The Forum’s document repository is a database of our publications with access to the full-text digital document.
The repository is the result of collaboration between the Global Forum’s Secretariat and FAO’s Library and Documentation Systems Division (WAICENT), who kindly provided us with a customized version of their Electronic Information Management System (EIMS). It is database driven and it interfaces with the web through XML.

A brief technical overview 

The EIMS provides the necessary workflow and the metadata element set needed to describe the digital documents. It also implements the Forum’s taxonomy and the AGROVOC Thesaurus used to describe the content of the publications.

The new repository supports XML exports compliant to both EIMS (Dublin Core) and AGRIS AP formats. Thanks to the AGRIS AP export, GFAR is now an AGRIS Center (for more information, see the AGRIS Network). 

RSS feeds have also been created to make GFAR publications available to other sites (see below).

The main functionalities of the system are:

• Documents can be published and made visible on the web world wide
• Documents are managed in the system following information management standards (i.e. AGRIS AP metadata model) and terminology standards (i.e. AGROVOC)
• Access to the document is preserved by producing the minimal metadata record
• Electronic documents are saved and preserved in a file server
• The search can be customized according to different needs

The documentation is organized according to the GFAR Taxonomy, which has also been mapped to AGROVOC.

The Forum also plans to prepare a glossary to define the main ARD keywords. Special queries will be formulated to provide access to documents from other electronic databases. 

Feeds from the repository: info for webmasters

Pre-defined feeds
Some pre-defined feeds of specific document sets are available here as cached RSS feeds (updated every 3 hours) for fast harvesting.

Dynamic feeds
Please use this feature with caution. Harvesting dynamic feeds at short intervals could overload our server which would force us to withdraw this service.
The base URL is http://egfar.org/egfar/inc/docRepRSS (this loads a default feed of the latest 50 records - feeds containing a larger number of records should be called only at long intervals, as they require time and high CPU usage).

A good way to filter the feed is to use the following parameters:
- QueryString (searches the title, abstract and author fields: for the moment, only a one-word search is accepted)
- Title (for the moment, only a one-word search is accepted)
- Pub_year (publication year)
- recordsperpage (the number of records you want to retrieve: 50 maximum)
- Pub_id (only if you know the ID in the repository: it is useful if you want to retrieve a single record)

The sort order is by publishing year/submission date, descending.

The RSS format used is RSS 2.0 + Dublin Core namespace (particularly, dc:creator and dc:subject).

A good example of a useful feed for "recent publications from GFAR" would be:
http://egfar.org/egfar/inc/docRepRSS?recordsperpage=20

_____________________________________
Related documents:

The GFAR Document Repository

Selection Guidelines for Submission of Documents in GFAR Bibliographic database

GFAR Bibliographic database submission guidelines