Multi-View Desk

The Design Challenges : tldr

Design challenges for working in XR with text includes use of space, legibility and physical affordances. The suggestion made here is therefore to provide a view of text in space for a Map type view with a plain background and to use the physical desk in XR, particularly for a ‘swipe’ gesture for the user to pivot their view into very different environments.

The Introduction

For a new user, the initial view when putting on a headset might be something like a rectangle overlaid the physical table with a traditional, maybe even 1970s computer monitor, which when turned on fades or ‘explodes’ into a large view and the desk wraps around. The intended effect would be for the new user to see the clarity and minimalism of the environment and then see it grow beyond what can be done in traditional, physical space. Just a thought.

The View

Having gone through the basic exercise of looking into how text should appear in XR and concluded that even though we have talked about knowledge ‘objects’ in an abstract sense, it takes a lot of space to have some kind of shapes in space, even lozenges or rectangles, so I have decided that I think it would be worthwhile to have at least one no-background view for a Map type view.

The basic view will be knowledge in space, minimalist. But how to accommodate other?

This led me to the notion of using space below eye/reading level for tools and info, a desk:

The Desk

The suggestion is giving the user a larger virtual desk as an information and interaction space (since it has a desk surface overlaid in VR) below reading eye-level, where the user sits inside what is shown as a circular desktop which they can virtually rotate to go through different information and controls.

The parts of the table which are physical might be shown as totally opaque whereas the continuing area around is semi transparent or a different color, for example, to make the user more aware of their physical environment.

This is inspired by a poster I saw on the London Underground (see below).

Default View. I can imagine the default view being something like Author’s Map view (in terms of initial color sets to experiment with), within arm’s reach. Controls should also be on the table for selection and manipulation of the objects, similar to ‘Show’ and ‘Layout’ in the Map etc. which I have already extensively experimented with and which can be a source for inspiration and contrast. A straight port of Map from Author to XR is not what I am suggesting, but it exists and works so we can interact and learn.

The ‘Rotate the Entire Desk & View’ Gesture

Swipe Control. To rotate the desk and view, you would put your hand down on the (physical with virtual overlay) table, palm sideways, and swipe right or left to rotate the entire workspace around you (see below).

This was inspired by the Softspace fist gesture which allows the user to move the entire knowledge set.

‘Alignments’: This way of working could then provide access to workspaces (let’s call them ‘alignments’ since that refers to perspectives and the moving shape maybe?) for a:

  • Map view, (at least one),
  • Writing view (vertical),
  • Timeline view,
  • Geographic view,
  • References (with citation maps of course), and
  • Active reading spaces where it becomes natural to move highlights, comments, notes and extracts into what is being worked on, as a paste into manuscript or left in space by the manuscript.
  • Doing an annotated bibliography could even merge two work spaces (reading and writing) perhaps, to optimize that workflow.

Note that when working in one such view it is not constrained to its slice of the pie so to speak, it can take over the entire view, the rotation pretty much acts like a toggle.

I can imagine interesting controls for changing the background view, maybe based on/locked on each of such workspaces, so that for example a rotated view to a relatively traditional writing view has a background 3D or image, whereas the Map does not. Plenty space for invention and innovation to discuss.

The Inspiration for Circle Desk

Sorry about the reflection, but here is the poster which inspired the circular desk:

The Swipe Gesture

Note that gesture clash is a familiar issue so the user might have to put their hand onto a specific spot for this to work reliably and not happen by accident.

The Gen-AI Visualizations

ChatGPT/DALL-E helped me with this visualization, which does not have the desk fully surround the user, which makes sense in a physical environment but in a virtual environment could happen. I asked for a woman user, since Emily modeled the interaction (above), but I think it’s nice that it chose a red-head for this, since I can easily imagine Dene in this environment.

Please note that in this visualization it’s still all about screens, that is the Gen-AI limitation on imagination. We should maybe have one such ‘alignment’ but others should be with a wide open space and more.

At least this visualization shows how one view is not only one slice, doing the action of swiping can (does not have to for all cases, depending on what is useful) change the entire view.

Further attempts, which do not show a fully 3D view but at least might give a notion of different kinds of views the user can gesture to get:

And in ‘nature’:

And on the moon. Despite the analog equipment, I think it provides a good perspective on many things in the circle and 3D content all around, should she wish. A swipe gesture and she could be in nature, another one working focused on a single screen writing in the traditional way and yet another in a massive knowledge map environment.

And when the desk and everything fades away and we work in a different view, where the user can still summon the visual overlay of the desk again (maybe through hand or arm menu?), to swipe gesture to another view:

Further views from 3D models

This model suggests a soft shape which the user can gesture around:

Maybe a semi-circle like this:

Design

To alleviate the issue of there being what looks like a table but there is no physical table underneath/in reality, maybe take a section of an octagonal table like this:

And make only the center part solid:

So then maybe even make the sides semi-transparent to indicate it’s not physically real but the user can do the swipe gesture and each section represents a view which will rotate inn:

1 comment

Leave a comment

Your email address will not be published. Required fields are marked *