Firstly, I’d like to echo the others and say that this is a fascinating post @edelsohn.
As I was reading through, the first thought that came to my mind was dynamic NFTs, which Chainlink has been enabling: What Is a Dynamic NFT? | Chainlink Blog. This is just a component of what you are articulating here, but it is an example of how an NFT might change and adapt, while still remaining the same NFT.
Broadly speaking, I recognise the problem you describe in this post and agree with the direction you propose when it comes to a solution. I have been thinking along similar lines but in different contexts – my own area of expertise focuses on people, rather than objects, but I think many of the same principles apply. I especially appreciate the emphasis you place on the importance of observation as well as perspective. Overall, I think that at a theoretical level all of this makes sense, however, the real challenge will be in how this translates from theory into reality.
When thinking about how I might add value to this post, I thought it might be useful to analyse the similarities (as @jmcgirk noted), between this discussion and one we are also having about soulbound NFTs as a way to capture and use identity on the blockchain (Discussion Post: Are Soulbound Tokens the best way to create a trusted layer on the blockchain?). I think this serves as a good practical example of the challenges you are talking about, except that it focuses on how to represent people on the blockchain, rather than objects, meaning that we talk about an NFT that changes with the object (person) it represents.
A ‘soulbound NFT’ can generally be described as a framework for attestations that encapsulates the digital representation of a person on the blockchain. A key problem is that we must establish both the identity of the person, as well as attributes about that person that are both objective and subjective and relate to social, human, and financial capital. Most importantly, these attributes will change over time. To do this, we need an oracle and smart contracts to create, maintain, and update attestations. An oracle can attest to a person’s real identity using existing government documents such as a passport or driver’s license, and other objective factors such as qualifications and income etc. However, when it comes to subjective attestations (see Kate Sills critique of soulbound NFTs), secure smart contracts can provide a credibly neutral mechanism by which people can make attestations about each other.
Do you think that soulbound NFTs, when described like this, provide a practical example of the problems and solutions you describe? There is more depth to your analysis, but I think it maps into this context pretty well.
I also have a few questions/comments about the key points you made.
- An object represented in the Metaverse must be arbitrarily and dynamically divisible or fusible with other objects on demand, as determined by the “laws of nature” in which the object exists in the respective Metaverse.
I am interested in why you choose to use the words ‘divisible’ or ‘fusible’ here. This may simply be a product of different backgrounds, but the words I would lean to are adaptable and transformable. Perhaps this is just semantics, or am I not fully understanding what you mean? I think the core concept you are getting at here, is that the NFT which represents an object must be able to change (adapt or transform) in accordance with both the object it represents in the real world, and according the “laws of nature” on the blockchain.
I think we could also have a discussion about how we define the laws of nature in this context, as they are presumably determined by people, and therefore driven by social consensus, rather than us discovering them through trial and error such as in the real world. In this sense, there are limitations to analogies between the metaverse (something we create) and the real world (something we exist in, and discover things about).
- NFTs should be created and correspond to the coarsest granularity that represents the object and its ownership necessary for the required, observable action.
I particularly like how this principle maps onto the people side of things, as I think it could be used for privacy related purposes which are a huge challenge. “An attestation about a person should be created and correspond to the coarsest granularity that represents what is being attested to, and its ownership necessary for the required, observable action.”
- The relationship between the representation or materialization of objects and the NFTs on the Blockchain must dynamically adapt to the ownership and the actions being performed; it’s not always a one-to-one mapping.
- Each NFT should correspond to the maximal complete subgraph of the components of the object that is necessary and sufficient for the operation to be performed on the object and the ownership of the components.
I am intrigued by the tension between representing the ship of Theseus as a single NFT on the blockchain, or representing each plank of wood as an NFT, and the ship itself becomes a composite of those NFTs. Do we then also represent the trees that are cut down to create the wood as NFTs? Would this NFT be destroyed or transformed during this process? This could be an interesting way to systemically account for previously ignored externalities.