Web2.0

Outcomes from the Web 2.0 Pow-wow 13th December 2007

The talk of the day mainly focused on Web 2 as a user-centric approach to creating web applications.

• The Richtags project have produced a wonderful interface to multiple EPrints repositories to facilitate social interaction.

• Connotea (Nature) already has an established EPrints user interface component

• ULCC has the SNEEP project and the Linnaean society repository. They have been developing comments and bookmarking facilities.

• The EDSPACE project is developing a user-centric approach to learning object repository.

• The FARAOES project are developing a repository user interface that follows the best-practice principles of Web 2.0 sites

How Web 2.0 can help with EPrints repositories

Web 2.0 Facilities such as commenting, bookmarking and tagging facilitate the generation of a large 'social' dataset, which can help researchers to connect with the kind of information that they seek much more efficiently. This is especially helpful in today's world, where the amount of information available to researchers is unprecedented.

Web 2.0 facilities could be

• Handled within the repository, so that (for example) items can be ranked by the number of comments, or their digg-style 'score'. The advantage of this is that the data is decentralised - control is not ceeded to some third party provider. The disadvantage is that the data is distributed between different repositories and it might be difficult to make sound inferences from the relatively small amount of data that resides on any one particular repository.

• Handled by external services like Digg, del.icio.us or Connotea (are already in some EPrints repositories) . The advantage of this approach is that there is little complexity in the repository itself, but equally, there is little benefit (value add) that the repository can offer the user.

Ways that Social Data Could be Managed Handled Within EPrints

The accumulated commenting and tagging data that such an approach gathers might be better managed in a separate user area - the user has their own area of the database where their comments an their tags are stored. There is already a user dataset in EPrints - perhaps its role needs to be enlarged.

Alternatively, these items can be managed as a separate table (by separate processes and user interfaces and services if necessary) and EPrints can just interpret the fields in the table as if they were readonly eprint metadata.

On the other other hand, EPrints can provide a category of metadata items that behave as normal (accessible through the standard API and user interface) but don't trigger changes in the metadata history or last-changed time. After all, an article doesn't change just because someone rates it.

Ways that Social Data Managed by External Services could be integrated into EPrints (??)

• See this Open Knowledge Foundation Blog Post for more information in repect of this.
• Google's OpenSocial
• Connotea
• Cite u Like

Efforts to Make Social Data Platform-Independent

APML (Attention Profiling Markup Language)

APML allows users to share their own personal Attention Profile in much the same way that OPML allows the exchange of reading lists between News Readers. The idea is to compress all forms of Attention Data into a portable file format containing a description of ranked user interests. - apml.org

So, APML acts as a track record of what you have been interested in socially across several sites during the last however many days. This attention profile could conceiveably be used as a way for helping social applications to fine-tune just what kind of relevant information they want to push at the profilee.

OpenSocial (An API from Google)

The web is more interesting when you can build apps that easily interact with your friends and colleagues. But with the trend towards more social applications also comes a growing list of site-specific APIs that developers must learn.

OpenSocial provides a common set of APIs for social applications across multiple websites. With standard JavaScript and HTML, developers can create apps that access a social network's friends and update feeds.

...

There are many websites implementing OpenSocial, including Engage.com, Friendster, hi5, Hyves, imeem, LinkedIn, MySpace, Ning, Oracle, orkut, Plaxo, Salesforce.com, Six Apart, Tianji, Viadeo, and XING. - Google's OpenSocial Page

---

Notes Transcribed from Whiteboard and Flipchart on the Day

These were taken from this Flickr collection.

From Notes 01:

What is Web 2.0?

• Changing workflow
• Voluntary vs. Involuntary information sharing
• User-Centric vs. Anonymous
• Building value from small collections => critical mass
• Can be difficult to sell to providers
• Serendipidy, Discovery - goes beyond search
• Open Architecture
  • Within Service
  • Between Services
• Difference between us identifying users and users identifying themselves

Concerns:

• Who owns tags etc.
• Barriers to deposit
  1. Solves my problem
  2. Becomes Social
  3. Emergent Information

Notes 02:

Free Access => Stability => Reliability

• Use information
• Authority about content
• Make people feel that they own their information
• Lower ingest(?) barrier
• Creative reuse
• Retention Policy
• Make the repository personal

Notes 03:

Sharing/Library of Configs

User-Centric Repository - Being done in:

• institutions
• by providers
• between repositories

Too Many APIs

How have we failed?