I am really far to be an expert in blockchain I rather know a bit about cryptography and user interface, but at the same time, it is possible to me to express my needs regarding the NFTs feature and the possibility to store information onto the blockchain which would ideally be not the hash of a file but the file, as text such as ASCII/base64 html file format existing for png/jpg/mp3/pdf/...
I stated a few weeks ago that, to be crystal clear with the people in Web 3.0 / Metaverse domains we should ensoul principles to contain and reduce loss of power (see, incorrectness of platform I on't mention) and that the remedy I was thought by my mind, was to make authors being public and human, code being open-source, application client-side without analytics services, cryptographic keys mostly stored offline or separated from centralized structure and NFTs being 1:1 on-chain, as the hash would just leave you a kind of dirty sock of your file you bought on XYZW... but we except our file to last forever, and for everyone.
Look, without getting over-complex, there exists archive.org, a very nice website for storing file for the public to exercise their free will to borrow them it could be nice to be able to borrow NFTs in kind of temporary lend/lock contract, very fair.
Also, in archive.org, you can search for files, and this is absolutely essential to have too on vsystems cloud feature since for an app to be client-side and reliable alone, one should not need to construct a centralized proxy/server for querying the DB. Thus, what I remember having had used PouchDB which is the web bro of CouchDB, is that it stores file in binaries even base64 ones (20% size reduce) and is faster to query by their deterministic key/value from the metadata (Therefore we avoid duplicate binaries file) which comes as a bundle of text in JSON with files being attached to as their can be a versioning too which can't be deleted but the metadata changed without versioning and it could be structured with tags, title, description, ... THE SAME archive.org do,THE SANER it can be.
Regarding storing binaries, and attaching them to json metadata in CouchDB and managing versioning, it works naturally as such, yet I am aware we use IndexedDB in Vsystems, but my interest is that they have PouchDB which can replicate parts of CouchDB and work fine in a web browser...
Yeah, in other words I mean, we could query metadata for free from the API, and if folks want query binaries they build their custom API... Regarding binaries, as I was limited by 73Kb on Hive/Blurt/Steem per post, I always managed to run with it without issues of size.
Hello,
I am really far to be an expert in blockchain I rather know a bit about cryptography and user interface, but at the same time, it is possible to me to express my needs regarding the NFTs feature and the possibility to store information onto the blockchain which would ideally be not the hash of a file but the file, as text such as ASCII/base64 html file format existing for png/jpg/mp3/pdf/...
I stated a few weeks ago that, to be crystal clear with the people in Web 3.0 / Metaverse domains we should ensoul principles to contain and reduce loss of power (see, incorrectness of platform I on't mention) and that the remedy I was thought by my mind, was to make authors being public and human, code being open-source, application client-side without analytics services, cryptographic keys mostly stored offline or separated from centralized structure and NFTs being 1:1 on-chain, as the hash would just leave you a kind of dirty sock of your file you bought on XYZW... but we except our file to last forever, and for everyone.
Look, without getting over-complex, there exists archive.org, a very nice website for storing file for the public to exercise their free will to borrow them it could be nice to be able to borrow NFTs in kind of temporary lend/lock contract, very fair.
Also, in archive.org, you can search for files, and this is absolutely essential to have too on vsystems cloud feature since for an app to be client-side and reliable alone, one should not need to construct a centralized proxy/server for querying the DB. Thus, what I remember having had used PouchDB which is the web bro of CouchDB, is that it stores file in binaries even base64 ones (20% size reduce) and is faster to query by their deterministic key/value from the metadata (Therefore we avoid duplicate binaries file) which comes as a bundle of text in JSON with files being attached to as their can be a versioning too which can't be deleted but the metadata changed without versioning and it could be structured with tags, title, description, ... THE SAME archive.org do,THE SANER it can be.
Regarding storing binaries, and attaching them to json metadata in CouchDB and managing versioning, it works naturally as such, yet I am aware we use IndexedDB in Vsystems, but my interest is that they have PouchDB which can replicate parts of CouchDB and work fine in a web browser...
Yeah, in other words I mean, we could query metadata for free from the API, and if folks want query binaries they build their custom API... Regarding binaries, as I was limited by 73Kb on Hive/Blurt/Steem per post, I always managed to run with it without issues of size.
See? Those images are only 10.3Kb and 12.0 Kb.
EDIT: Useful links: