Difference between revisions of "Basic Concepts"

Revision as of 15:22, 11 October 2011

gCube once deployed and installed implements a service that is characterised by some foundational concepts, i.e.

Virtual Research Environments and Community Portals

Communities of scientists and researchers use the gCube middleware and Grid resources to set up Virtual Research Environments (VRE) for Content exchange and collaboration. The gCube empowered VRE provides a framework of applications, services and resources to support the underlying research processes. The purpose of a VRE is to help researchers in all disciplines by managing their increasingly complex range of tasks.

Portals provide the required tools for content search, browsing, annotation, metadata management and other value added capabilities. They usually consist of a collection of reusable Web modules, called Portlets.

Content, Information Objects, and Collections

The Content concept encompasses the data and information that the VRE handles and makes available to its users.It is an umbrella concept used to aggregate all forms of information objects that a VRE collects, manages, and delivers, including such objects as annotations, and metadata.

Content is composed of a set of information objects organized into collections.

The notion of Information Object represents the main entity populating the Content of a Virtual Research Environment. An Information Object can be a complex, multimedia and multi-type object with parts, such as a sound recording associated with a set of slides, a music score, political and economic data associated with interactive simulations, a Ph.D. thesis which includes a representation of a performance, or a simulation experiment and the experimental data set adopted, a data stream representing the pool of data continuously measured by a sensor.

Collections represent the “classic” mechanism to organize Information Objects and to provide focused views of the VRE Content. They are sets of Information Objects aggregated for some management purposes, e.g. behave as a single entity for access rights management. Another typical role assigned to collections is organizing the search space into restricted and/or topic-based sub-sets of it, thus allowing improving the search results.

Communities exploit any VRE collection through the features offered by the Web Portals.

Content Annotation

As a definition of annotation we could say that it is a written explanation or criticism or illustration that is added to an information object. Informally, annotations are comments or notes made by users in order to expose their ideas about a document. The annotations’ role in a digital library is very important since they promote collaboration between the users of the library. Through annotations, new ideas and concepts can be discussed, something which may help the understanding of an annotated object and the advance of knowledge about it.

In gCube a number of different types of annotations are provided. The first distinction has to do with the content of the annotation. So, there are text annotations and associations. By adding a text annotation a user may add a comment or a note about a gCube information object. On the other hand, associations are used in order to create conceptual links between two or more information objects.

Annotations can be applied to the whole document or just to a part of it. In the first occasion text annotations or associations are denoted as generic or non-anchored annotations and in the second as anchored ones.

Generic annotations should be used when the comment or the association is referred to the whole entity. General comments, notes and additional information about an information object belong to that category.

Anchored annotations introduce a more interesting concept of annotating an object. Now, the author is able to bring in information about a specific part/ fragment of an information object. Presumably, this extends the capabilities of users as they may indicate in a visual and descriptive way which part of the information object their comment refer to.

Metadata

The “classic” definition of metadata is “data about other data”.

Metadata are used for describing different aspects of the data they are associated with, such as the semantics, provenance, constraints, parameters, content, quality, condition, and any other characteristic that is considered important to record. These data can be used in different contexts and for a diversity of purposes; usually, they are associated with an Information Object as a means for facilitating the effective discovery, retrieval, use and management of the object.

The gCube system does not impose any constraint on the metadata it supports, the system is capable to handle any piece of text. Each metadata object contains data described according to a specified schema and/or ontology in the large (any conceptualisation that defines the terms about a domain). In fact, a metadata object usually manifest in a metadata record, i.e. a number of items of information where each item as well the values assigned to the item might come from one or more controlled vocabularies. An example of metadata schema is Dublin Core, i.e. a vocabulary of fifteen properties for use in resource description. An excerpt of a Dublin Core metadata record in XML is as follows

...
<dc:creator>Rose Bush</dc:creator>
<dc:title>A Guide to Growing Roses</dc:title>
<dc:description>Describes process for planting and nurturing different kinds of rose bushes.</dc:description> 
<dc:date>2001-01-20</dc:date>
...

However, multiple metadata schemas/ontologies exist and can be defined, each tailored or specifically designed to satisfy the needs of a well identified community.

Each Information Object might be equipped with an arbitrary number of metadata objects compliant with a specific schema.