. The internet drives everything to the lowest common denominator.
The Spotify scam is incredible. Negotiate a 'deal' with the people that own you. Negotiate with hat deal. In favour of your owners and against the artists. Justify not paying more to the content creators based on making no money but the money you are 'losing' is going directly to your owners.
> "The artists today that are making it realize that it’s about creating a continuous engagement with their fans. It is about putting the work in, about the storytelling around the album, and about keeping a continuous dialogue with your fans.”
FU.
A musician's job is to create music. Period. It's hard enough on it's own.
When it comes to Spotify’s payment model I would rather that my $10 a month be split between the artists I listen to (based on number of plays) rather than the artists only getting $0.006 to $0.0084 for every play. I think for the more niche artists that can’t rack up the plays this model could be more sustainable. Thoughts?
Yes they can. Spotify fortunately is not the only way in which musicians can reach and audience and for some musicians it is still an art rather than a sausage of which they need to produce an extra quota to boost every Spotify quarterly earnings report.
Quality takes time, especially in music.
I wonder if that CEO has ever made any music themselves?
You are missing my point I think. Musicians do not necessarily make music for their fans and to 'engage' with people. It is art. It doesn't necessarily have to please anybody but the person that made it and it is still perfectly valid. Yes, people have to eat. But plenty of art was made by people without any form of compensation or interaction with their 'fanbase'. That's just mr. Ek saying what would help him.
I think the main issue is that people don't buy albums anymore. They only buy/listen to singles. So it doesn't make sense to work on a full album and release it a the same time. Because of the new channels, it makes more sense to release individual songs at regular intervals.
Imho allow free streaming and pay for downloads of what i like. Happy to pay £10-15 direct to artist with 100% going to them. Streaming can be covered by adverts.
Simples, but apperently so far :(
Otherwise paid performance amd commissions etc are always ways to generate more revenue.
It's interesting he's focused on quantity, and not quality. Maybe it's Spotify's suggestion algorithm but too much of what I hear sounds disposable. Not bad. Not a lack of talent. But a lack of ambition and a willingness to take chances.
Imagine we treated audio production the same way we treated video/movie production. The whole paradigm between starving musicians and Hollywood film stars is ridiculous. And to be honest, the two trades are not that much far removed from each other.
I think they are confusing musician with "content producer" Musicians make the music that we end up listening to 20 years later, content producers are the ones that create crap for us every day so we don't get bored that day.
Musicians should be getting closer and closer to 100% royalties as the infrastructure gets cheaper and cheaper. But, tech CEOs all like crappy music. Which is why they should never be in control of the livelihoods of creatives.
The reason there are any royalties to distribute at all is that Spotify has managed to create a distribution service that ppl are willing to pay for.
You can get free distribution for your music easily. Just put it on any torrent site...
I went back and read the original interview and I still don't know what he was actually trying to say. He stressed that artists need continuous engagement with their community and the old business model didn't work anymore. But I really can't tell if he was trying to say staggered, small releases, and do away with the big promos for albums, or if he really was saying "stop being lazy and churn out more product." He could easily claim the former for PR reasons, but the language he uses in the interview is incredibly ambiguous I honestly just have no idea what he's talking about. The sensationalist reporting obviously wants to play up the controversy, acting like the least favorable interpretation is unambiguously the correct one, but that's precisely why the public no longer trusts news media.
Ah yes: the fader, a smaller music publication, is just an example of why we should disregard media as a whole. /s
I don’t think the article is that sensational, I just think that you might have a predisposed belief that news media is always (or at least most of the time) sensationalized.
This is a such a clickbait title. Here I thought Spotify was requiring musicians to require more frequently, when all he meant was that if you release once every 3-4 years you're not going to see good results.
I publish one five-track mixtape as a solo artist once a year; and whatever singles I end up just naturally doing, as well. It’s a nice method that has successfully netted me some actual revenue, shockingly.
Same for software. Sometimes I dream of basic income and everyone having time to do just what they really believe others and they themselves will like to use/listen to/watch/experience, but it probably won't play out like that in my lifetime.
It takes enormous hubris to build a business as a parasite of music and its creators, and subsequently whip them for more productivity. I don't think I am the only person that associates the name "Daniel Elk" with the word "douche-bag" in my head.
