Guest post by Jesse Walden: What a blockchain for music really means
ED note: Jesse Walden, co-founder of Mine and Mediachain, will be speaking TODAY at Music 4.5: Blockchain and the trust issue. The blog post below has been reprinted with permission from the Mine Labs blog. To hear Jesse and our stellar speaker panel dissect blockchain as it relates to the music industry, get your tickets to join us today at the offices of Reed Smith, 599 Lexington Avenue, New York.
What a blockchain for music really means – dispelling the myths and explaining how we can actually build one by Jesse Walden
Before starting Mine, I co-founded Cool Managers, an artist management firm where I represented independent musicians including Solange Knowles, Blood Orange, and Majical Cloudz. Through that work I became entangled with a host of problems to do with music data, which D.A. Wallach eloquently summarized in his Bitcoin for Rockstars:
The problem is simply that no central database exists to keep track of information about music. Specifically, there are two types of information about a piece of music that are critically important: who made it and who owns the rights to it. Right now, this information is fiendishly difficult to track down, to the great detriment of artists, music services and consumers alike.
Cool Managers has spent untold hours corresponding with labels, publishers, and performing rights organizations (PROs) about ownership information. In one case, a stakeholder discovered song ownership percentages totaling 107% out of 100%:
Even after reconciling with 8 other parties by email, a massive problem remained: there is no good way for anyone else to know the correct ownership percentages of the song.
This is because that information doesn’t have a globally referenceable home. Instead it is trapped in proprietary databases, spreadsheets, email inboxes and long-form contracts maintained by separate organizations.
This affects the entire music supply chain, and most critically, digital service providers (DSPs) who are trying to pay artists and rights owners.
Spotify recently paid a $30M settlement to resolve a lawsuit with the National Music Publishers Association for royalties withheld because they didn’t know who to pay.
The “Global Repertoire Database” initiative was started in 2011 to aggregate ownership data in a central database, but infamously failed in 2014 after millions of dollars of investment. Sure enough, no party wanted to cede control to a new entity and the resulting political friction ended in failure.
Blockchain technology is what enables Bitcoin to allow financial exchange without a middleman. It is effectively a decentralized database where participants follow a protocol to record the ownership of tokens of value and their exchange, without the need for a central entity like a bank to provide trust.
An imaginative person will jump to extend the metaphors of such a system to other domains, and of course to the music industry. A music blockchain would be a single place to publish all information about who made what song, without having to trust a third-party organization.
However, before contemplating such a solution, it is important to distinguish between two distinct but often conflated problems in the music industry—because one must be solved before the other.
Data vs. Money
The data problem leaves music platforms with the question: whose song is this? Bad or missing ownership data is the result of poor data hygiene and fractured databases. It prevents artists from getting paid and makes life difficult for platforms trying to pay them.
The money problem leaves artists with the question: where is my money?This is a result of opaque accounting processes that obscure calculations from artists and their accountants.
Some have imagined solving the money problem using cryptocurrencies and smart-contract platforms like Ethereum to enable more efficient, standardized and transparent payments, but these technologies do not address the data problem of knowing who to pay.
If we want to enable maximum value flow and creation, we’ve got to solve the data problem first. Given that context, we should view a blockchain solution as a simple metaphor for shared, networked, media metadata.
How we can build it
A shared metadata network needs the following properties in order to solve data fracturing in the music industry:
A Decentralized Network
A network enables multiple participants to share data in a single logical space. Decentralization obviates the need to assign a central gatekeeper, removing the political friction of centralized approaches like the Global Repertoire Database.
Unique ID Resolution
Search enables two or more organizations to discover and share a common identifier for a song. The identifier can be random so long as it can also be discovered by alternate IDs such as ISRC, ISWC or other internal fields or keywords.
A modular and lightweight translation system would enable interested parties to communicate, while preserving their proprietary formats. The network becomes incrementally compatible over time, without requiring agreement about a mandatory new standard:
All data published to the system must be signed by the contributor using cryptography. This ensures data is attributed and can be filtered by identity. For example, you could query for data from whitelisted entities and ignore data from unknown participants.
Scalable, cost efficient and performant data storage
The innovation of blockchains is enabling distributed consensus without needing to trust a central authority. This is possible at the cost of performance, making blockchains like those powering Bitcoin or Ethereum actually terrible databases for the scale of music metadata. Writes take a long time, capacity is a small, and each update has a financial cost. Every participant must replicate all contents of the chain, even if there is useless data. While this is important to prevent “double spending” of digital currencies, it is unnecessary for tracking the chronology of media metadata. What’s needed instead is a decentralized datastore capable of scaling at low cost, while preserving the complex relationships and revision history of structured media metadata.
Industry orator Benji Rogers has been a loud advocate for “minimum viable data,” and what he’s talking about is the answer to the question: whose song is this? Ownership information is the only data needed to enable a wildly more efficient marketplace for artists, DSPs and the music industry.
Who should participate?
The demand for music data is evident, but the incentive to supply it is less clear. A number of incumbents benefit from the status quo, but can we look to certain innovative music services as catalysts?
Royalty Collection & Asset Management Startups
They can take advantage of a shared metadata network by offering entrants best in class tools to work with it. An analogy is the success of Google’s Gmail, which gained a large market share by offering superior spam filtering and search tools on top of open email protocols.
Platforms like Spotify and Soundcloud have an incentive to find a reliable, long-term solution to the fractured data problem in order to avoid future lawsuits. Spotify seems to be leading the charge, having recently committed to “fix the global problem of bad publishing data once and for all”. They also have the scale and technical resources to ensure the availability and operation of the network.
Finally, a shared metadata network needs trusted arbiters to facilitate conflict resolution. Today this function is performed by PROs and other services that administer copyrights. It’s not illogical to imagine shrewd organizations of this variety choosing to focus on data auditing, privacy and resolution of their client’s claims.
Where does this get us?
With these properties and just some of the participants on board, we would already have a compelling demonstration of what a decentralized music metadata network looks like for new registrations.
To scale past this stage, the network must reach an inflection point where there is more cost to keep existing records off the network than to get them on it — a network effect.
There is a historical pattern of the media industry betting on closed systems again and again—dating all the way back to AOL vs. the open internet — and losing. We can do better this time.
Much like the open internet enabled powerful network effects for social platforms, an open metadata network can enable network effects for media.
Today a developer can code an app in a few hours, but it can take months to legally add content of any kind to it. An open data network could containattribution for a viral GIF, creative commons licensed journalism, as well as information needed to license a song. Further, if that licensing information were machine readable, another developer could automate payments for the content using cryptocurrencies and smart contracts.
At Mine, we believe that an open, decentralized data network is an apolitical venue for media organizations, internet scale platforms, creators and innovative developers to collaborate, innovate and build a new generation of media applications that play on fair terms.
Mine is leading the open source development of Mediachain, a decentralized data network that aims to make it simple for organizations, creators, and developers to share and reuse information about creative works.
As a shared metadata network for music, Mediachain offers a uniform interface to data contributed by multiple participants with no central authority.
If two parties have information about the same song, Mediachain’s resolver allows them to share a common identifier for the work. A modular data translator allows participants to preserve internal data formats while communicating with others, and a cryptographic identity layer enables filtering data by contributor.
Because Mediachain is open source and decentralized, all participants remain in control of their data and there is no central point of failure.
You can read about how Mediachain works here and dig in further on our blog and Github — and if you’re excited about shared media metadata and want to learn more or participate, please say hello [at] mine.nyc or join us on Slack.