A minimal NFT, with more tokens than atoms on Earth | Oct 2021

You may have heard the almost deafening buzz around NFTs. Curious, I grilled Harang Ju on what’s getting people so excited. A lot of it failed to resonate with me, but one thing that did was the idea of having a demonstrable, unique link to something you care about. So, generalizing that, why not consider links to pieces of text? Why not links to the unique hash codes for literally anything?

BabelNFT was born a week later!

 
 

With a poetic similarity to Borges’ Library of Babel, the NFT we made allows any possible value up to 256 bits long, the length of token IDs for NFTs. There are approximately 10^77 possible tokens, and while most are utterly meaningless, there exists a token for nearly every person, place, idea, creation, in every language. There is no associated metadata, nor any image, video, or literally anything else. There is only the token ID, and you assign whatever meaning to it you like.

The simplest way to interpret the token ID is as text: so we decode the raw bits into text by the same means your emails are decoded into human readable format. This allows anything that can be written in an email to have a corresponding token, so Hello!, 🔥, and 어이 each have their own token. However, this has the length constraint of 32 bytes. For anything longer, you have to reduce the size, but SHA-256 hashing is perfect for it.

 
 

What I like about the idea is it makes it more clear (for me, at least), where the value resides in NFTs. First you have to buy into the idea of a specific universally visible ‘registry’ of links, generally without anything more to ownership than that. If/once you do, then you can decide the value of different links however you want. I enjoy the idea of owning links to some words and phrases I care about, so I minted a bunch of stuff like ?, Rick and Morty, and Milky Way. A friend of mine minted John 3:16 and Kanye West, and another minted Love. The cool thing is you get to decide.

Check out the project, and mint your own (or just read more about what it is) at LibraryBabel.xyz.

Previous
Previous

Tepe project: asynchronous trustless data sharing

Next
Next

A sprint through some machine learning paradigms in TensorFlow, in the form of colab notebooks | Nov 2021