AI Summary:
The meeting focuses on developing and refining a VR/XR demo presentation for upcoming demonstrations scheduled for Wednesday (Fabien) and the next day (Frode). The group is discussing how to structure the demo experience, what to include, and how to make it user-friendly for people who will be trying it.
Key points from the discussion:
1 They’re developing a sequential demo experience for presenting their XR work, with guidance from Mark Anderson.
2 They discuss the importance of clear onboarding and instructions for VR users, especially first-time users.
3 Features they’re working on include:
◦ A physical table indicator in VR
◦ Wrist controls that can be customized
◦ Document/PDF interaction capabilities
◦ A manuscript/cube system
◦ Movement of reference cards in 3D space
4 There are technical challenges they’re addressing, like ensuring the interface is intuitive and determining the best sequence for introducing features.
5 They acknowledge that first-time VR users experience significant context switching that makes remembering verbal instructions difficult.
6 The demo is meant to be archived as a “quarter one” snapshot of their efforts.
7 Later in the meeting, they discuss AI-generated illustrations, with Karl explaining LoRA (Low-Rank Adaptation) for training specialized AI models.
Frode Hegland, Fabien Bénétou, Peter Wasilko, Karl Arthur Smink, Tom Haymes, and Brandel Zachernuk.
Demo Structure and Refinement
• The team is creating a sequential demo experience where users progress through different stages
• They’re focusing on making the first experience approachable before introducing complexity
• Fabien has already built most of the technical infrastructure, and the meeting is about polishing presentation
Key Demo Components:
1 Welcome/Introduction: Instructions for basic interaction, showing users that they have a sphere on their left wrist they can touch to view commands
2 Physical Table in VR: A feature where users grab a yellow bar and move it to match their physical desk, providing spatial grounding
◦ The bar changes from yellow to gray when positioned correctly
◦ This helps users feel more comfortable and oriented in the VR space
3 Wrist Controls: Customizable controls on both wrists
◦ Left wrist shows available commands/code snippets
◦ There’s discussion about whether to teach customization early or later in the experience
4 Document Interaction: PDF/document interaction with highlighting capabilities
◦ Users can highlight and interact with text
◦ There’s a highlighter tool they’re refining
5 Manuscript/Cube System: A 3D object that can unfold
◦ The cube appears to be related to manuscript editing or viewing
6 Reference Cards: Small floating text cards representing documents or sources
◦ Users can move these around in 3D space
◦ There’s discussion about color-coding and improving legibility
User Experience Concerns:
• The team discusses the challenges of onboarding new VR users
• Karl emphasizes that “context-dependent memory” means instructions given before putting on the headset are often forgotten
• Peter notes issues with boundary warnings appearing when trying to interact with elements
• They discuss visual feedback for interactive elements (highlighting, color changes)
• They debate text length and readability in VR
Future Improvements:
• Better visual cues for interactive elements
• Improved text legibility
• Color-coding for reference materials
• More intuitive object movement
• Better hand tracking and interaction feedback
• Possible branching paths in future versions rather than just linear progression
AI and Visual References Discussion:
• Later in the meeting, they discuss using AI to generate illustration reference materials
• Karl explains LoRA (Low-Rank Adaptation) for training specialized AI image models
• They discuss posemy.art as a tool for creating 3D figure poses
• There’s interest in developing consistent visual styles for illustrations
The meeting appears to be part of a larger project related to authorship and knowledge work in XR/VR environments, with specific focus on text and document interaction in 3D space. The group seems to be creating tools for scholars or writers to work with documents, references, and manuscripts in virtual reality.