[Cryptography] A naming and key distribution infrastructure for the Mesh

[Sid Spry]

On Wed, Sep 23, 2020, at 9:18 AM, Paul Wouters wrote:
> On Tue, 22 Sep 2020, Phillip Hallam-Baker wrote:
> 
> > So I don't want Alice's address to be alice at example.com. I want her to be @Alice.
> 
> We have millions of alice's who want to be @Alice. It can't work like
> that. Look at a 12 year old who wants to get a gmail address. The good
> namespace is already taken and they have to come up with weird stuff
> that none of their friends can remember anyway.
> 
> I agree that you don't want an email address as unique qualifier because
> people currently can't guarantee they can keep this identifier for life.
> 
> But you have a uniqueness and name mapping problem.
> 

Federated systems would solve this. It'd work kind of like email addresses,
but you need a more robust technology like XMPP. Users could be globally
addressable with their fully qualified federated name, e.g. alice at wonderland,
or their local name within the node.

What makes this better than email addresses is that the user would have a
closer relationship with their federated node.

There's still problems, but it seems like a good start. In practice fully P2P
systems don't seem to scale well.