Does The Spotify Gimlet Purchase Signal The End Of The Open World Of Podcasting?

from the hopefully-not dept

If you follow this kind of news at all, you probably have heard that Spotify has recently purchased two podcasting companies: Gimlet Media and Anchor. Gimlet makes a ton of high quality, highly produced podcasts (it's like the HBO of podcasting), while Anchor is a combination of a podcasting advertising network and a set of tools to let anyone create their own podcasts easily (it's like the SqaureSpace of podcasts). On the one hand, it's good to see podcasts getting some attention and interest, and Spotify is clearly one of the largest services for listening to audio files -- though much more so on the music side.

My concern, however, is about the potential walling off of the podcast world. The whole concept of podcasts from the early days was the idea that anyone could create them and anyone could access them. That's been changing a bit of late. There have been a growing number of exclusive and walled off podcasts, including on Spotify (but also on Stitcher with its Stitcher Premium and Slate with its Slate Plus program -- and likely others as well).

And obviously, it's nice to see experimentation around business models regarding podcasts, but as some are already pointing out, this could be another nail in the coffin for the idea of an open web.

In an interview about the Gimlet deal, Gimlet's co-founders insisted that they didn't think it was going to lead to this sort of fragmentation for existing podcasts, but future podcasts might be a different story.

Alex Blumberg: Yeah, I mean, well, Spotify said themselves this morning on the earning call that they have no intention of taking the shows that are out there that Gimlet produces and putting them behind a paywall. So those will continue to be freely distributed.

Matt Lieber: Or make an exclusivity to Spotify, the existing shows.

Alex Blumberg: Yeah. Yes. The existing shows will not be made exclusive to Spotify. They will continue ... you’ll continue to get them where you get them now. And yeah, going forward, I think it’s going to be a mix. This is a new world, and we’re trying to figure out how it works. And so it’ll be a mix of exclusive things that we make exclusively for Spotify, like we’re doing right now with Mogul or things that are windowed or things that are a mix of the two. I think there’s gonna be a lot of experimentation.

Frankly, the idea of windowed podcasts doesn't bother me that much -- releasing versions a bit earlier to subscribers seems like a reasonable business model for some kinds of content. But creating full exclusives, building a fragmented world of podcasting with walled gardens sounds like a disaster for the medium. There are already people complaining about all the walled gardens in video entertainment content. We don't need to add to that with podcasts, when the podcast world has already shown that an open, standards-based system built on RSS works just fine.

Filed Under: exclusives, open internet, paywalls, podcasts, rss, walled gardens
Companies: anchor, gimlet, spotify

Max Schrems Files New Privacy Complaints That Seem To Show The Impossibility Of Complying With The GDPR

from the what-a-stupid-law dept

We've written many times about privacy activist Max Schrems, who almost single-handedly brought down the silly privacy safe harbors between the EU and the US. Last year, we wrote about his newest project called noyb.eu, which stands for "None Of Your Business."

Last week, Schrems and noyb announced a big list of GDPR complaints filed in Austria, against basically every streaming media company, none of which -- they claim -- are in compliance with the GDPR. Schrems also provided everyone with a handy dandy chart showing the basic details of the results of the GDPR requests they made to eight different streaming platforms, where they fell down, and how much they might be on the hook for:

If you'd like to see the actual complaints, here they are for Amazon, Apple, DAZN, Flimmit, Netflix, Soundcloud, Spotify, and YouTube.

I have lots of thoughts about this, so let's list them out:

1. This demonstrates the near impossibility of complying with the GDPR: While I'm sure many will view this as a positive for the GDPR, in that Schrems is going after a bunch of big companies who many people love to hate, I'd argue that these complaints really show just how ridiculous the GDPR is in practice. At least with the larger companies on this list (Amazon, Apple, YouTube, Netflix, and Spotify) it is ridiculous to argue that any of them were deliberately avoiding the GDPR requirements. All of those companies have been well aware of the GDPR for years and spent the past few years spending many, many millions of dollars preparing for the GDPR. All have decently large teams focused on doing everything they can to comply, in part because of the possibility of massive fines if they fail.

   The fact that those large companies, who have all the resources in the world, are still deemed by Schrems to fail on nearly every aspect of the GDPR suggests, pretty clearly, that it is nearly impossible for anyone to truly be GDPR compliant in any reasonable sense.

2. The nature of the complaints shows just how silly the GDPR continues to be: Taking the Apple Music complaint as an example, the company did allow noyb and its client to download all the data it had, but noyb is demanding significantly more information under the GDPR -- much of it is information that would effectively be impossible to provide in the first place. For example, the complaint notes that Apple didn't provide "information about the purposes of the processing." But... isn't that the kind of information that anyone signing up for Apple Music already knows about when they sign up? Apple is using your information to provide you access to music and to recommend other music to you. What good does it do to have that information need to be spelled out once again at a later date to avoid massive billion dollar fines?

3. The possible fines remain completely insane: Note the numbers on the "maximum penalty" associated with these complaints. Under the GDPR, a company can be fined either €20 million or 4% of annual global turnover whichever is greater. So those eye-popping numbers are basically that 4%. Remember, most of the companies here bent over backwards to try to comply, with most of them setting up useful systems that allow users to download all of their data, even if noyb didn't like the format that data was in. And yet they might still face billions in fines?

4. GDPR could destroy some of these companies: It is surprising to see two companies -- DAZN and Soundcloud -- not respond at all to these requests. Both of them are based in the EU (though DAZN may escape via Brexit shortly, but it operates in many EU countries). I would think, at the very least, these companies would have in place some method of responding to GDPR requests. Soundcloud, despite its level of popularity, has struggled even to stay alive -- and came very close to shutting down a year and a half ago before getting a last minute reprieve from some investors. Either way, the company is clearly struggling, and the fact that both of these company's "maximum" possible fines are €20 million suggests that this is "greater" than 4% of their annual turnover. In short, this is likely a crippling and possibly company-destroying amount for these smaller operations. I'm still surprised neither responded to the requests at all -- but it's going to be difficult for either to stay in business facing these kinds of headwinds thanks to the EU's overaggressive regulations.

I'm sure that many don't seem to care that this might cause problems for these companies, or that it may be literally impossible to comply with these regulations. But we should all be concerned if regulations make it effectively impossible to be in business on the internet in the EU. We should be even more concerned that -- as many of us predicted -- regulations like the GDPR seem to have a high likelihood of completely destroying smaller players, like SoundCloud. The huge fines for the big companies are eye-popping and totally disconnected from any actual "harms," but most of those big companies can grudgingly afford to pay them. That seems unlikely for the smaller players meaning that -- once again -- the EU seems to be clearing the field of smaller internet companies, and leaving in place only the giants, from whom the government will just keep siphoning off cash.

One final point on all of this: I recognize that there are lots of legitimate concerns about privacy in this day and age -- and, in particular, how various data collection companies are using our private data. And I've long been on record that companies should be not just a lot more transparent about the data they collect and how they use it, but also should push control over that data out to the end users. But, looking over this list, none of these are companies that I'm particularly worried about concerning how they use my data. Yes, there are potential privacy concerns here, but the idea that SoundCloud or Spotify contains data so sensitive that they should be fined massive amounts for not making it "intelligible" just seems disconnected from any real harms and any real concerns.

Indeed, my concern with this type of litigation is that it actually waters down and distorts the real concerns we should be having over privacy in the internet era. Netflix not giving me all of the data on what I've been watching via streaming doesn't seem like a particularly big consumer concern -- and yet if it sucks all the air out of the room, it makes it that much harder to deal with real privacy questions raised by internet giants.

Filed Under: data protection, eu, gdpr, max schrems, privacy, streaming
Companies: amazon, apple, dazn, flimmit, netflix, noyb, soundcloud, spotify, youtube

Facebook's Latest Privacy Screwup Shows How Facebook's Worst Enemy Is Still Facebook

from the get-your-act-together dept

There's another Facebook scandal story brewing today and, once again, it appears that Facebook's biggest enemy is the company itself and how it blunders into messes that were totally unnecessary. When the last story broke, we pointed out that much of the reporting was exaggerated, and people seemed to be jumping to conclusions that weren't actually warranted by some internal discussions about Facebook's business modeling. The latest big scandal, courtesy of a big New York Times story, reveals that Facebook agreed to share a lot more information than previously known or reported with a bunch of large companies (though, hilariously, one of those companies is... The NY Times, which The NY Times plays down quite a bit).

The social network permitted Amazon to obtain users’ names and contact information through their friends, and it let Yahoo view streams of friends’ posts as recently as this summer, despite public statements that it had stopped that type of sharing years earlier.

As Kash Hill notes in a separate story at Gizmodo, this suddenly explains a story she had explored years ago, where Amazon rejected a review of a book, claiming the reviewer "knew the author," (which was not true). However, the reviewer had followed the author on Facebook, and Amazon magically appeared to know about that connection even if the reviewer did not directly share her Facebook data with Amazon.

The NY Times report further explains another bit of confusion that Hill has spent years trying to track down: how Facebook's People You May Know feature is so freaking creepy. Apparently, Facebook had data sharing agreements with other companies to peek through their data as well:

Among the revelations was that Facebook obtained data from multiple partners for a controversial friend-suggestion tool called “People You May Know.”

The feature, introduced in 2008, continues even though some Facebook users have objected to it, unsettled by its knowledge of their real-world relationships. Gizmodo and other news outlets have reported cases of the tool’s recommending friend connections between patients of the same psychiatrist, estranged family members, and a harasser and his victim.

Facebook, in turn, used contact lists from the partners, including Amazon, Yahoo and the Chinese company Huawei — which has been flagged as a security threat by American intelligence officials — to gain deeper insight into people’s relationships and suggest more connections, the records show.

As Hill noted on Twitter, when she asked Facebook last year if it uses data from "third parties such as data brokers" to figure out PYMK, Facebook's answer was technically correct, but totally misleading:

Specifically, Facebook responded: "Facebook does not use information from data brokers for People You May Know." Note that the question was if Facebook used information from "third parties" and the "data brokers" were just an example. Facebook responded that it didn't use data brokers, which appears to be correct, but left out the other third parties from which it did use data.

