Closed adamkrellenstein closed 5 months ago
Wanted to suggest that the other day.. I tried it back in March, it's easy to use. It's not replicated, or at least it wasn't back in March, so if data persistence is a requirement maybe alternatives should be considered as well.
Yeah, what do you do if all of the nodes hosting a certain file (or a part of a file) go down??
I guess we'd have to require every Counterparty node to host every file...
Right now, it's obvious that there IPFS is not enough mature to be used in production. But it seems to me very promising, and I think it's worth to test it combined with OP_RETURN as an alternative way to encode transactions data.
@ouziel, see https://github.com/namesystem/blockstore/issues/80, blockstore has OP_RETURN too.
"Right now, there’s an implementation on the Namecoin blockchain, but the protocol could just as easily be ported to sidechains, a logic-heavy blockchain like Ethereum, or protocols like Counterparty that build embedded chains of consensus on top of the Bitcoin blockchain (let’s call them topchains)." - http://blog.onename.com/blockstore-bitcoin/
@rippler, IPFS seems to me more powerful and flexible than blockstore and more adapted to our needs (for instance IPFS don't need an external blockchain to exist, is an immutable decentralized DHT).
@ouziel-slama I like IPFS, I set it up it as soon as it was announced, but it's going to be a few quarters (or more) before it gets persistence and replicas. They're planning to have their own coin for it (perhaps it'd be convenient if it existed on CP). Edit: note on persistence and anonimity: https://github.com/ipfs/ipfs/issues/50#issuecomment-90629889
IPFS may be supported as part of blockstack.
https://forum.blockstack.org/t/general-architecture-of-blockstack-initial-version/54
We need something with guaranteed availability, ofc. :/
Use for storing larger data payloads?
Primary concerns: