Iterative music culture, generative AI and the Web3

A recent project calledĀ TunesĀ used AI to generate 5,000 unique NFTs. Theyā€™re songs, or rather, shells of songs – missing artwork, audio and an artist. Thatā€™s intentional. They serve as prompts for people to iterate on, tapping into a recent trend in the NFT space popularized by another project calledĀ Loot. Letā€™s dive in.

An example of one of the Tunes NFTs

Loot

At the end of August an NFT project calledĀ LootĀ dropped. People could claim Loot NFTs for free by minting – only paying for the gas cost. LikeĀ NFT avatar projects, people did not know what they would get exactly after minting. Unlike other NFT projects, Loot was stripped from everything except text.

After minting, youā€™d get a list of gear. Text in an image. Thatā€™s all.

People loved it, because it felt like an invitation to imagine what you could build from this starting point. Quickly, an ecosystem emerged around the project.

Adventure Gold created a token to set a standard for projects building on Loot in the future. Role creates characters which can equip the Loot. Realms attempts to map out a world for Loot to exist in. And there are many other projects.

Mirror, the writing platform this article was first published on, ran a similar project calledĀ HeroesĀ which would create randomized pen names and identities: text-only.

Itā€™s different from many other NFTs, because instead of selling you something that is finished, you get the building blocks. Since everythingā€™s on a blockchain with smart contracts, everyone can plug in and start building and expanding the project.

What if this idea was applied to music?

Tunes

Eight of the Tunes listed on OpenSea

Tunes started with a similar premise. 5,000 possible songs to mint for free, but without audio, cover art, or an artist. Building blocks.

A bot calledĀ ArtunistĀ was set up on the projectā€™s Discord to generate artwork. People can submit the ID of the NFT they minted and then get artwork generated for it. The results are impressive.

Cover art generated for Tunes by Artunist bot

The project has since expanded to include ā€˜Songs for Tunesā€™ which combines the generative artwork with artists and the music they made, with some being sold as NFTs for 1ETH ($3000~ at time of writing) like thisĀ beat by oshiĀ or this songĀ generated by the AI bandĀ (Twitter).

New games, new economies

The reason why Iā€™m highlighting these projects is to show that there are new games to play, completely new avenues to explore. From the outside, the crypto space mostly gets attention for financial aspects while the cultural aspects donā€™t get picked up properly. When they do, itā€™s almost always in a financialized contextā€¦ If an artist drops a new release on Spotify, we donā€™t say ā€œArtist X is releases album on streaming service that raised over $2.1B in 18 roundsā€ yet thatā€™s exactly whatā€™s happening in the web3.

The actual exciting part is not really the money. Itā€™s that anyone who perseveres can spin up a new project that can tap into any other project connected to the same blockchain. In that sense blockchains with smart contracts like Ethereum are global computers for us all to utilize. The web3 is iterative and music culture in the space is starting to embody that principle.

The first music artists who will never bother with the contemporary streaming landscape are likely already here, experimenting in the web3 and trying out other modes of collaboration, community-building, and are starting to make a living by doing so.

And to put it inĀ Thanakron Tandavasā€™ words: thatā€™s fucking cool.

Image by @tandavas on Twitter

NFT Avatars and Onboarding Subcultures to the Web3

You may have heard of CryptoPunks and Bored Apes. Theyā€™re NFT series of 10,000 unique and randomly generated images of characters. Theyā€™re also 2 of the most well-known examples of ā€˜NFT avatarsā€™ – a trend which has exploded in the crypto space of 2021 and is set to permeate subcultures once they move into the web3.

Hereā€™s how, and why.

PFPs & NFT Avatars

NFT avatar projects are casually referred to as PFPs: ProFile Pictures, since thatā€™s a common use case for them. They most often have up to 10,000 template characters that are ā€˜mintedā€™ by people wanting to own one. Minting costs money (cryptocurrency) and in the process of minting randomized traits are applied, ranging from extremely rare to common. These may be variables like skin colour, headwear, haircuts, shirts, background colours and anything else you may think of thatā€™s appropriate to the project. Many projects will have 10 traits with 10 variables per trait.

Once minted, the character is created and will remain unchanged, so before you mint, you usually only know what the style of the project is, but not what your NFT character will end up looking like. Once all NFTs from a series have been minted, the value of some of the PFPs may increase, since people may try to purchase ones that specifically appeal to them (likeĀ Jay-Zā€™s CryptoPunk).

Being early

The creative web3 space is still relatively small. Many of the more successful projects specifically cater to the crypto community and its aesthetics. As interest has peaked, so have the number of PFP projects that now flood the space. Many struggle to get all their NFTs minted, often resulting in minted NFTs being sold for a price lower than the minting cost, which further slows down the minting process.

Yet weā€™re early. Many types of aesthetics that are popular in subcultures, for example as album artwork, in music videos, or as tattoos, donā€™t necessarily resonate well with those already onboarded to the web3. While some of the aesthetics emerging from the web3 space will go on to become cornerstones of emerging subcultures, a more diverse variety of aesthetics will become popular in the web3 as more people get onboarded.

A PFP project making slow progress on minting seems bad, but if it finds ways to onboard the subculture in which its aesthetics are rooted, it can succeed over time and enjoy a potentially more meaningful success than it could by shilling to speculating crypto bros (f/m/non-binary).

Onboarding subcultures

The social dynamics of whatā€™s happening right now are not unique. The clearest memory I have of subcultures shifting to a new type of internet is from around 2009 when artists started switching out the MySpace Music players on their profiles for SoundCloud embeds. SoundCloudā€™s player was clearly superior. While MySpace limited you to a maximum number of tracks per profile, SoundCloud allowed you to upload 4 tracks of any length per month for free in those days. The benefits were obvious and thus once a few musicians in a genre embedded the SoundCloud player on their MySpace profile, youā€™d see it spread like wildfire through their subgenre.

The type of people who were the first to switch out their player, the innovators, are now onboarding to the web3. Many of them are already experimenting with NFTs and DAOs. PFPs allow them to signal the web3 to others in their cultural space, just like how the SoundCloud player signalled a shift from the age of downloads to the age of streaming. PFPs are not enough though, since their utility is not as obvious compared to an embeddable player that was easily twice as good as what preceded it.

For subcultures to onboard to the web3, there are two main hurdles to overcome:

  • Proof of stake (PoS). The energy use associated with Proof of Work blockchains likeĀ BitcoinĀ andĀ EthereumĀ has made many people unwilling to touch any type of crypto. So, I expect a few factors in the next year will drive more people to the web3: 1) Ethereum switching to PoS; 2) maturing ecosystems around PoS blockchains such asĀ TezosĀ andĀ Solana; 3) more accessibleĀ layer 2 rollupsĀ for more diverse use cases.
  • Usability. It currently takes about 30-60 minutes to onboard someone to the web3 and buy their first NFT, like anĀ ENSĀ domain (see my primer below). They have to open a wallet, verify their identity on an exchange, buy crypto, wait for transactions to clear, etc. Itā€™s hard to figure things out. Itā€™s easy to make expensive mistakes. Many people who are already onboarded forget how hard it is: try onboarding a friend whoā€™s completely new to it. Sit next to them. Walk them through all the steps.

Subcultures and PFPs

A year from now, things will likely be a lot easier for newcomers. Besides work being done on the previous two bullet points, tech giants like PayPalSquare / Cash App, and likely Facebook / Instagram entering the space will help decrease the number of hurdles (at the cost of decentralization).

It will increase the viability of sounds and images that are specific only to certain subcultures and decrease the necessity to make plays that cater to the wider web3 community. So if youā€™ve been hesitating to start a PFP project or a DAO, because you donā€™t think the people are onboard for it yet: start small and do it anyway. Consider a lower number than 10,000, and donā€™t rush, donā€™t shill too hard. Your audience will get here eventually and youā€™ll be the CryptoPunk equivalent of your domain. The OG PFP NFTs of your subculture.

I donā€™t know about you, but Iā€™m so ready for crypto goth, crypto gabber, and actual crypto punk. See you soon.

The web3ā€™s user-friendliness barrier for music

Setting up a Decentralized Autonomous Organisation (DAO) is a clunky affair when compared to more centralized variants. But maybe that’s the point?

I was recently asked when DAOs will feel more like installing an app and there are definitely accessibility hurdles to be overcome for these organisations to become a common phenomenon in music. Here’s my thoughts.