And this is why Facebook is, once again, its own worst enemy. It answers these kinds of questions in the same way that the US Intelligence Community answers questions about its surveillance practices: technically correct, but highly misleading. And, as such, when it comes out what the company is actually doing, the company has completely burned whatever goodwill it might have had. If the company had just been upfront, honest and transparent about what it was doing, none of this would be an issue. The fact that it chose to be sneaky and misleading about it shows that it knew its actions would upset users. And if you know what you're doing will upset users, and you're unwilling to be frank and upfront about it, that's a recipe for disaster.

And it's a recipe that Facebook keeps making again and again and again.

And that's an issue that goes right to the top. Mark Zuckerberg has done too much apologizing without actually fixing any of this.

One bit in the NY Times piece deserves a particular discussion:

Facebook also allowed Spotify, Netflix and the Royal Bank of Canada to read, write and delete users’ private messages, and to see all participants on a thread — privileges that appeared to go beyond what the companies needed to integrate Facebook into their systems, the records show. Facebook acknowledged that it did not consider any of those three companies to be service providers. Spokespeople for Spotify and Netflix said those companies were unaware of the broad powers Facebook had granted them. A Royal Bank of Canada spokesman disputed that the bank had any such access.

Spotify, which could view messages of more than 70 million users a month, still offers the option to share music through Facebook Messenger. But Netflix and the Canadian bank no longer needed access to messages because they had deactivated features that incorporated it.

This particular issue has raised a lot of alarm bells. As Alvarao Bedoya points out, disclosing the content of private communications is very much illegal under the Stored Communications Act. But, the NY Times reporting is not entirely clear here either. Facebook did work hard for a while to try to turn its Messenger into more of a "platform" that would let you do more than just chat -- so I could see where it might "integrate" with 3rd party services to enable their features within Messenger. But the specifics of how that works would be (1) really, really important, and (2) should be 100% transparent with users -- such that if they're agreeing to, say, share Spotify songs via Messenger, they should absolutely be told that this means Spotify has access to whatever they have access to. A failure to do that -- as appears to be the case here -- is yet another braindead move by Facebook.

Over and over and over again we see this same pattern with Facebook. Even when there are totally reasonable and logical business and product decisions being made, the company's blatant unwillingness to be transparent about what it is doing, and who has access to what data, is what is so damning for the company. It is a total failure of the management team and until Facebook recognizes that fact, nothing will change.

And, of course, the most annoying part in all of this is that it will come back to bite the entire internet ecosystem. Facebook's continued inability to be open and transparent about its actions -- and give users a real choice -- is certainly going to lead to the kinds of hamfisted regulations from Congress that will block useful innovations from other companies that aren't so anti-user, but which will be swept up in whatever punishment Facebook is bringing to the entire internet.

Filed Under: data sharing, mark zuckerberg, people you may know, privacy, tracking
Companies: amazon, facebook, netflix, new york times, spotify

Unintended Consequences: How The GDPR Can Undermine Privacy

from the be-careful-what-you-wish-for dept

We've highlighted a few times now, just how problematic the GDPR is. This is not because we don't care about privacy -- we do very much. We just think that the GDPR's approach is not a very good one with a lot more downsides than upsides -- and, it's unlikely to do very much to actually protect your privacy. For example, we just wrote about the GDPR being used (successfully!) to try to erase a public court docket.

But not only do we think that the GDPR doesn't actually protect your privacy, it might actually put it at much greater risk. Take the story of Jean Yang, who noted that someone hacked her Spotify account and then, thanks to GDPR requirements, was able to download her entire Spotify history.

That's because, under the GDPR, platforms are supposed to make all of the data they have on you easily downloadable. The theory is that this will help you understand what a company has on you (and, potentially, to request certain data be deleted). But, it also means that should anyone else get access to your account, they could access an awful lot of important and/or personal data. Your Spotify interactions might not seem like that big of a deal, but plenty of other services could expose much more sensitive data (and, who knows, there are situations where your Spotify data might be quite sensitive as well).

As Jean notes in a later tweet, this kind of thing could really come back to bite other services, such as Lyft or Uber. She jokes: "Would be pretty bad to get hacked and kidnapped in the same day."

There are possible technological solutions that could help (again, as Jean suggests), such as using multi-factor authentication to access your own data (one-time passwords, Yubikey, etc), but it's telling that few companies (or regulators!) have really thought about that, because that vector of attack probably hasn't occurred to many people. But, it probably will now.

This is, of course, yet another good example of the unintended consequences of regulating technology, even with good intentions. Very little thought has been put into the second and third order effects. Instead, you have a bunch of policymakers who think "platforms collecting too much data is bad, thus, we have to let people check on their own data." It never occurs to them that this now creates a brand new route to accessing potentially valuable, sensitive and private data.

And, as an end result, a regulation designed to increase our privacy... could sometimes have the exact opposite effect.