COVID-19 is going to completely wreck the music business and the spotify guy is right: only endless social media campaigns with procedurally-generated soundtracks will survive in the future.
I make music for fun. No gyrations or even complete death of the music business will stop that. That would require something to happen to the "real" music business. (instruments and such)
Gorillaz used to do their shows almost entirely hidden from the audience and the characters would be projected onto screens. If we get to the point where we can generate actual good music we can do it in a live setting with avatars.
Ah, apologies. I had algorithm artists on my mind for whatever reason. Yes, that would still be a live concert and as someone who loves concerts I agree that there is no substitute.
Personnaly some of the weirdest and deepest hours of my life were live music. It catchs your soul and also bonds people together. No spotify playlist can do this I believe.
The only live music I'm getting these days is when my neighbors hang out on their respective porches. They are Berkeley's best trombone-and-banjo duo, probably. The last concert tickets I bought, for which I paid a rather incredible $540/pair, have been postponed to 2021 and realistically there's no reason to believe the concert will ever happen.
You know that feeling you get when you see, say, CSI "building a GUI interface in Visual Basic to track the killers?" That's how I feel whenever a post about music hits the front page of HN.
Save for a few genuinely educated and empathetic comments (thankfully towards the top for once), these comments (such is tradition) are a sea of Dunning-Kruger effects in full swing.
I'd like to respond to a few confident-and-wrong assertions I'm seeing rear their heads again:
No, "most musicians" don't make most of their money from gigs. That one band you know is not indicative of an entire diverse industry, nor is it a proper sample size. Those are amateur musicians, not pros. A pro musician who makes the bulk of their cash from gigs is doing it wrong. And not all musicians are instrumentalists: some are primarily writers, some are producers, or audio engineers, or synthesists, or whatever. There's so much diversity in this world that you're not aware of. If your idea of a "musician" is "someone carrying an instrument," that's a good sign that you know as much about this world as your family member who says "you're a software developer, why can't you fix my TV" as if knowing about computers means you know about hardware and electronics -- it just betrays a huge amount of ignorance on your part.
No, streaming does not pay people better than album sales. This is just...I don't even know where to start with how wrong this is, so if you believe this just know that you're dead wrong and I'll leave it at that. Look into my previous comments for more info if you'd like.
No, streaming is not supposed to be "exposure." It is a replacement of both album sales and radio plays. Artists and musicians constantly have to defend against the moving of goalposts -- first it was album sales nearly going away, then gigs becoming "exposure" for their albums, then their album sales becoming "exposure" for their gigs, then album sales became totally replaced by streaming and still being seen as "exposure" for their gigs/merch. It's a constant chipping away of previously-decent-income-streams that we're then told to be grateful for because they can apparently "expose" us for better payment. Which takes the form of...what? The only truly reliable thing left is licensing/royalties, which is also very heavily under attack.
Just -- people, please understand that you just probably don't understand very much about the complexities of a market with which you have zero experience. Your armchair analysis that makes sense only within the context of near-total ignorance isn't helping. HN tends to be full of smarter-than-usual people, so at least be smart enough to know when you don't know what you're talking about, and try approaching these topics with more empathy and humility.
Music is a commodity, a very easy to produce and distribute commodity. The facts are that the market was artificially controlled for the later half of the 20th century. Musicians, throughout history, have seldom made money. The later half of the 20th century was the anomaly, not the market of today or of prior...
As a musician and someone who spent years in the "biz". Its never going to be like it was in the 70's. Those days are gone.
I don’t think this is true remotely. In the last few years my only music purchases have been album releases on Bandcamp or independent record label sites, for musical acts like The Field, Swans and Nation of Language.
They release every 3-4 years and I buy albums. Works great.
Spotify seems to be about 50B on the stock market, so 9% of that is... 4.5B.
It seems that nearly 100% of his known money is virtual. I'm not sure he takes home a huge salary rather than paying out to artists ("Ek and his business parter Martin Lorentzon founded the company in 2006" so he didn't invest a lot to get those 9%). The company is also unprofitable currently, but having seen what non-profits like Mozilla take home I guess that doesn't say much (they also don't make any profit, and yet...).
Not sure if the actual salary is known. It seems like the kind of thing that would be, but duckduckgo annoyingly only turns up articles about this latest news.
The music isn’t your product, the music is your marketing. The shows, the merch, your influence - that’s your product.