2006-10-01: Design of dataset wiki pages

From Datafedwiki

<Back to Task List | Create New Task | Edit with Form
Title: Design of dataset wiki pages
Date: October 1, 2007
For User: DataFed
Status: InProgress
About Project or Tool: {{{about}}} (Add New project or tool)
Objective: [[:{{{Objective}}}]] (Add New Objective)

---

<Back to All Tasks

A long-term goal of DataFed was to allow the collection of various data types beyond the structured data accesible through standards-based data access. These odd data, related to a given dataset, may include bibliographic references, pictures, standard webpages.

Also, there is a need to collect and display application and usage pattern of a given dataset within DataFed and outside. User-provided lists of bugs, wishes, and discussion items (blogs) are also desireable.

A good model for the linking pattern is Google Scholar. It maintains bibliographic references (forward links), citations (backward links) and other useful attributes. For each bib entry it links to citations, related articles and general web search. It also exports content in multiple bib formats.

Candidate applications for mashing:

• Mediawiki - homepage of each dataset and it also houses the feedback and discussion
• DataFed - provides static dataset registration as well as usuage statistics
• Google Scholar - scientific references to this dataset
• Google - Web pages and images
• Blogs, del.icio.us, flickr,... - harvest these to retrieve relevant data.

Approach to mashing:

Each dataset page may contain different views of the dataset-relevant info.

Whenever possible the mashing needs to be through dynamic queries which are maintained automatically. However, it is also clear that for some of the application linking needs to incorporate human filters to eliminate spurious finds.