Tony Lashley is the founder of Marine Snow, a new approach to social networking and streaming for music fans where users can both listen to and stream music on an independent platform devoid of corporate interests or marketing. He's previously worked at Spotify, SoundCloud and for Frank Ocean’s BLONDED, and with a history in curation and experience working with data, he aims to create a more equitable and immersive music experience for both artists and fans that maintains value and appreciation for music as an art form.
Last year, Lashley published an eye-opening essay called “Curatorial Governance” discussing how tastes are shaped and value is ascribed to music online, in the hands of crowds, experts and platforms. How much do experts matter anymore? Should the value of something be judged purely by fans? How can the music industry create models of compensation and value that better serve both artists and fans? And how does Web3 figure into all this? Music journalist Andrew Ryce spoke with Lashley and discussed how the very idea of curation itself is constantly changing and shifting — hopefully for the better.
How do you think Web3 — or, more specifically, a DAO — could change the concept and execution of curation?
I think curation, for a really long time, has been top-down. The idea of a very small pool of people deciding a lot of what people consume is never really a good idea. Injecting the voice of the people, and allowing for individuals who maybe would be talented curators on their own, to have the opportunity to pull new artists that we never would have otherwise been exposed to, and creating systems of incentives to reward people for doing that is really, really powerful. With Web2, in the past, if you were to find a new artist, you might get, like, digital clout as an incentive. And that's not really a good incentive, for a variety of reasons. So allowing people to be rewarded and capture some of the value that they actually create is really important, and allowing disparate voices to enter the fold through people who otherwise wouldn't be curators… That's a unique opportunity.
What are some examples of the good incentives that you're talking about?
The space is pretty nascent right now. And I don't think that the way it is today is how it will be, or should be, tomorrow. But think people like Haleek Maul rising on Catalog... he's a really talented artist who was once signed and then was unsigned, and I'm not sure that he would have ever surfaced on a Spotify playlist or any other medium. So in terms of artists like him getting an opportunity, the rewarding is important — the systems of rewarding people who bring in work. It's kind of like scouts, or a street team, in the music industry, but putting the hands and/or the power in the hands of anybody who wants to be in that role rather than choosing those people from the top down, and actually paying those people a fair amount.
Do you think it's healthy to have that kind of approach where you're rewarding people — is that gamifying it?
I don't think gamifying is an inherently bad thing. I think it just depends on what direction it goes in. I'm a big believer that most of these things are inherently neutral. And it depends on the values that imbue the actions that occur within those systems. If you're talking about gamification of getting healthy or learning a new language, that's not not a bad thing, right? It helps align the short term and long term, using game mechanics to have you feel a sense of short term reward to help you along for the long term. It's really just about: what are the long term values that you're aligning towards? If you can align towards the right values, then it's a good thing. You're aligning towards the wrong values, like free market values — let's just say popularity — then I think it's a bad thing.
In terms of art and DAOs, are we still operating in a free market value situation?
I think so. There's a big libertarian streak that exists in crypto and Web3, that inherently thinks the free market is objectively good. Just like how I think Web3 is neither inherently good or bad, I don't think that free market values are good or bad, per se. I just think you need antidotes to them — like, these values already exist, so you need to counterbalance. I also don't think that something being popular makes it good the way that a lot of people in the world do, where they just associate fame and celebrity and popularity with something being "good."
If we're going to use these structures to curate and theoretically support artists, if the goal is not popularity, then what is the goal?
I think it depends on the institution or the set of individuals. I think it's okay for different individuals to have different goals, we don't need to all have the same goal. My personal goal is to shine a light on things that I think are historically, critically, or culturally important that haven't been assigned the proper value… but "proper" is pretty subjective. Hopefully, collectively, you can figure out how to value artists like Sylvester — people who've had more influential roles in music history than most people realize — or also hyperpop artists who are really important to 19-year-olds that don't get enough shine? I think measuring value in terms of downstream influence is really what I'm interested in, as opposed to just quantity of consumption.
So value to you is also measured culturally, not just in terms of finance or views and clicks.
I think it's measured historically, culturally. But ideally the goal is to align financial value with all of those things. And I do think in the long arc of time, or the longest arcs of time, financial value tends to move closer in those directions. Not perfect, right, but eventually. I don't necessarily define value as being purely financial, but my goal is to align the non-financial forms of value.
With your example of Sylvester and trying to bring more attention — and therefore, more value — how do you quantify that value? How do you put that on paper? If you're talking about influence as opposed to money or sales?