Firstly, I do think we’ll see DAOs become mainstream, but not until some of the least user-friendly aspects are resolved. Setting up wallets, gas costs and treating mobile as an afterthought are 3 points of friction that come to mind. I’ll dive into them here.

Secondly, the point of a DAO is to organise as a group on your own terms, using your own tools, and participating in the value you generate together. Part of that value may be represented as shares in the project (in the form of tokens) and part of it may be cash (in the form of cryptocurrency). That’s complex and usually requires all sorts of documents, contracts and perhaps even notaries. So in that sense,Ā the expectation to make DAOs as simple as installing an app is unrealistic, but further maturing of tooling will definitely make things easier.

Wallets

In order to participate in DAOs you’ll need a ‘wallet’. I recently recorded aĀ 10 minute how-to, that actually turned out to be 30 minutes long, which highlights some of the complexity well.

This wallet allows you to participate on the blockchain utilized by the DAO. It lets you make transactions, interact with smart contracts, send and receive tokens (incl. NFTs), as well as cryptocurrency. It also functions like a single sign-on that you can use to interact with blockchain-based tools, platforms, and communities.

There’s a way around the wallet requirement: you could do things ‘off-chain’ for people without a wallet, which means the organisation centrally records transactions & what people are owed without writing it to a blockchain. Whoever has a wallet can move all their assets ‘on-chain’. Many tools, like Collab.land, already do this by recording your tokens alongside a user ID such as your Discord account until you withdraw them to a wallet that has a blockchain address.

šŸ”® I expect Facebook to eventually integrate wallets into their offering. This means that people would be able to open a wallet on the social network and be able to use Facebook’s one-click login on web3 platforms too. This will remove a lot of friction from the space (although I really don’t think you should be using Facebook).

šŸ”® I also expect banks and financial tools like PayPal & Cash App to eventually have wallets for Ethereum & other popular blockchains. They’re already doing Bitcoin.

Gas costs

Shortly after I published my video about how to set up a cryptowallet, the transaction fees (gas costs) on the Ethereum network climbed to around $50. That means that even if you’re trying to perform something that’s free of charge, you may still have to pay $50.

Gas costs have recently been a lot lower, but no matter the sum, it will take education to explain to people why certain interactions require money on an internet where everything has always felt like free. It depends who I’m explaining it to, but I tend to explain that instead of selling your data to advertisers in exchange for a service, you pay the network directly to process your transaction and host all the data. But how does a musician explain this to their fans?

Here, too, doing certain things ‘off-chain’ can help cut costs, but decentralization gets sacrificed.

šŸ”® Ethereum’s transition toĀ Proof of StakeĀ will help a lot. I also expect ecosystems around other blockchains likeĀ TezosĀ andĀ FlowĀ to mature, which have both found their own ways to keep gas costs low.

Mobile as an afterthought

Mobile phones are the world’s most important personal computer. For most people around the world, they’re the only real personal computer they have access to; that’s just theirs and doesn’t have to be shared.

Unfortunately, many web3 projects highly prioritize the desktop experience. It’s understandable why: that’s the environment in which builders are creating, so especially in this early stage it’s hard to expect everything to be perfect on mobile phones. There’s a thin line between afterthought and neglect however and if we’re serious about onboarding people to web3, we have to do a lot better on mobile.

šŸ”® The size of the opportunity of making the web3 accessible to the mainstream will drive organisations to prioritize mobile. Currently, a lot of the pieces are still being put in place, but I expect a period of growth to follow especially once some of the other clunky aspects mentioned in this article get ironed out. That’s when the web3 goes mobile.

What does it mean for music?

My guess is we’re about 1.5 to 2 years away from things like NFTs and DAOs going mainstream. The biggest obstacle is user-friendliness. DAOs don’t have to be as easy to install as an app, but right now it will be hard for artists to onboard their audience to a fanbase DAO. While artists with ‘crypto native’ fanbases may be able to gain traction with a DAO, the barriers will remain high for most other artists. What those artists could do is tap into creator economy trends and tooling in order to set up the foundation of their community in a more centralized way and then extend it with web3 aspects that may not appeal to everyone straight away. The same goes for event organisers.

Web3 primer: setting up your cryptowallet & buying your first NFT

Recently I’ve written a lot about web3 use cases, from organising raves as DAOs, to 1,000 true fan organisations, to the basics of bootstrapping a DAO.

In order to get more of you actively participating, this post goes into something more practical: setting up a wallet & acquiring an NFT. I’ve set it up as a video:

A step-by-step guide on:

  • How to set up a cryptowallet using Metamask
  • Adding cryptocurrency to your wallet using Coinbase
  • Buying your first NFT on ENS
  • Monitoring transactions on Etherscan
  • Giving your wallet a nice address like basgras.eth or musicxblockchain.eth

Links:
https://metamask.io/
https://www.coinbase.com/
https://ens.domains/

I spent about $50 on gas fees and .eth addresses in order to create this video. This is why I have aĀ Patreon community for MUSIC xĀ whose contributions help me cover the costs associated with running the sites & newsletter. Thank you! šŸ–¤

Tip: if you want to follow along using a cheaper method, you can also use the Tezos blockchain. It uses a ‘Liquid Proof of Stake‘ consensus mechanism, which is more efficient, so lower gas fees and cleaner environmentally-speaking. For Tezos, you can use the Temple wallet and register your domain on Tezos Domains.

Blockchain basics: how to start a DAO

My recent writing has focused on the community dynamics of blockchain-based ‘Decentralized Autonomous Organisations’ (DAOs). I’ve explored:

In this article I will attempt to explain some of the more technical aspects in clear terms for people with little to no experience with these topics. I’ll be diving into the steps outlined in a tweet by Jess Sloss of Seed Club, a DAO that builds and invests in communities.

There’s a comment section below. If anything is unclear or could be worded better: let me know either with a question or by spelling things out more clearly yourself.

How to bootstrap a DAO

People familiar with English-language startup terminology will be familiar with the term bootstrapping: to start something using nothing but your own funds (or in some cases: zero funds). Continue reading to learn how a DAO might do that.

You can’t have a DAO without a great community. I won’t go into that for this piece, but recommend reading How to grow decentralized communities by pet3rpan (before you click out of this website, consider joining the newsletter, so we can reach you in case you get lost in a rabbit hole ;-)).

šŸ“„ Drop an NFT or series (on chain revenue)

I think by now, for many people, non-fungible tokens or NFTs have become synonymous with auctionable digital artworks. This is not incorrect, but it’s a little bit like saying MP3s are music, while actually it’s a technology that has lots of uses in terms of audio encryption. A slightly better way of thinking about NFTs is as collectibles.

Non-fungible tokens allow for tracking ownership, as well as functionality like ‘splits‘ which are commonly used to make sure the original author gets money (in the form of cryptocurrency) every time their NFT is resold. This is done through smart contracts: little computer programs linked to a blockchain database that run whenever certain actions are performed or conditions are met.

Although the most publicised use case is 1 NFT of a unique artwork being sold, there are also countless examples of collectibles where 10 people can buy NFTs that represent identical artworks (e.g. this NFT by musician Sevdaliza). The former case would be described as a 1/1 and the latter as a 10/10 run, like a collectible. A series could be a set of NFTs, like a bunch of 1/1s, multiple 10/10s, or any mix like a 1/1 and a 5/5 drop.

This creates on-chain revenue: value stored on the blockchain that the DAO will use to let the community participate and distribute ownership. That revenue is stored in cryptocurrency.

A recent music-related example of a DAO that funded itself with an NFT sale is Songcamp. With the on-chain revenue, it could afford to cover the fees associated with ‘minting’ (creating) an NFT for the participating artists in its first songwriting batch.

šŸŽ Give NFTs to dope people (on chain community)

On chain community means that you have a way to track, via blockchain, who are the people in your community. Since tokens like NFTs allow people to see who owns them, it’s an easy way to trace ownership back to a DAO (the link between the DAO and the recipient is forever recorded).

  • Step 1: create an address for your DAO on a blockchain by setting up a wallet which allows for transactions and storage.
  • Step 2: create NFTs with that address.
  • Step 3: send the NFTs to addresses of people you want to add to your community.