Filed Under: breaches, data, data protection, gdpr, hacked accounts, privacy
Companies: spotify

How The Record Labels Screwed Up The Music Industry, And The Tech Industry Saved Them

from the start-thinking-about-the-customers dept

If you've been following how much the record labels stumbled around the internet for the past couple of decades, then you know the basics here. But time has a way of erasing some of the nuances of history, and I find it incredible to watch the RIAA and the record labels these days walking around proudly acting as if they were the ones who "saved" the music industry by embracing streaming services that now make up the bulk of the recording industry's revenues. Indeed, as we've pointed out for years, the recording industry has a very long history of overvaluing the music and undervaluing the services that people want. They've spent so long insisting that the music is the sole source of the value of what they produce, that they always downplay (or entirely erase) the rest of the equation: getting the music to fans in a manner that is convenient, reasonable, and non-burdensome. Instead, they always focus on killing the golden goose -- insisting that any successful music tech service pay them more and more until they're squeezed dry.

Over at Motherboard, Ernie Smith, has a good history of how the recording industry screwed up streaming in the early days (unfortunately he does what most people do and refers to what's really the "recording industry" as the "music industry" -- and also simplifies the history to be just one round of mistakes, rather than many, many mistakes leaving a graveyard of dead tech companies in its wake -- but the overall article is still excellent). It's a very instructive piece in detailing exactly how the record label bosses were so focused on making sure that they had control and limits, that they didn't care at all about providing a service that people actually wanted. Much of it focuses on the two idiotic music label-approved streaming services that the industry tried to launch MusicNet and PressPlay (which we dubbed MusicNot and PressPause way back in 2001). Smith details how both services were built entirely focused on "how do we protect our revenue stream" rather than "how do we serve the customer." Thus, you had lots of DRM and lots of stupid limits:

Their offerings were slightly different—for a $9.95-per-month fee MusicNet allowed for 100 temporary downloads and 100 on-demand streams at launch, according to Billboard, while Pressplay allowed for 300 streams and 30 downloads and offered limited CD-burning capabilities at higher price points—but the tissue tying the two approaches together was DRM.

It certainly wasn’t music, as the services made no effort to collaborate with one another. In the post-Napster era, the combination of limited libraries and competition from peer-to-peer file sharing services put the companies at a major disadvantage.

Incredibly, inside the industry, executives so believed their own bullshit that they really thought these consumer-unfriendly services would win over piracy:

Stephen Witt, in his 2015 book How Music Got Free, portrayed Universal Music’s then-CEO, Doug Morris, as being overly excited about PressPlay, to the point where Recording Industry Association of America President Hilary Rosen, frequently portrayed as an “enemy combatant” of the Napster era during this time, had a hard time talking him off the ledge. (Rosen, it should be said, was simply sharing the company line because it was her job.)

“On several occasions he told Rosen to stop talking to Napster, to stop negotiating with the Fannings, to stop worrying so much, because he had something that would ‘make it all go away,’” Witt wrote. “In later years, PressPlay would be a reliable starting point for listicles of the ‘Top All-Time Tech Busts.’”

If you don't remember Doug Morris, he was the guy who, back in 2007, gave a stunning interview where he insisted that the reason why the labels missed the boat on the internet was... because they didn't know anyone who understood technology and then blatantly admitted that he didn't know how to hire someone who would have understood technology. To me, that seemed like evidence for why the board should have fired him and brought in someone who actually understood the market. For Morris it just meant he remained a revered leader of the recording industry.

But really the thing that screams out through Smith's piece is the same old thing: the record labels seem to believe that the only thing that matters -- the only thing that provides any value at all -- is the music (which, coincidentally, is the one thing they have some element of control over thanks to copyright). And thus, the idea that the technology or service matters, or that making a service that people actually want is considered barely even worth considering.

Indeed, reading through the article, it basically shows how the tech industry -- in the form of Apple and Spotify -- more or less rescued the recording industry from their own incompetence. And the incredible thing today is that the record labels still whine and complain that the tech industry that saved them is "taking advantage" of them, rather than showing them the way forward.

Filed Under: business models, doug morris, drm, golden goose, music, overvaluing content, streaming
Companies: apple, musicnet, pressplay, riaa, spotify

Patent Troll Sues Spotify, SoundCloud And Deezer Over Patent On A 'Music Organizer And Entertainment Center'

from the oh-for-sure,-man dept

Another day, another story of another patent troll. This one is about MOAEC Technologies LLC, a "patent licensing" company that exists solely around four related patents for a "music organizer and entertainment center." Last month, MOAEC sued Spotify, SoundCloud and Deezer over these patents. It's interesting that the lawsuit came just a few weeks before Spotify's IPO, as we've seen a bunch of companies sued for patent infringement right before their IPOs -- but it didn't prevent Spotify's IPO from happening.

All three lawsuits focus on US Patent 6,232,539, which is described this way:

A music organizer and entertainment center provides a center having a microprocessor, sound card functions and high-volume data storage and retrieval units for playing back music according to a variety of predetermined categories. Music can be played back in random form or can be played back according to a particular pre-selected order. The categories are provided by service provider who delivers selected titles and/or songs to the end user. The songs are typically loaded using a custom CD-ROM provided from the service provider. The music is provided in data-compressed form and is decompressed and processed through a sound card during playback. The categories can include a variety of parameters such as title, artists, date, speed, dance characteristics, subjective energy level and music style, such as easy-listening, upbeat, etc.

