Group Report: Applications

Introduction to the Group Work

• ways in which VR is particularly of benefit;
• does (or how does) the VR solution differ from a desktop interface?
• what hardware and software systems and tools are used and what are the limiting factors?
• how does this application impact on future education (e.g., use as an educational delivery vehicle, or use for educating those in specific disciplines)?
• (un)suitability of existing tools;
• can general-purpose VR systems be adapted for applications, or are more customised solutions necessary?
• problems of cost

We will be trying to amalgamate, structure, and perhaps prioritise, the issues brought to the workshop.

Possible recommendations include:

• what topics should these cover and what emphasis should they have?
• should there be moves at a national level to conclude site license deals for specific software systems?
• would it be valuable to conduct evaluations of VR hardware and software systems?
• might it help to have a national resource or centre for information/advice about VR (or does the Web make this idea obsolete)?

Roger Hubbold, University of Manchester

Group Discussions

The group discussed the issue: is VR really useful?

There is a problem with the current state of the art in that users really need to be programmers. As a result, the development times are lengthy and thus much more time is spent developing than evaluating (90/10 balance was suggested as being typical). There is a real need for "non expert" systems.

With the current state of VR technology the benefits are perceived rather than quantifiable. It was felt that waiting until the benefits could be quantified would be damaging to UK industry and HE. In order for UK industries to keep up with progress and maintain their competitive edge, it is vital that students are taught to integrate VR techniques with their other tools in computer graphics, visualization and multimedia.

The perceived benefits are seen as being:

• performance improvement - in design reviews; risk assessment (with EU regulations on health and safety this is a major benefit)
• Computer Supported Co-operative Working (CSCW) can offer savings in time
• access to expensive/inaccessible sources - example include virtual laboratories, virtual field trips (collect data in the field, measure at leisure), inside the patients head

End users are however sceptical at this stage and need to be shown that the technology can be effective and under what circumstances. Good case studies are needed. It is not clear that we can deliver at this stage.

There is a need to provide access to the technology:

• there is a need for low cost systems for wide use
• need access to some high end systems
• site licences for software might be appropriate, though at this stage there is probably no universal best buy
• there is a need for access to novel technology, for example 3D input

There is a need to provide access to the data:

• building virtual environments is expensive
• it would be useful to have shared models/data
• standards for data exchange (VRML) are likely to be important and other standards (PREMO) need to be considered, some standards may need to emerge to meet other needs
• the data required can be large
• there are privacy issues to consider (medical data) as well as other legal ownership issues

Software Issues:

• we need to address the differences between importing applications to VR versus importing VR into
• applications
• there is a need for application constraints and behaviour in virtual environments
• hence, we may need the integration of VR and the application
• portability is a major issue

The use of VR needs to be an integral part of the application if the technology is to succeed. Data needs to be altered and exported back within the system. Data integrity and version control are issues here.

Educating the Next Generation

• this is essential if the UK is to be competitive in this area
• there is evidence to suggest that such education will yield take-up in industry

Other issues include IPR.

There is also a concern that too many things are called "VR". We need to move to an application focus.

Recommendations

1. Conduct an evaluation of software

   There is a need for an evaluation of systems. This would provide a wealth of timely information for the community. There is unlikely to be a single solution to solve all problems but there may be some systems where we should be looking to obtain a CHEST agreement. Any evaluation needs to consider:

   • model building
   • run time
   • formats and standards, e.g. PREMO, VRML, STEP (insufficient is known here )
   • evaluation should look at strengths and weaknesses of packages
   • how the underlying data are handled
2. Conduct a series of application studies

   These should build on other work already underway or completed.

   • looking at the needs and benefits in different application areas and application-specific standards, such as STEP
   • look at the cultural issues involved and conflicts of interest
3. Involve UK HE in definition of standards e.g. PREMO, VRML2, ...
4. Develop a guide to good a practice including ethical issues
5. Recommend to TASC (JISC Technology Applications Committee which oversees JTAP) the funding of some VR application studies - we need to see how VR can be integrated into applications
6. VR Club (initially by EPSRC?) to organise workshops etc...
7. Collect exemplar applications and "object" libraries perhaps through CTI centres
8. Set up a VR training and education centre with short travelling courses - ideas based on successful JISC models for visualization through AGOCG and NTI/HPC training and education
9. Production of generic training materials, e.g. related CHEST packages
10. EPSRC programme to investigate integration of models and VR
11. Support for inter-disciplinary research with VR (computer science, cognitive, user disciplines..) within EPSRC and between Research Councils