Visualisation Use Cases

RDF data is often presented visually. For simple cases, node-and-arc 
diagrams (things connected to other things) are often used, with some 
automatic layout for managing the number of connections to any 
particular node. Understanding the information presented depends on 
several factors. For example foafnaut - http://foafnaut.org - presents 
a person with some data about them, a picture of them, and information 
about how many people they know or are known by, or for how many people 
there are photos of them together available.

For generic RDF data (i.e. where the information recorded isn't known 
in advance) it is more difficult to determine what connections are 
interesting. Being able to select among those available is valuable. 
Being able to merge or de-merge connections which are subtypes of each 
other is a useful strategy for dealing with large amounts of granular 
information. This is something that is often not represented explicitly 
in current systems, either visually or otherwise.

there is a SWAD-E workpackage on visualisation and accessibility - so 
far we have produced some preliminary ideas: 
http://www.w3.org/2001/sw/Europe/reports/w3c_note_sw_accessibility/ as 
well as the image annotation workshop which explored some ideas in this 
area: http://www.w3.org/2001/sw/Europe/reports/dev_workshop_report_1/

(sorry these are such rough thoughts. I thought it was better to get 
them out than polish them in private, since other people may know this 
already or be happy to find the pointers...)

cheers

Chaals

On Wednesday, Sep 17, 2003, at 08:12 Australia/Sydney, Jim Allan wrote:

> Visualization also allows the viewer to get a gestalt (or trends and
> relationships) of the data presented that a more granular (individual 
> data
> points) view does not provide.
>
--
Charles McCathieNevile                          Fundación Sidar
charles@sidar.org                                http://www.sidar.org

Received on Tuesday, 16 September 2003 20:29:16 UTC