Re: draft proposal for SWUI workshop at ISWC

Lloyd and monica,
I am belatedly getting to look more closely at the proposal. Initial
thoughts, with possibly more tomorrow. Thanks for driving this! Duane.

1. Do you have a format in mind at this early stage? One of the things that
was considered important by reviewers in last year's submission was the
inclusion of the interactive discussion element. Is it worth elaborating on
how the workshop might go about its aims, at least in a general way?

2. I'm not sure I'm clear on the phrase "We aim to map the "problem space"
that binds these issues together." This is probably related to my first
point.

3. There is an additional workshop reference you may want to cite:

User Aspects of the Semantic Web. Held in conjunction with the 2nd Annual
European Semantic Web Conference, Heraklion, Crete. May 29, 2005. Online:
http://kmi.open.ac.uk/events/usersweb/.

I would also have included the one I did last year at UPA for the usability
community, but I'm behind in posting the notes that came out of it... Sigh.

4. In advance of the workshop last year, we developed a general
categorization framework for what we were seeing in papers and prior
workshops. This might help provide some starting points for this year, as
well:

- General issues that impacts user acceptance of semantic behaviors - user
perceptions and social/cultural issues that affect user experience in
semantically-enabled applications (such as trust, collaboration,
transparency and authorization of agent action), requirements for future
applications, and/or considerations that need to be made when thinking about
user needs and design guidelines.
- Ontology models/frameworks - approaches for creating, managing, or
presenting ontologies, controlled vocabularies, and "folksonomies." This may
include models illustrating how concepts/contexts could be represented, the
implications of defining "meaning webs" and how they can be maintained so as
to remain relevant, and how they may be derived in an automated or
semi-automated way ("concept extraction") to support semantic-enablement of
static content repositories.
- Query or parsing models - how queries are constructed or user language is
interpreted and how it can/should be processed by the computer. The
relationship between unstructured free text search and search/navigation of
controlled vocabularies, as well as . The focus is on mapping between the
end user's perception of language and the end goal - information or task
completion - and the role that context can play to assist the user.
- Interaction models - specific methods by which users can interact with a
site or application in a novel way, or in a way that exploits the unique
benefits of the semantic web. The models aim to describe approaches to
interaction, rather than specific interaction implementations themselves.
The interaction might be small scale (e.g. particular screen messaging,
search support, handling drop-down lists or buttons, or presenting links for
navigation), or large (e.g. cascading menuing/window arrangement, complete
environments, device-specific segmentation/display).
- Applications - sites or applications that is present a complete user
experience which is leveraged by the use of semantic information and
technologies. Where possible, these will be briefly illustrated or
demonstrated.
- Interaction components - particular components or sets of components
supporting user interaction or experience within a site (rather than an
entirely semantic application), for example a visualization widget/applet, a
control for filtering selections from within trees or navigation, selection
elements that can be integrated with PDA or mobile phone applications, or
particular view formats that could be incorporated with other components in
an application.
- Enabling technologies - details of a technology, application or design
approach that provides infrastructure upon which semantic applications can
be created.
- Design/evaluation processes - processes to gather user requirements,
encourage user participation in design, evaluate usability with actual
users, and/or report on outcomes of user involvement. In particular these
need to highlight methods that respond to the particular needs of analysis
and design for semantic environments: high user context, dynamic data
integration and presentation, the use of ontologies and inferencing, etc.

Received on Tuesday, 4 April 2006 08:58:41 UTC