Beyond Caif - Cow / Escape from Client Server

From IIW
Jump to: navigation, search

Session topic: The amazing hyper connected API + XDI based world / Beyond calf-cow escape from client server

Convener: Judi Clark, Doc Searls

Notes-taker(s): Mark Sabadello

The goals of the session were to:

  • Imagine the VRM-enabled network we want to build and its properties
  • Think "outside the cow", i.e. build a system that does not rely on a central authority

Ideas raised during the session:

  • Today's online world is a set of domains, and the web is just one app on the Internet, which was originally meant for endpoint-to-endpoint communication
  • On the web, there are always dependencies which we cannot escape
  • But the offline world is different, e.g. you get into your car and you can do with it what you want
  • The web allows us to publish, but not to selectively share and protect personal data
  • The web does not provide good mechanisms for referencing, sharing and linking the data we are dealing with
  • Must get the right data to the right people to empower us and even save lives
  • To achieve this, stable identifiers are needed (DNS and IP addresses are not stable)
  • XRI could provide such identifiers, and XDI could provide the technology for referencing and linking data
  • Content-centric identifiers such as magnet: URIs also exist


Data is like seashells.. I have the largest collection of seashells, I keep them on the beeches around the world!


To get rid of cow/calf model, we need different architectural models, e.g. Federations, Personal Data Stores, FreedomBoxes, etc. instead of centralized silos.


We need to be more inclusive of things other than hosts and other than individuals.

For example, include groups, organizations and the Internet of things in our vision.

Also abstract concepts, not just concrete things from the real world.

Need to be able to discover and describe all those entities.


Necessary technologies that were mentioned during the session:

XDI, RDF, Semantic Web, Persistent Identifiers, Torrents, Global Registry, Distributed Hash Tables, Distributed Edit Histories


Other important ingredients for the ecosystem:

Contracts, Agreements, Disclosure, Informed Consent, Provenance, Infrastructure