Collaboration in Shared Space

The core idea of collaboration in shared space is simple and powerful: instead of each person looking at their own separate view of a design, several people occupy one shared model at the same time, each able to move around in it and see the others as presences within it. A shared spatial session synchronises the model and everyone's position across the network so that when you look toward where your colleague is standing, you see a figure there; when they point at a wall, you see them pointing at that wall; when they walk to the window, you watch them go. The design stops being a private picture on each person's screen and becomes a common place the group is in together.

This matters because so much of design collaboration is spatial and deictic - it runs on 'this', 'here', 'that one', 'over there', accompanied by pointing, gaze and gesture at shared things in a shared space. Around a physical model or on a real site this is effortless; over a flat screen share it collapses into 'no, the other window, the one on the left, no, your left', because there is no shared space to point into. A shared spatial session restores the thing that screen-sharing loses: a common three-dimensional space everyone inhabits, in which pointing, looking and moving mean the same thing to everyone and refer unambiguously to the same real place in the design. For a coordination review, where the whole task is people gathered around a model resolving spatial questions together, that restoration is genuinely valuable.

Participants can be co-located (in the same physical room, sharing the space both really and virtually), remote (each in a different place, meeting only in the virtual space), or a mix of the two. The remote and mixed cases are where shared space offers something no other tool does: people who are physically far apart standing together in the same unbuilt building. That is a real capability with no equivalent in a drawing, a render or a screen call - and it is the heart of why collaboration is one of spatial computing's more compelling uses. But it depends entirely on how well each person is represented to the others, which is the next question, and on everyone being genuinely 'in', which is the honest problem after that. Whatever the group agrees while standing in the model, the binding coordination and dimensions are still resolved against the verified BIM and drawings, not settled by where an avatar happened to point.

For a shared session to work, each person must be represented to the others, and the quality of that representation - how strongly you feel the others are really there with you - is called presence, or more precisely social or co-presence. At minimum, each participant is shown as an avatar: a figure, often just a head and hands or an upper body, positioned where they are in the shared space and moving as they move. Even a crude avatar carries a surprising amount of the information collaboration needs, because the two things that matter most for spatial discussion - where a person is looking and what they are pointing at - are exactly what an avatar's head direction and hand position convey. When you can see that your colleague is looking at the same beam you are talking about, the conversation flows the way it does around a real model.

Presence is built from these cues: gaze direction (where the head and, if tracked, the eyes are pointed), pointing and gesture (hand position and movement), position and movement through the space, voice coming from the right direction (spatial audio), and sometimes facial expression. The more of these are conveyed faithfully and without lag, the more strongly people feel genuinely together, and the more natural the collaboration. The less faithful they are - a frozen cartoon avatar, no gaze, laggy movement, voice with no direction - the more the sense of being together frays into the familiar feeling of a slightly broken video call, just in 3D. Good collaborative design invests in exactly these presence cues, because they are what turn a synchronised model into a shared meeting.

But presence has honest limits worth naming rather than glossing. Avatars are usually cartoonish or partial, which can feel odd, impersonal or even faintly unsettling; they rarely convey the full richness of facial expression and body language that real co-presence carries, so subtle social signals are lost. Eye contact, the felt sense of someone's mood, the small human cues that make a difficult conversation work - these are thin or absent. So while presence in a good shared session is genuinely better than a screen call for *spatial* discussion (pointing at the design), it is often worse than a screen call, and far worse than being in a room, for *social* and *emotional* communication (reading the people). A designer choosing shared space should do so for its spatial strength and stay aware of its social thinness, especially for delicate client conversations where reading the room matters as much as reading the model.

The uncomfortable truth about collaboration in shared space is that its magic assumes everyone is equally immersed, and they very often are not - producing an asymmetry that is one of the most socially awkward situations in all of spatial computing. The classic scene: one person wears the headset, fully inside the model, gesturing and exclaiming, while everyone else in the room stands around watching them wave their arms at empty air, seeing at best a flat, wobbling mirror of the immersed person's view on a screen. The immersed person is having a rich experience the others cannot share; the others are half-excluded spectators. This is not a rare edge case - with limited headsets, it is the *normal* situation, and it quietly undermines the collaborative promise.

Several kinds of asymmetry compound the problem. There is the headset-versus-screen split just described. There is the immersed-versus-real-world split: the person in the headset cannot see the real room or the real faces of the people beside them, so they may talk over others, miss reactions, or feel oddly isolated even while surrounded by colleagues. There is the co-located-versus-remote split: people together in a room share glances and side comments the remote avatar cannot, so the remote participant can end up peripheral. And there is plain technical asymmetry: different headsets, lag, someone whose avatar keeps freezing. Each of these chips at the feeling of a genuine shared meeting, and pretending they do not exist is exactly the hype this course warns against.

The honest response is to design around the asymmetry rather than deny it. Give the non-immersed people a good shared view, not a nauseating helmet-cam - a stable, well-framed screen of the scene, ideally with its own controllable camera. Where it matters, get everyone in (enough headsets for the whole group) or deliberately take turns so no one is a permanent spectator. Brief the immersed person to narrate and stay aware of the people they cannot see. Make sure the remote participant is well represented and actively brought into the conversation. And be willing to conclude, task by task, that a plain screen share where everyone is equal is simply the better collaborative tool for this particular meeting - because sometimes it is. Shared space earns its place when the spatial, in-the-model quality outweighs the awkwardness of getting everyone in; when it does not, the honest choice is the humbler tool, with the binding coordination resolved, as ever, against the verified model and drawings.

Set against that honesty, there is a use where collaboration in shared space delivers something genuinely hard to get any other way, and it is the one most relevant to Indian practice: bringing physically distant people into the same design at the same time. Modern design is distributed - the architect, the client, the structural engineer, the services consultant, the fabricator and the site can easily be in different cities or countries - and the classic ways of collaborating across that distance are thin: emailed drawings that everyone reads differently, screen shares that collapse into 'the other window', flat video calls that show faces but not the space. A shared spatial session lets those distant people stand together in the model and resolve spatial questions by pointing at the actual thing, which is a real gain in a distributed workflow.

For India specifically this fits the grain of how much practice already works. Design and construction teams are frequently spread across cities; a great deal of the country's building serves clients and a diaspora who are physically elsewhere - a family in the Gulf or North America building or renovating a home back in India, an NRI investor who will not see the site for months. For these, a remote shared walkthrough - the client standing in their future home from another continent, pointing at a room with the architect beside them in the model - is not a gimmick but a solution to a genuine distance problem, and it can be reached with affordable consumer headsets rather than exotic hardware. Even one-to-one, a remote client review in shared space can do what a screen call cannot: let the client *be in* the space rather than look at a picture of it, which is exactly the client-communication value this course keeps returning to, now extended across distance.

The mature position holds both truths at once. Collaboration in shared space is genuinely one of spatial computing's strongest and growing uses - restoring shared, pointable space to distributed teams, and extending the stand-in-your-home value to remote and diaspora clients - and it is honestly awkward, unevenly available, socially thinner than a room, and often beaten by a plain screen share when getting everyone equally 'in' is not worth it. The skill is the one the whole course teaches: judge, meeting by meeting, whether the in-the-model, point-at-the-design quality earns the friction of a shared session, lean in where it does - especially for remote and distributed collaboration - and reach for the humbler equal tool where it does not, always keeping the binding coordination and dimensions with the verified BIM, drawings, engineers and surveyors rather than with wherever an avatar last pointed.