Digging into the actual claims, the real focus here appears to be on the ability to select "a category" of music:

a graphical user interface display having a plurality of selectable screens, at least one of the selectable screens including a plurality of category buttons constructed and arranged so that when a predetermined of the category buttons is activated, music selections having category flags matching the predetermined category of a respective of the buttons are selected and listed on the display.

Everything else in the patent appears to just be around playing music. But it's difficult to see how this even remotely passes the Alice test for a generic computer function. The nearly identical filings try to insist that this patent revolves around some magical invention:

The claims of the ’539 Patent do not merely recite the performance of some business practice known from the pre-Internet world along with a requirement to perform it on the Internet. Instead, the claims of the ’539 patent recite one or more inventive concepts that are rooted in computerized electronic data communications networks, and an improved method to deliver content and provide interface among different accounts and computing systems.

The claims of the ’539 patent recite an invention that is not merely the routine or conventional use of electronic devices for music playback. Instead, among other things, the invention adds new features to deliver content, integrate application interfaces and other protocols together on shared networks. The ’539 patent claims thus include improvements for, for example, storing music information in association with a set of predetermined flags, including an ownership category flag, and permitting user-customizable playback of music, based in part on the flags.

Accordingly, each claim of the ’539 patent recites a combination of elements sufficient to ensure that the claim in practice amounts to significantly more than a patent on an ineligible concept.

Of course everything described above was not limited due to the lack of ideas to be able to do any of this, but just the lack of ready general purpose computing equipment. In other words, there's nothing unique or non-obvious is the patents. The reason we didn't have this earlier was that the underlying technology wasn't ready for it, which is part of what should invalidate these patents under Alice.

That is... if these companies actually fight the lawsuits. As always, it is often easier and cheaper to settle a patent lawsuit than to fight it, even if you would win. Though, settling will only attract more of these kinds of nuisance lawsuits.

Filed Under: music, music categories, music organizer, patent trolls, patents
Companies: deezer, moaec technologies, soundcloud, spotify

Killing The Golden Goose (Again); How The Copyright Stranglehold Dooms Spotify

from the because-of-course dept

For many, many, many years, we've talked about how the legacy entertainment industry will seek to kill the Golden Goose by strangling basically any innovation that is helping it adapt to new innovations. We saw the same pattern over and over and over again. The simple version of it goes like this: the legacy entertainment industry sits around and whines about how awful the internet is because it's undermining its gatekeeper business model that extracts massive monopoly rents, but does nothing to actually adapt. Eventually, companies come along and innovate and create a service (a) people want that (b) actually is legal and pays the legacy companies lots of money. This should be seen as a win-win for everyone.

But the legacy companies get jealous of the success of the innovator who did the actual work. They start to overvalue the content and undervalue the innovative service. The short version of this tends to pop up when a legacy entertainment exec says something like "why is innovative company x making so much money when all it's doing is making use of our content?" Of course, if the service part was so obvious, so easy, and so devoid of value, then the legacy entertainment companies would have done it themselves. But they didn't. So with the jealousy comes the inevitable demand for more cash from the innovator. And, usually, demands for equity too, which the innovator has basically no ability to resist, because they need to have a "good" relationship with the content companies. But the demands for more (and the jealousy) never go away.

The end result, of course, is that tons of innovative businesses that created amazing services that people liked get crushed. Completely. Venture capitalist David Pakman (who founded one of the companies, which I used way back in the day, that was eventually crushed, called MyPlay) detailed how the legacy recording industry used this strategy to bury more than 150 companies over the past two decades. It's the same story over and over again. Any company becomes too successful and the legacy copyright holders squeeze them to death, whining the whole time about how they don't pay enough. As Pakman wrote:

The music industry complains loudly about the “leverage” these giants have over them. First they criticized Apple iTunes for not agreeing to raise prices above $0.99, then they went after Pandora and other webcasters by insisting webcasting rates were too low, then they attacked Spotify for not paying them enough, then they insisted Apple Music pay them more than Spotify did, and now, just as the YouTube licensing agreements are coming up for renewal, they complain YouTube doesn’t pay them as much as Spotify.

But this is a “crisis” of their own making. Many of us argued for years that it was in the industry’s best interest to create a healthy ecosystem of hundreds or thousands of successful companies, all enjoying successful businesses around music. But those arguments fell on deaf ears, and instead the industry fought repeatedly to raise royalty rates over and over again, despite evidence that not a single company ever achieved profitability.

In my mind, it would have been in the best long-term interests of the recorded music business to enable the widespread success of thousands of companies, each paying fair but not bone-crushing royalties back to labels, artists and publishers. But the high royalty rates imposed upon startups, even after clear signs over the past 19 years that the strategy killed companies, has prevented a healthy ecosystem from emerging. It’s a bed the music industry made for itself, and now it is left to lie in it.

