status-im / specs

Specifications for Status clients.
https://specs.status.im/
MIT License
14 stars 14 forks source link

eip-1459 #79

Closed decanus closed 4 years ago

decanus commented 4 years ago

This PR contains everything related to EIP-1459 which will be used to bootstrap for waku.

oskarth commented 4 years ago

This PR should also update the current bootstrap section:

https://github.com/status-im/specs/blob/master/status-client-spec.md#bootstrapping

And possibly discovery section:

https://github.com/status-im/specs/blob/master/status-client-spec.md#discovery

This is where it'd be very useful to have a v1 tag, because then the spec diff is clear

decanus commented 4 years ago

@oskarth see changed to the client spec, short and sweet for now.

oskarth commented 4 years ago

Even if we happen to use the same implementation mechanism for bootstrap and general discovery, lets make sure we keep these sections separate, as they are separate components. Us using the same mechanism is a convenience thing.

decanus commented 4 years ago

So @fjl and I had an interesting discussion in Berlin today, about all the tooling they use for example to build the merkle tree etc. What is interesting is they crawl the DHT to find the nodes which they later add to the bootstrapping, decided to just note this here and make @fjl aware of the fact that we are using this issue.

oskarth commented 4 years ago

Ping @decanus @adambabik what's needed to get this merged?

decanus commented 4 years ago

@oskarth I don't believe its actually implemented anywhere. Once it is I am happy to merge.

adambabik commented 4 years ago

It's not implemented anywhere yet, for now I am trying to upgrade to geth 1.9.11 (https://github.com/status-im/status-go/pull/1899) as a prerequisite.

oskarth commented 4 years ago

@decanus IIRC consensus was to merge this as raw draft, could you adjust accordingly please?

decanus commented 4 years ago

@oskarth so the problem we still had which I remember is how to merge, because the EIP sped is marked as final, but now we are adding a draft thing to a final spec.

oskarth commented 4 years ago

The solution proposed IIRC was to add a column indicating "raw" state. This document is more of a compilation than a spec on its own.

decanus commented 4 years ago

ping @andremedeiros @yenda