Portability Architecture TF

From Social Web XG Wiki
Revision as of 11:03, 16 August 2009 by Mcarvalh (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Interested in Participating

Please add your name to the list if you're interested in participating

Topics to Discuss

Task Force Charter

Quoting from the SWXG Charter:

We will not create new technology. However, we will report on the landscapes of existing technologies and examine their compatibility and inter-relationships. We will also determine how they fulfill core goals of the W3C such as accessibility and internationalization, as well as make recommendations for future standardization work.

Quoting the description of the Technical Report from the Main Page:

A description of active and past technical work being used by industry that will seek to explain how the W3C and community at large may leverage the best of breed and highlight possible weaknesses.

These are highlighted here as they potentially conflict with some of the ideas proposed below for discussion.

Mission

The goal for this task is to interconnect the different information and notification systems related to social heartbeat and social graph for some groups of people. The long term goal is to have the open formats and protocols that allow to have different hubs and silos for this kind of information related to social graph, presence, microblogging, etc with secure and privacy enabled in a very usable way.

SocialHub.001.jpg

Whilst this is a worthy task I (Peter Ferne) think it goes some way beyond the remit of the SWXG as stated in the charter. This is not of course to say that members of the SWXG should not be working on such a task, only that it is perhaps properly regarded as separate from the work of the group. Or perhaps something similar could be included by way of a vision of future interoperability.

As I (Peter Ferne) understand it our goal in the Distributed Architecture and Data Portability Task Force (which needs a snappier name IMHO) should be to discover, understand and document how people are seeking to address these issues. We need to catalogue existing formats, protocols and what not, analyse how and where there are overlaps or gaps, and propose the work required (within the W3C) to make the vision of open exchange of social data and open interoperability of social network services a reality.

Distributed Architecture

Areas of interest:

  • Expanding the OpenMicroBlogging work into something that works for more than just microblogging, providing cross-site friendship links / subscriptions.
  • XMPP and Google Wave
    • Look at Google Wave's use of XMPP (Wave Spec) for social graph syncing.
  • Atom and Atom Publishing Protocol
    • Have a REST interface to publish and consume notifications and different formats of information.
  • Widgets
    • Look carefully at OpenSocial, see how it interops with existing FOAF and PortableContacts descriptions, and W3C Widgets, internationalization, accessibility
  • Portability and Distributed social graph storage
  • The web as a framework: "code visits data versus data visits code"


Portability Roadmap

  • Identification of information and protocols and services for interchanging social information and social graph.
    • Formats
    • Protocols
  • Define an open format (Foaf based) and protocols to access it in several ways.
    • Open Data Format for social network interconnection
    • Protocols to access this Open Data Format
    • Best Practices of usage.
  • Define a distributed architecture for distribution, federation and aggregation for social information based on the previous defined format.
    • Distributed architecture for distribution, federation and aggregation for social information based on Open Social Information
  • Have some reference implementations for the architecture.

Deliverables

  • Actual state of the art in formats and protocols related to social graph and social information.
  • An updated vCard RDF spec
  • Having FOAF work with Portable Social Contacts
  • Having FOAF work with vCard
  • OWL/RDFS/SKOS mapping inbetween specs
    • Maybe build upon SISM (Social Identity Schema Mapping) - SKOS-based mapping schema for interoperability between social vocabs
  • Protocols and architecture for open interchange of social information ( based on Distributed Hash Tables (DHT) and map-reduce ? ).
  • Distributed aggregation, filtering and searching of semantic information coming from microblogging.