There's something in academia that is definitely an inspiration for me, the H index, though it is not perfect. An H index is a measure of an influential paper and what it does. Let's use CRISPR as an example. CRISPR is probably one of the most advanced and important technologies of the past 20 years, but we don't measure its importance by the number of people that have read the paper, right? We measure it by the number of papers that cite that paper. And the number of papers that cite the papers that cite the paper. If you can show the downstream network map, or family tree, of all the things that a work influenced, then I think that's how you begin to quantify it. An H index is just a way of assigning a value to a paper and then assigning a value to the author of a set of papers. So I'm very interested in how you measure that influence. It's a pretty hard thing, but say Spotify has some of this data, and they never actually chose to really use it, but the economic data they got when they acquired the Echo Nest — they use it to measure related artists. I think you can measure influence to some degree, and then inform what you curate based on that.
So you could also define this kind of vague or intangible value with data.
I don't think it will be 100% defined by data. And maybe the data is discussion, which is why I use Echo Nest as an example, because part of what they were doing is pulling information from the blog universe. And Hype Machine is similar. I think the data is going to be quantitative aggregations of qualitative pieces of information. People talking — and then the distillation of people talking.
Talking about music specifically, you had a really good example with Playboi Carti in your essay and I'm wondering, maybe a few years into the future, as far as you could see via changing trends — where do you think curation is going? And is who holds the power changing at all yet?
I'm trying to be the change that I wish to see. Ultimately, I think the way that the world is, is that there’s a synthesis of those various sources of information. I don't think we're going to enter a world where any one way wins the day. But I think we are starting to wake up to the contemporary over-indexing towards the voice of the crowd, when it comes to the distillation of that being an algorithmic culture and understanding. Take Facebook for example: their committee to determine whether a post should stay up, or all of these social media companies in the last election having to reckon with the idea, "Maybe we shouldn't just be laissez faire." Or, “Maybe we need some other sources of information to find truth other than the voice of the crowd.” And I think synthesis is going to continue to be the name of the game. In the grand arc of history, compared to the past, the crowd will continue to get more power. But in the recent arc of internet history, I actually think experts are going to get more power, but it's just going to be a different type of expert. It’s going to be an expert through the proving of data — an expert through action, rather than an expert through appointment.
Do you think as more and more people get online, that we are now searching for some kind of expertise to guide them when everyone is showing their opinions all the time?
I think in an era of media overabundance people are searching for a way to distill the overabundance, and it tends to be through voices that they trust. So yeah, there's filtering, but that can also happen through systems. I don't think it's any one group of people, whether that's critics or people just relying on their networks or just on their own intuition. It's all those things at once, building systems that synthesize and create repeatable processes for synthesizing the best of all those sources of information. I think that's what people are really craving. People are not craving just expert opinion because they also want to know what the streets are saying, and conversely they don't want to know only what is doing well on the Billboard charts. They also want to know what people who really care about music think is good.
So what would an ideal synthesis look like to you?
Without spilling too much of the tea in terms of what we're working on, I think a really key part of it is creating systems that allow for the continual appointment of new experts. And defining those experts in a variety of different ways, whether that's breadth of consumption, or time spent researching. And maybe if you can put numbers to that, then you can appoint, like, a 17-year-old in their basement as an expert, because they have done all the work. Even if they're not Andrew Ryce or Jon Caramanica, let's say they are like the future Andrew Ryce or the future Jon Caramanica. And so how do you identify those people much earlier and plug them into a system? It's an interesting question to me. How do you create systems of voting and delegation that allow people to have their individual voices heard, but also [that] are not swayed by cascades of information where one person says something and so everybody else says something? I think voting systems of delegates, and systems of creating new experts are important early steps. And then from there it's about how do you weigh it — like, the quadratic stake versus the delegate stake. And then it's just like 20% this, 30% that, and tweaking the numbers until you get something that you feel works.
Does Web3 offer these solutions, or a place for these kinds of systems to exist?
I think being on-chain is important in terms of data sources for creating experts. Because maybe you can define an expert as somebody who has been to two Boiler Rooms, plus has listened to these artists for x amount of time plus has read Pitchfork for this amount of time plus has crafted their own. Or Medium or Mirror posts that are this many words, plus have engaged in this or that discourse. Any one single source of truth to define expertise is not going to work — you're going to need to pull it in from multiple places to really capture what it means to be an expert. I think Web3 presents a really interesting opportunity because it's decentralized.
Do you think these idealized systems, or even the baby steps that people are making towards them, could be applied to a music event or festival model?
I think it's about standing for something. You can have a corporate brand, even, and still stand for something. I think there's going to be a mixing of expert opinion and the voice of the crowd in the curation of an ideal festival. And if you can state a set of values that comes from experts and then use the voice of the crowd to find the examples of things that align with those values, I think it's a really powerful way of combining those two.
To use Refraction as an example, more and more people are joining the DAO and it will be a mixture of the experts and the crowd as the crowd gets more involved. But it's also not like all decisions are sourced directly from the crowd — the curators are still involved. Which is closer to what you're talking about, where it's a synthesis of things, right?