Not only does this crush lots of interesting companies, the history of this sort of destruction has served as a giant warning sign to entrepreneurs. Years back we wrote about entrepreneur Tyler Crowley explaining how this kind of history makes entrepreneurs steer clear of doing anything with music. His original post is sadly gone from the internet, but we've still got some quotes that highlight the key points. His argument was that there are different options for entrepreneurs, which he describes as "islands" with different rules and conditions to "dock" at those islands:

For tech folks, from the 35,000' view, there are islands of opportunity. There's Apple Island, Facebook Island, Microsoft Island, among many others and yes there's Music Biz Island. Now, we as tech folks have many friends who have sailed to Apple Island and we know that it's $99/year to doc your boat and if you build anything Apple Island will tax you at 30%. Many of our friends are partying their asses off on Apple Island while making millions (and in some recent cases billions) and that sure sounds like a nice place to build a business.

And what about "Music Biz Island"? Well, the labels have made it clear you don't want to dock there.

Now, we also know of Music Biz Island which is where the natives start firing cannons as you approach, and if not stuck at sea, one must negotiate with the chiefs for 9 months before given permission to dock. Those who do go ashore are slowly eaten alive by the native cannibals. As a result, all the tugboats and lighthouses (investors, advisors) warn to stay far away from Music Biz Island, as nobody has ever gotten off alive. If that wasn't bad enough, while Apple and Facebook Island are built with sea walls to protect from the rising oceans, Music Biz Island is already 5 ft under and the educated locals are fleeing....

That doesn't seem healthy for music. And that brings us around, of course, to Spotify. The music streaming giant has filed to go public, and Ben Thompson over at Stratechery has done a bang up job highlighting why even this hugely "successful" music platform looks like a disaster from a standard internet investment perspective. Its margins suck. Its margins suck so bad it's still unclear if Spotify can ever make money. Because it's the same old story that we described above, where the labels (who own a large chunk of Spotify -- remember the equity demands?) are crushing the company in a way that is unlike basically every other successful internet company. Internet companies are built on the idea of huge margins, because the marginal of one more customer is minimal.

But not with Spotify. Spotify has to give over so much of its revenue to the labels that it's nearly impossible for it to ever be a viable business. Ben points out that the revenue and costs numbers show that Spotify operates like any "well-managed SaaS company." But it has a "marginal cost problem" in that the label deals (even ones that were restructured recently) guarantee that nearly all of the money that Spotify gets... goes right out the door to the labels.

Spotify’s margins are completely at the mercy of the record labels, and even after the rate change, the company is not just unprofitable, its losses are growing, at least in absolute euro terms:

Ben has all this laid out in his usual nice charts and graphs and such that are worth checking out.

But, what this all comes down to, yet again, is the stranglehold of a messed up copyright system. The fact that the labels can kill the golden goose over and over and over again is because of one simple reason: the artificial monopoly handed to them by the copyright system, and the power it bestows. It's a market distortion. This isn't to say that there shouldn't be any copyright (let's see if our usual trolls make it this far in the post or if they've already dashed off a comment about how I want no copyright at all...). But it certainly demonstrates how the copyright system is so weighted to favor the copyright holder that they can strangle basically any business that touches on copyright, and make those markets entirely different from basically any similar business that isn't encumbered with copyrights and legacy businesses who, having failed to adapt themselves, now demand a king's ransom from the companies that did all the adapting for them.

Filed Under: business models, copyright, economics, golden goose, margins, music, recording industry
Companies: spotify

Spotify Finally Realizes That Streaming Isn't Reproduction Or Distribution

from the copyright-conundrums dept

Hold on tight: we're going to get down into the weeds a bit on a copyright issue. In early 2016, we wrote about the "insanity of music licensing" as it related to streaming music, and Spotify in particular. This was in response to a series of class action lawsuits filed against Spotify by songwriters, claiming a failure to properly license so-called "mechanical rights." As I noted at the time, I talked the case over with a large number of copyright lawyers -- and many were left scratching their heads regarding what the lawsuit was actually about. Spotify, of course, is famous because it's a licensed music streaming service. That's it's whole thing. But, as we discussed, part of the problem is that there are a ridiculous number of different possible licenses out there -- many of which were designed for different types of technologies, and, when it comes to internet services, some people seem to assume that the services need to license roughly "all of them."

So, it was always known that a company like Spotify needed to secure negotiated license to use the sound recordings (that's from whoever holds the copyright on the actual recording -- not the composition). And they had to get public performance licenses from Performance Rights Organizations (PROs). But the question in these lawsuits was about an entirely different license -- the "mechanical" license. As you may have guessed from the name, a "mechanical" license comes from way back in the day, when the companies manufacturing records needed to get a license from the composer for the sake of reproducing and distributing the songs which those songwriters wrote.

And here's where we do some copyright 101. Section 106 of the Copyright Act includes a definitive list of six rights that copyright may grant the holder an exclusive right to (absent something such as fair use). These then are the specific exclusive rights under copyright law:

1. to reproduce the copyrighted work in copies or phonorecords;
2. to prepare derivative works based upon the copyrighted work;
3. to distribute copies or phonorecords of the copyrighted work to the public by sale or other transfer of ownership, or by rental, lease, or lending;
4. in the case of literary, musical, dramatic, and choreographic works, pantomimes, and motion pictures and other audiovisual works, to perform the copyrighted work publicly;
5. in the case of literary, musical, dramatic, and choreographic works, pantomimes, and pictorial, graphic, or sculptural works, including the individual images of a motion picture or other audiovisual work, to display the copyrighted work publicly; and
6. in the case of sound recordings, to perform the copyrighted work publicly by means of a digital audio transmission.

So, when Spotify is licensing sound recordings from copyright holders, it's licensing #6. And when it's getting PRO licenses, it's covering #4. So, here's where things get trickier. The mechanical license -- usually granted by the composers to whoever is manufacturing, say, a record or a CD, and done via someone like the Harry Fox Agency (HFA), covers the rights for #1 and #3 -- to "reproduce" and to "distribute." That makes sense in an age of vinyl records, tapes or CDs.

As we noted back when these lawsuits first came out, some of the copyright lawyers we spoke to couldn't figure out why Spotify would even need a mechanical license in the first place. After all, it's just "streaming" music. It's not "reproducing" the work, nor is it "distributing" the work. Or is it? That question gets pretty murky, pretty fast. You can argue that the process of streaming involves bits being "reproduced" and "distributed" -- but a counter to that is that that's happening in such an ephemeral way, in which the "recipient" ends up with nothing, that it's not at all the same thing. Indeed, this issue came up in an important case a few years ago, concerning the "transitory copies" that were made by Cablevision with its remote DVR. In that case, it was determined by the appeals court that transitory copies are not "copies" under #1 above. That is, they are not "reproductions." Here's from the ruling:

"Copies," as defined in the Copyright Act, "are material objects ... in which a work is fixed by any method ... and from which the work can be ... reproduced." ...The Act also provides that a work is "`fixed' in a tangible medium of expression when its embodiment ... is sufficiently permanent or stable to permit it to be ... reproduced ... for a period of more than transitory duration."... We believe that this language plainly imposes two distinct but related requirements: the work must be embodied in a medium, i.e., placed in a medium such that it can be perceived, reproduced, etc., from that medium (the "embodiment requirement"), and it must remain thus embodied "for a period of more than transitory duration" (the "duration requirement").... Unless both requirements are met, the work is not "fixed" in the buffer, and, as a result, the buffer data is not a "copy" of the original work whose data is buffered.

That said, it did appear that Spotify had made (via HFA) at least a half-hearted effort to secure mechanical licenses anyway, often using the "Notice of Intent" process, which allows for compulsory mechanical licenses if certain rules are followed. Of course, the half-hearted nature of this appeared to have some significant gaps (gaps that HFA later started scrambling to fill).

Either way, earlier this year, we wrote about Spotify settling the big class action lawsuits it was facing, and agreeing to pay out a bunch of money. As we pointed out when that happened, it appeared that no one seemed to really want to fight out the issue in court of whether or not Spotify even needed to pay mechanicals, noting that such a fight would, undoubtedly, get messy.

Well, things are getting messy.

After that settlement, not all composers were happy. Some of them, including Bob Gaudio, from the Four Seasons, sued Spotify over the same issue -- and argued that the settlement of the previous lawsuits was an "empty gesture" given just how massive the copyright infringement was on Spotify. In other words, these new cases seemed a lot less willing to settle. And that means, the issue we never thought was going to get tested in court... is looking like it's going to get tested in court. As first covered by Eriq Gardner at the Hollywood Reporter, Spotify has now directly argued that its service does not reproduce or distribute the songs it streams. From the filing:\

Plaintiffs allege that Spotify “reproduce[s]” and “distribute[s]” Plaintiffs’ works..., thereby facilely checking the boxes to plead an infringement of the reproduction and distribution rights. But Plaintiffs leave Spotify guessing as to what activity Plaintiffs actually believe entails “reproduction” or “distribution.” The only activity of Spotify’s that Plaintiffs identify as infringing is its “streaming” of sound recordings embodying Plaintiffs’ copyrighted musical compositions.... But “streaming” – by its very definition – cannot infringe upon either the reproduction right under 17 U.S.C. § 106(1) or the distribution right under 17 U.S.C. § 106(3). As a consequence, Plaintiffs’ allegations simply do not inform Spotify how Spotify is alleged to have violated the law.

As Spotify's lawyers admit (as we described above), it's clear that streaming implicates #4 and #6 of copyright's exclusive rights -- and it has licenses to cover both of those areas -- but not #1 and #3:

In short, the act of streaming does not reproduce copies of sound recordings or musical compositions, and equally does not distribute copies of either sound recordings or compositions. This is not to say that streaming has no copyright consequence. Streaming does result in a public performance of both sound recordings and musical compositions. See 17 U.S.C. § 106(4) (granting copyright owner the exclusive right to “perform the copyrighted work publicly” in the case of musical works) and § 106(6) (granting copyright owner the exclusive right to “perform the copyrighted work publicly” in the case of sound recordings).... But Plaintiffs do not allege that Spotify has violated their public performance rights, and such an allegation would be futile. Spotify has paid hundreds of millions of dollars to license the public performances of the musical compositions it streams, including those allegedly owned by Plaintiffs, through negotiated licenses administered via ASCAP, BMI, SESAC, and GMR – a group of licensing agents known collectively as “performance rights organizations” or “PROs.” In sum, Plaintiffs’ Complaint, which rests upon Spotify’s “streaming,” leaves Spotify guessing about what it has done that, in Plaintiffs’ view, constitutes a violation of Plaintiffs’ reproduction and distribution rights.

