[drupal-devel] Google Summer of Code: Social context of content

[Kieran Lal]

Hello, please review this proposal and provide feedback for  
consideration as a Google summer of code project for Drupal.

The social context of content is important in evaluating content.

User questions this is trying to answer:
Who is the person everybody reads?
Who was there when the subject of the content happened?
What are your friends reading?
What is relevant for your next social meeting?
What it is the socially relevancy of media such as pictures or video?

The goal of this project is to allow for the descriptions of content  
in a social context and to automatically discover the social context  
of content.  The first phase will be to create a standard set of  
social tags for describing individuals, their relationships, and  
their activities.  These descriptions will be done using the  
taxonomic tagging module in Drupal and it's forthcoming customer  
relationship management module.   The automated discovery of the  
social context of content will be the second phase.   This will allow  
social context to be derived from information that is available but  
not coordinated.  This will include user page views, authorship,  
number of reads, response rate weighting, interaction of buddy list  
contacts, social weight of readers, responders social weight ,  
events, locative information, volunteer, and invitation information  
which are all currently available in Drupal modules.

Why this is innovative?  Content management has currently focused on  
supporting varieties of content, transcoding content, and integration  
of content with other systems but has done little to state why it is  
socially relevant to people.  This project aims to enable producers,  
consumers, and managers of content to describe content in a socially  
relevant manner.  It also aims to discover how users want to describe  
the social context of content and how they prefer to consume that  
information.  Drupal's well defined schema allows for discovery of  
social context in structured data that is presented to users in an  
unstructured way.

Deliverables
Part I: Social context tagging
An open standard for referencing individuals, their relationships,  
and their activities.   This will be an open standard for Community  
Relationship Management based on the CiviCRM open standard, SugarCRM,  
foaf, xfn, rojo (proprietary), aura (codecon 2005), and SalesForce  
for non-profits.
A database file with this standard taxonomy, and possible folksonomy  
extensions.
An interface for the module which allows social context tags to be  
effectively displayed with content.
An interface for allowing users to add social context tags or extend  
their own tags.
Part II: Automatic discovery of social context.
A module that drills across content management tables for social  
context information and creates a table with the total social context  
of content nodes.  This will be done in off-line batch modes as a  
series of data warehousing style queries.  This will be architected  
to include queries to other data marts.

Interested Googlers who would benefit from this open source project
Danah Boyd from Blogger and Ellen Spertus from Orkut would be able to  
benefit from research into social context prototypes.

Academic references
This project was inspired at the Computer Human Interaction  
conference 2005.
One paper from Microsoft described the need for social context of  
content in Microsoft Outlook rather than meta-data such as file size,  
date, and urgency.
One paper in the short papers section described how a photo of young  
people in a car on a trip could be socially relevant because it could  
have been the author's nephew who died on a similar car trip.  This  
illustrates the power of social context of content.
A paper from Microsoft analyzing Usenet post meta-data showed that  
power users were most interested in what the most respected people in  
a community were saying.
References to papers and academic relevancy will be added to this  
proposal.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://drupal3.drupal.org/pipermail/development/attachments/20050610/85d3d5e0/attachment.htm