Let's say Refraction wants to curate based on a general value, like internationalism and global music. Then there are specific values: maybe you need more South Korean artists, as one example, because they think South Korean artists have had an interesting outsized role in global music altogether. You can set that value as an expert, but I'm not South Korean. I don't speak Korean. So I'm not going to actually find the South Korean artists the festival needs. I think these questions are particularly acute when it comes to globalization, and language in particular, because if you look at the critical landscape, it's all super English-language driven. And what we define as good or having value is very Anglocentric, even in music today. If you look at Pitchfork or places like that, they're starting to touch Spanish language music a little bit, because it's reached escape velocity — they have no choice. But what's popping in Turkey, Morocco, South Korea, Cote D'Ivoire? I think that presents a good opportunity to set the value, and then allow members or other people to go out and find things that represent that value. If you have a fraction of South Korean or Korean members, they can find what's good in Korea, and you'll be in a much better place than if it was just, like, 20 people in North America and Europe deciding what was valuable.
Music events, or even an album release, are all temporal things. I'm wondering, for you, as someone who is interested in history and influence, how do you measure value and keep it long-term beyond just those flash points where there's a release, when people are talking about it because it's new?
I would actually argue that that's not the right approach, as a festival, to capture short term value. I booked concerts when I was in college and I think a really powerful strategy for us was to book artists that we thought were going to be long-term valuable, because then we could point back a couple years later and say, "Remember when we brought X? Look how big they are now," and it made it much easier to sell whatever we want. You have actually the most power when you can point back and say, “Look, we booked Flying Lotus—his first college headlining show ever,” now, and two years later, we can say, "Trust us." I think it actually makes it much easier if you can focus on long term value to eventually capture value in the long term. Right? If you're constantly trying to optimize what's hot right now, you have to get somebody to believe you when you say this is hot. And you won't know if that's true or not until later anyway unless you've heard of every single artist, whether that ‘later’ is two days later or two years later.
How can artists and curators — because they also help and participate in this — create and maintain value beyond PR cycles and news cycles for artists?
Artists that are really focused on the long term of their careers and building worlds and doing what they want to do, agnostic of what the public perception is at any given time, as long as their vision is clear and these metrics are clear... I just went to the Drain Gang show in New York, and I was talking to my friend. I was pretty emotional going to that show. They've been making music for ten years. And what struck me was that their fans are like 19, 20, 21. And I'm almost 30. And I realized that I was their age when I first started listening to those artists, but they didn't have any of the popularity and notoriety that they do now.
I think people don't realize that they've been making music and doing their own thing for a really long time. And what I admire about those artists is that people revere them as individuals, they also have reverence for them as a collective, as Drain Gang. And then they also have reverence as a part of an even larger world of sadbois and Swedish hip-hop. And they have managed to drill down and build worlds at every level in a way that I find impressive. They're super independent, they're not trying to make as much money as they can as quickly as they can. It's very clear to me that they are interested in making art that resonates with them, and they don't really worry about anybody else. And so, paradoxically, I think the way you achieve long term value by the market conditions at any given point in time, is having your own values that you can espouse and perpetuate through your music — I think that's ultimately what gets people to really care.
To use the language of value, it's almost like investing in your art through time and effort, and [it] eventually paying off, right? But ignoring the market — it's not a stock market thing.
Bingo, that's exactly right. It's not — to use a stock market metaphor, although that can be very perilous — it's not worrying about quarterly earnings, but thinking about the longest arc of your career or value. What is the longest arc? In theory, it should be infinite. And what I mean by that is, you should never expect people to love it, and I think if you never expect people to love it, eventually people will love it if you're continuing to get better. And you're continuing to espouse a set of values and aesthetics that you think are meaningful, that are different compared to what's out there today.
Like you said, everything is nascent, and there's a lot of promises that haven't been seen to completion yet. Do you feel like, especially with Web3, that we are heading towards an improvement on the current quo when it comes to valuing art and curation?
Net neutral. Neither an improvement [nor] a corrosion. I think there's some good things, there are things that are good. There's some things that are bad. And I hesitate to use a very broad generalization to describe what is the agglomeration of many different sets of projects — I think some are good, some are bad. I think with Web3, what people need to keep in mind is they don't want to fall into the perils of Web2 or any other system because I think it's a little bit unfair to say Web3 is so much better versus past systems. As long as the values are outside of free market values, particularly in a space that's so short term speculation driven and financially driven. It's very easy to focus on short term value, because they're what's gonna immediately give you the highest ROI, but not maybe the highest ROI in the long arc of time.
Artists Connie Bakshi and Ellie Pritts reflect on how AI influences their work and perspectives on storytelling, identity, and the evolving role of the artist following their curation in Livepeer x Refraction's Limitless AI exhibit in Bangkok.
