From my brief time trying to understand the various tools and tech available for NFT's, I stumbled on the area of lazy minting; but with the seemingly few sources of information on the topic (the main source I could find is this video by Open Zeppelin which I highly recommend) I wasn't able to fully clarify my understanding about a few things.
In particular, for "on demand" lazy minting (e.g. ECDSA where the NFT data is encoded in the blockchain on demand, not ahead of time like storing root node of a merkle tree) in the context of original art/data source NFT's, I wonder if this removes the benefits of traditional NFT minting in regard to proof of origin?
Here's an example to illustrate my thoughts:
The process of "regular" minting associates the particular data with the signers keys, and a timestamp on chain. This enables the artist to prove he was the first to put it on chain, and if the artwork is original (i.e. no obvious alternate sources found elsewhere, copyright, etc..) then in theory it should act as a pretty good claim on the original source (which should get stronger as more people do it / the more standardized it is to do this).
But with dynamic lazy minting, since nothing's on chain yet someone could take advantage of the system by creating his own signed copy of any NFT that's listed off chain like this and saying that they are the original creator. There could be mitigation measures like listing timestamps and authenticity check periods on the centralized marketplace platforms, or just the reputational risk involved - but AFAICT there's no truly robust solution. Like in the aforementioned case there's nothing stopping someone from selling/listing copied work in a place without these measures. Okay sure, maybe people would only trade on "trusted" platforms that implement those measures, but then it's facing the same old centralized power problem. Without being on chain, it's not trustless. Especially for small/unknown artists trying to get started you can probably see how this would be a real problem.
So if this all makes sense, then I believe the best compromise to achieving on demand lazy minting would be to batch encode the data where multiple NFT's get encoded on the chain at once in the same transaction (not minted yet, just encoded). I assume this would be relatively cheap on gas and whatnot, and could be accomplished by posting another merkle root for each batch for example.
I'm very new to this topic (Eth, and NFT's) and just have dug in over the last day or so. Part of why I'm questioning my understanding is that apparently some NFT marketplaces (like mintable allow users to mint for completely free. So how does this work? I've tried finding out but search results are flooded with non-technical articles and such. The best explanation I could find so far was the topic of lazy minting (this is how I stumbled on it) and in particular that Open Zeppelin video. But as detailed above, AFAICT lazy minting alone doesn't fully cover this feeless process. I'm hoping someone could explain how these marketplaces work with how the feeless process is enabled -- is it more complicated or are they just not that concerned with this "proof of origin problem"?. Or otherwise please clarify/correct my thoughts and point out what I'm missing. I intend on tinkering around and just want to make sure I understand how this all pieces together before I get too far in.
