Flow's approach to oracles and various off chain data

Hi - i’m interested to know how Flow handles off-chain data such as that used in NBA Top Shot - how have you approached the data on-boarding and verification?

There are plenty of examples of off-chain data / on-chain state including DAOs linked to LLCs and Chainlink’s entire proposition, but is there something unique about Flow’s approach?

Thanks

Welcome to the Flow community @mrbulb! Thanks for posting. :slight_smile:

NBA top shot only stores text-based metadata on chain. All the images and videos are in a regular database.

We currently have conversations going about on chain metadata, but those are still in progress: IMPROVEMENT: NFT Metadata · Issue #9 · onflow/flow-nft · GitHub

Of course, options like IPFS or the others you mention are also possible, but I don’t think a project on Flow is using them yet, but we’re definitely moving in that direction!