If you are following our NFT self-experiment, you already know what an NFT is and what it stands for. To recap (and add some technical perspective), an NFT is a unique assignment of a specific number and further data, such as an artwork or a link thereto, to an owner (or their public key) in the blockchain. It is unalterable and unchangeably stored on the blockchain. For the NFT to be transferable and to be able to provide information on the current owner of the NFT, the assignment must be managed by a program that creates – or "mints" – the token. For this we need a "smart contract".
What is a smart contract?
A smart contract is not a contract in the legal sense, but a program that runs on the blockchain. Smart contracts have two essential characteristics: they can receive and send transactions, and they are unalterable.
A smart contract basically runs in its own wallet on the blockchain. It has its own private/public key pair and can provide, read and store data. Interactions with a smart contract take place via transactions with the wallet of the smart contract. These transactions contain corresponding commands that are "understood" by the smart contract. The specific number of a token, the address of the assigned owner of the token and other data of the token or the smart contract are saved in the "storage space" of the smart contract. For a smart contract to be "fit" to be used to create an NFT, the smart contract's code must support specific functions (and "fit" in the respective blockchain where it will be used). For the Ethereum blockchain, a unified standard for NFTs is precisely defined under "ERC-721.6". The standard enables a reliable and uniform communication with the program / smart contract so it is clear how an individual or the NFT platform/blockchain can interact with the NFT.
Of course, someone with the requisite skills could program a smart contract entirely by themselves, which when using the Ethereum blockchain follows the ERC-721.6 standard.
An alternative, however, is to use one of the many existing templates of smart contracts that are freely available and accessible to be used in online libraries under open source licences and add some individual data/codes, such as the name of the token, functions of transactions and potentially a link to the artwork that is assigned to the token.
Legally speaking, the smart contract might be copyright protected
From a legal perspective, questions have arisen: As with any other software, it cannot be ruled out that a smart contract as a program is sufficiently individual to be protected by copyright. This does not change fundamentally even if the software is offered publicly for free use under an open source licence. But what do the licence conditions for the open source products contain? Do they sufficiently cover the intended use? These are only rhetorical questions at this point, because the conditions under which open source software is offered for free use are quite different. So here is just a cautionary note. The answers will depend on which product the user ultimately chooses.
Using an existing smart contract
Still, both options (full and partial programming) have one – in our view and for our self-experiment – significant disadvantage: the entire source code of the smart contract would have to be stored in the blockchain. Storing this amount of data would incur considerable costs and consumer a lot of power, making it neither cost-efficient nor environmental friendly. A third option is to use an existing smart contract. Some providers allow minting of new tokens via an online form.
For our NFT self-experiment, we chose to use an existing smart contract, specifically one from Rarible, as it was easily accessible by our wallet, also uses IPFS (more about that below), follows the ERC-721.6-standard, and is simple to use. Again, general terms and conditions had to be accepted. We checked them especially with regard to whether the use of this software would grant any rights to the work of art to be uploaded. That was not the case here. Nor did the typically extensive General Terms and Conditions contain any unexpected special provisions. It is pointed out very clearly that the sole responsibility for the "unique Collectible" created with the help of the tools offered lies with the user who uploads the "Collectible" ("There can be no guarantee or assurance of the uniqueness, originality or quality of any Collectible or Collectible Metadata."). That seems perfectly fair. And of course, an artist who transforms his artwork into an NFT or links it to an NFT with the help of such services must take responsibility for not infringing any third-party rights.
Nevertheless, we became aware of a particular legal aspect of this new distribution method.
The danger of misleading advertising
If an artist offers an oil painting in a gallery and the gallery owner physically hands over the artwork to the buyer, there is no doubt about what the object of purchase is. Even in the case of online purchases of an original oil painting – at an auction, for example – the usual illustration and description of the work in the printed or online auction catalogue makes it clear what is being sold. But is this also clear if an NFT was offered for sale as a "Collectible"? Does the buyer then know exactly what they are getting? Doubts could arise here, and it may be assumed that newcomers will not have a clear picture. Some may equate the NFT with the work of art. They might be disappointed when they discover afterwards that the NFT only contains a link to a work that can be downloaded digitally somewhere, but they neither receive the file (in high resolution) nor a printout, and certainly no licence to use it. There is definitely a risk of misleading advertising here. This could also consist in the omission of the required clarification. The legal consequences of such unfair behaviour, including claims for injunctive relief, damages, etc. are evident. It is therefore up to the seller to clearly describe the content of the NFT in order to avoid misleading the buyer. This should also be kept in mind when promoting NFT sales outside of the platform. This obligation is particularly relevant with the emergence of this new technology, because it must be expected that the average consumer will have little concrete knowledge about it and therefore a high need for information in order to avoid purchase-relevant misleading statements.
What data is packed in a freshly minted NFT?
In principle, it is possible to include an entire data set of an artwork (i.e. the GIF in our case) in an NFT and thus into the blockchain. Still, in view of the high energy consumption and transactional costs, for our self-experiment we decided to limit the data in our NFT to a minimum.
Besides the owner's address and the token ID, only a link (more precisely in our self-experiment a URI) to the actual artwork is stored. In this link, which contains the metadata of our NFT, we further included the name of the artwork, added a link to where the artwork is stored (instead of a description) and formulated as a short licence agreement:
[click to enlarge]
URI, hash – who?
Our aim was for the NFT to remain unambiguously assignable and permanently linked to the artwork and the associated data. The solution was to have a hash value (a unique and reproducible alphanumeric value from a specific data set) calculated from the artwork. If the algorithm used is known, anyone can check whether a particular work of art is assigned to the respective token holder, as long as the hash value is stored in the blockchain. The Interplanetary File System (IPFS; a worldwide peer-to-peer network for decentrally storing public files) calls up / reaches data not under their storage location, but under its hash value, such as the link to our artwork and associated data. The hash value forms our URI (Uniform Resource Identifier). This has some advantages:
- the link to the metadata (including the artwork) is uniquely identified as a hash value (URI);
- if the file should ever be deleted from the IPFS (this usually does not happen arbitrarily, but whenever a file is unused for a long time), you can always reload the same data in IPFS and the link becomes retrievable again (because the hash value of this data is identical);
- the hash value that is noted/secured in the NFT can prove a data record (in our self-experiment, the data record of a specific work of art and that it's linked to the NFT).
This may be especially relevant if the owner of the NFT also owns the digital work of art or as a way to prove that the digital art was in fact created by the named "artist".
Voilà, our NFT was created – or freshly minted, as the NFT pros would say!
By using an existing smart contract and inserting a bit of data about our artwork, we found what we thought was the best way to store the artwork and its associated metadata and identify it in the NFT – as a URI. No serious legal hurdles or uncertainties have arisen at this stage of the self-experiment. However, in the case of a new type of product, such as art-related NFTs, and new distribution channels, such as the offer of art-related NFTs via online platforms, it is important to inform the customer exactly what they are getting for their money. A clear description of the NFT and its content is necessary to avoid claims of misleading advertising.