If Plaintiffs are indeed alleging an instance of copying (as opposed to streaming) that allegedly violates Plaintiffs’ reproduction and/or distribution rights, they can amend their Complaint to say so. Depending on what that allegation is, Spotify will advance a number of defenses. If Plaintiffs’ real complaint is with, for example, a particular instance of alleged temporary copying, then, among other things, Spotify may assert a defense of fair use. Spotify may also assert defenses of compulsory license, implied license, negotiated license with copyright owner or co-owner, statute of limitations, and others, depending on the nature of the specific allegations. But Spotify should not be forced to guess and aim its defenses at an unknown claim. Spotify believes that every element of its service is either fully licensed or otherwise permitted by law, and Plaintiffs’ Complaint offers no notice as to what Spotify conduct it thinks violates its copyrights or what conduct would require additional licenses.

Spotify also argues that the point of this lawsuit is really to try to convince composers to opt-out of the settlement from earlier this year, and to join in this action, which may explain why Spotify is taking an aggressive stance here. But this case, should it move forward, could become quite important. Because if a court somehow goes against earlier precedents and argues that streaming actually impacts four out of the six different exclusive rights granted under copyright law, that would be quite amazing.

As we've noted many times in the past, one of the problems with copyright law is that each time a new technology comes along that threatens "the old way" of doing things, rather than rethinking copyright itself, Congress tends to just duct tape on a new right that requires licensing. And that works fine until the next innovation comes along. Now, in the internet age, where things look a little like their analog predecessors, but also with other features, it's no surprise that some copyright holders are trying to argue that ALL THE LICENSES must apply. But if you actually look closely at what Spotify does, what copyright law covers in terms of those six exclusive rights, and what previous cases have held, it does seem likely that streaming should really only implicate two of the exclusive rights.

Filed Under: bob gaudio, copyright, distribution, mechanical rights, music licensing, reproduction, streaming
Companies: spotify

The Music Licensing Swamp: Spotify Settles Over Failure To Obtain Mechanical Licenses

from the how-could-they-not-have-done-this? dept

A year and a half ago we wrote about a lawsuit, filed by musician/songwriter/Techdirt-hater (with a few perhaps surprising exceptions) David Lowery against Spotify, for failing to pay mechanical licenses. As we noted at the time, the more interesting thing to us beyond the lawsuit itself was how it demonstrated what an amazing clusterfuck music licensing is. That's because copyright law has not done a very good job keeping up with the times as technology changes (understatement alert).

Basically, each time a new technology undermines the way licensing worked in the past, Congress ends up duct-taping on some new kind of licensing regime. There are a bunch currently, nearly all of which can be traced back to different technological innovations from the past century and a half. And, then, the internet came along. And it wasn't entirely clear how the licensing regimes of things like radio, television, player pianos, and satellite radio fit into the internet. And, some seem to think the answer is: they ALL apply. At the very least, I don't envy the "licensing" team at the various music tech companies.

In our initial post, we noted that the issue seemed so complex that after talking to half-a-dozen copyright lawyers, no two could agree on what was actually happening with the lawsuit, or even if it was a legitimate case. The underlying issue had to do with mechanical licensing (a type of licensing which, as it's name suggests, goes all the way back to the early days of "mechanical" reproduction of compositions), and we were wondering how it could possibly be that a company as big as Spotify, whose entire story rested on the idea that it had properly negotiated licenses, had somehow failed to properly secure mechanical licenses. And, yet, a few months later, we noted that the Harry Fox Agency, an organization that many companies, including Spotify, Apple and others, use to handle these kinds of licenses, appeared to be scrambling to send out notices of intent (NOIs), which was something that should have happened way earlier.

After Lowery's lawsuit got combined with another similar lawsuit, it's now been announced that Spotify has settled the combined lawsuit and created a $43.4 million fund to pay for the mechanical licenses it failed to obtain properly in the first place. Now, there are still some who argue that mechanical licenses shouldn't even be necessary for a streaming service, but it doesn't appear that anyone has the desire to fight that one out in court, and it's understandable why. Doing so would almost certainly lead to any service making that argument getting slammed by musicians for trying to avoid paying songwriters.

Either way, Spotify has paid its way out of this and I remain baffled by the fact that it hadn't just done the right thing in the first place -- though I'm still curious if the real culprit here is the Harry Fox Agency, and if Spotify and HFA have had a long conversation or two about how this all came down. The real lesson in all of this, though, is that music licensing continues to be a complete murky, swampy mess, almost designed to make it that much harder for licensed music services to exist. While Congress dithers with silly ideas about "moving" the Copyright Office, if it wasn't to actually reform copyright laws, it should start by fixing and modernizing the crazy and overly complex licensing regimes.

Filed Under: copyright, david lowery, licenses, mechanical licenses, music
Companies: harry fox agency, hfa, spotify