Now there is a link between your address and theirs, through the NFT. You can see this happening in the above screenshot, but strip away the interface of the auction house and you get something like this.

Here you can see the transfer of a token from one address to another, here indicated as club.eth, which is the same @club from the Zora auction house screenshot and actually also the Seed Club referred to at the start of this post.

A community or service can let you sign in using your wallet (e.g. Metamask) which is a little bit like the type of ‘Single Sign-On’ you’re used to around the web from Google, Facebook, and Twitter. It can then check your wallet for any NFTs or other tokens (I’ll get into this) and grant you special privileges, ranging from simple access to more advanced features.

I was recently lucky enough to get voted into Mirror, a kind of crypto version of Medium, but way more interesting (thanks for the votes!). To participate in the vote, you have to connect your wallet. If you win, Mirror transfers an access token to your wallet. On Twitter that looks like this:

On Etherscan, a tool to read about transactions on the Ethereum blockchain, the above looks like this:

Here you can see 1 address sending 10 tokens to 10 addresses through the execution of 1 smart contract (you can read the code of that contract here). Bonus points if you can figure out which address I hold. šŸ˜‰

Overwhelming? No worries, the user experience is easier than setting up an internet connection or email in the 90s. In the end you just need a browser extension like Metamask’s to log in to Mirror and when it sees you hold the correct token it presents you this simple interface for creating your account:

šŸ›« Launch Snapshot + token gated Discord (gov. infra)

To set up the DAOs ‘governance infrastructure’, you can use a tool like Snapshot to let people submit and vote on proposals, plus you create a community for token holders (I’ve described token gating in the previous paragraphs). The latter is commonly done through Discord.

Here’s an example of a proposal for CabinDAO: a community that is creating a cabin residency program for select creators.

Here the community (or DAO) is voting on a linked proposal. It’s essentially deciding to commit a certain amount of funds (15 ETH) and community tokens to the program.

There’s a list of voters – 15 in total. They’re shown as addresses on the Ethereum blockchain and since Jon Gold has registered his name through ENS, I can recognize him and search for him elsewhere. For example, I can see he used his $WRITE token to join Mirror a few months ago. The votes are ranked by the number of community tokens someone holds (the bottom 13 are cut off). I haven’t looked into exactly how CabinDAO has distributed tokens so far, but usually they’re awarded to early community members and rewarded for participation, contribution, or in exchange for things (or cryptocurrency).

šŸŖ‚ Airdrop ERC-20 tokens (governance to the ppl)

I’ve explained non-fungible tokens already, but haven’t gone into detail about other types of tokens.

ERC-20 is basically the technical standard for token implementation on smart contracts on the Ethereum blockchain. Remember Mirror awarding 10 people with tokens to join their service? It happened in 1 transaction through the execution of a smart contract.

While no two NFTs are alike (and commonly use the ERC-721 standard), ERC-20 tokens are fungible, meaning that they can be interchanged with one another. In simple terms, if I send you 1 $WRITE token mentioned above and you send me 1 $WRITE token, we end up with the same in the end. Trading NFTs would typically leave us with two distinct items.

Through your community’s smart contract, these tokens can give you voting rights or participation rights in a DAO, e.g. access to a Discord server or the ability to vote on proposals on Snapshot or similar.

This is where you might award the buyers of the NFTs with a certain number of tokens created uniquely for your community through aforementioned smart contract, e.g. if it were for my newsletter’s community, I might call them $MUSICX tokens. You’d also give early community members and other supporters some tokens in your community. This incentivises them to get active and start participating in the governance.

This process of distributing tokens among your community is called ‘airdropping’. Now, there’s just one thing remaining:

Use ETH / Tokens to go do cool shit

Like the Friends With Benefits DAO, you could let people buy their way in through exchanging a cryptocurrency (ETH) for tokens ($FWB). This means as a DAO, you have a certain liquidity from token sales. So as a community, you can use tokens to incentivize certain actions (e.g. creating a residence program for artists in a cabin) and you can use ETH to cover certain costs, from renting the cabin, to infrastructure, to perhaps paying a few developers to build your website.

That’s it. All of the above is using the Ethereum blockchain, but there are other blockchains out there that support similar functionality.

Go organise your community and if you’d like to invite me – send me a token at basgras.eth or a tweet @basgras.

Mike Shinoda auction on Zora

NFTs are blockchain’s hottest new use case for music. They should not come as a surprise.

Linkin Park‘s Mike Shinoda just sold a digital piece of art for $30.000 and took to Twitter explaining some of this thoughts in a thread:

“Even if I upload the full version of the contained song to DSPs worldwide (which I can still do), i would never get even close to $10k, after fees by DSPs, label, marketing, etc.”

The ownership of this piece of art is tracked through a non-fungible token on a blockchain. Blockchains are commonly used as distributed ledgers: databases operated by networks of users, like Ethereum. They keep records of any changes to the ledger and can track things like ownership of tokens or cryptocurrency, e.g. Bitcoin.

But so what if a piece of art is recorded into a distributed database? Why the hype?

The current cultural moment is strongly influenced by the pandemic. Artists saw a big drop in income. Streaming revenue isn’t cutting it for most. So the big experimentation began. Artists searched for revenue through things like livestreaming, fan clubs, ticketed virtual meet & greets, online courses, and NFT auctions…

Why are people buying content that can easily be duplicated?

Many a music industry conference panel has bemoaned the fact that people are willing to buy a cup of coffee or bottle of water, but won’t spend that money on a download and instead chose to pirate it (in the days long before Spotify counted 150M paying subscribers). Two decades later and many of the same philosophical debates about the price and value of music continue. Meanwhile, gaming, an industry that faced the same piracy issues as the music industry, pragmatically pioneered ways to get people to pay for completely virtual items.

Gaming gave the ownership of virtual items a valuable context. People who spent many hours a week inside games would find value in virtual real estate or vanity items that translates into real world currency. This is not something recent. In 2013, someone paid $38,000 for an in-game item in Dota2 – an item which doesn’t improve a player’s performance, but just makes them look cooler. In 2010, virtual real estate by the name of Club Neverdie in online game Entropia sold for $635,000.

Now, ten years later, we’re seeing the same dynamic emerge for music. Owning an NFT doesn’t necessarily mean that nobody else can enjoy the work of art associated with the token, much like with physical art that’s exhibited. With the emerging metaverse, some are expecting NFTs to become its property rights.

NFT x Metaverse

The idea of the metaverse essentially boils down to a virtual shared space. One prominent example of this concept is Roblox, which is a gaming platform in which people can build their own experiences that are all interconnected through Roblox’ economy (its currency being Robux). Another is Fortnite, which has some of the ingredients already, but hasn’t yet developed a marketplace with low barriers to entry like Roblox has. Despite that, one of the best primers on the topic of the metaverse is the below interview with Tim Sweeney, CEO of Epic Games, which owns Fortnite.

It’s the convergence of various pandemic-accelerated trends (VR / XR, virtual economies, crypto) and the expectations of people in these domains that is currently driving NFT art’s success stories ($750,000 CryptoPunk sale, Panther Modern‘s $666 sale, virtual critters for $100,000 a piece). If you want to know what the future holds, look at what the smartest people in the room are doing, because they’ll be the ones building that future.

12 years after the initial release of Bitcoin and the world’s introduction to blockchain, crypto is starting to emerge as an anticipated layer of connectivity for transactions occurring in the metaverse. With a market cap higher than Facebook at the time of writing, Bitcoin has made many early adopters very rich (as have other cryptocurrencies). Besides figuring out how to build an infrastructure in which they can effectively use their blockchain-riches, we’re seeing this money flow into other spaces, like art (and soon Tesla).

Simplified: to understand some of NFTs’ success, you should look at the crypto space as a metaverse without an interface that looks like a video game. The participants of that space are still players: they’re building their own world, their own infrastructure. They care about what they look like in that world, just like how people in virtual worlds care enough about their looks that they’re willing to buy in-game currencies like Robux (to the sum of billions of USD in 2020). Owning art is cool – it gives you standing in your micro-community which is part of larger meta-communities (e.g. a gaming clan is a community inside the community of one server of a game, which is a community inside the global player-base of that game).

And sure, there’s altruism too, because it’s cool to support art. However counting on altruism tends to spawn panel discussions to compare bottles of water to digital art. Focus on non-altruistic value.

Join the MUSIC x newsletter.