Peercoin & keybase.io

Hey all,

Sorry for the long absence - between some craziness at work, then a long vacation, I’ve been away for a while.

I’m picking things back up where I left at on PeerApps, and going to be focused on QAing the Peercoin v0.5 release to ensure everything added for PeerApps is solid.

In the meantime, I could use your help.

Keybase.io is a natural complement to PeerApps. It allows a user to establish an identity associated with GPG keys and social accounts (if desired), and they’ve recently added support for Bitcoin. If we can get them to add support for Peercoin, I can use their identity system rather than rolling my own, which will be a huge boon to the project - not to mention the other numerous advantages it would bring.

If you have a github account, express your support for this feature request located here - let’s show them the Peercoin user-base is alive and worth building an integration for!

Thanks, friends.

I hope you don’t mind, but I made this post on Reddit to help drum up some excitement and get some more comments. You basically announced the Ebay-like project in your Github post, so I felt it was ok to post about it…

All good, thanks Sentinel!

[quote=“emeth, post:1, topic:3583”]Hey all,

Sorry for the long absence - between some craziness at work, then a long vacation, I’ve been away for a while.

I’m picking things back up where I left at on PeerApps, and going to be focused on QAing the Peercoin v0.5 release to ensure everything added for PeerApps is solid.

In the meantime, I could use your help.
[…][/quote]
Hey!
I think I speak for all here when I say: we are glad that you are back!
Next to @JordanLee’s Peershares your project is one of the most innvative attempts to make use of unique features of Peercoin.

This is not intended to disesteem projects like Peerbox, oh no! They are very important. But I consider them “Peercoin core” projects e.g. like Peerunity that intend to increase the utility and security of the “Peercoin protocol”. Without them Peercoin wouldn’t be what it is - a secure and sustainable coin!

PeerApps on the other hand is striving to make something based on that protocol, a “Peercoin application” - so “PeerApps” is wisely chosen :wink:

“Peercoin core” is the ground to build on and gladly it’s a solid one.
And “Peercoin applications” are built on that ground.

It looks like you have some support at https://github.com/keybase/keybase-issues/issues/1697 :slight_smile:

Welcome back!

Glad to see you again!

If Peerapps as you said on GitHub was to require a Keybase account, would that mean we depend on Keybase in the future? Is the central nature of Keybase not a problem?

Sounds useful though, and you probably know what you’re doing. I don’t yet understand these two projects well.

Great question!

Some good info here on keybase mirrors:

In the short term, keybase will indeed remain a single point of failure.

However, I intend on adding a concept of supernode to the PeerApps framework, where a user running both a full Peercoin node and a full PeerApps node can turn on PeerApps minting.

When they do so, they will become a processed data store for the last 14 days worth of data in the network (becoming a backup for the parasitic data stores) and for GPG public keys and their linked Peercoin addresses (becoming a backup for Keybase). In doing so, they will act as a gateway to the PeerApps network for light mobile clients, and collect a % fee from activity done by mobile users.

Regarding account creation, once supernodes get implemented I’ll add back in the ability to support accounts created straight on the blockchain in addition to KeyBase - since the supernodes will store the processed state of all published GPG keys on the Peercoin blockchain. The goal is to avoid requiring all users to process the entire blockchain, but to allow them to just process about 14 days worth and be up and running.

A final comment - I don’t think the threat is very severe on relying on a centralized service for identification, since it will be validated with signed messages. The only power Keyerror will have is to disable identities (by deleting them), which they have no incentive to do so - Facebook Connect doesn’t disable accounts because they used the Facebook Login feature on a sketchy site. If that becomes a problem though, we can fall back on requiring users to scan the entire blockchain for GPG keys until supernodes get implemented.

Any word from them?