How Much Memory Does a Temporary Team Need to Become a Community?

AnAn emergency response team assembled from strangers at the scene of a disaster has no shared history, no time to build one, and often no idea who else is on the team until they're standing next to them. It still has to coordinate, sometimes within minutes, on decisions that matter. A six-month project team disbands the day the project ships, and most of what it built together — the shorthand, the running jokes, the shared sense of who to ask about what — disbands with it. An open-source contributor fixes one bug, has a single exchange with a maintainer they'll likely never speak to again, and is gone. None of these groups has the luxury the rest of this cluster has been assuming: enough shared time to build a slowly-accumulated shared mental model, or to move a newcomer through legitimate peripheral participation toward full membership. The question worth asking isn't whether these count as real teams. It's how much memory, if any, a genuinely temporary group actually needs before it starts behaving like a community rather than a crowd that happened to be in the same place.

Edmondson's "Teaming," Not Teams FoundationalKnowledge that endures for decades — core principles

AmyAmy Edmondson coined "teaming" as a deliberate verb to describe coordination that has to happen without the stable team structures — fixed roles, a settled hierarchy, time to build rapport — that most accounts of teamwork quietly assume¹. Teaming works out who does what in real time, often with people who have never worked together and may never work together again, and it depends on a different skill set than stable teamwork does: fast, low-cost ways of signalling competence and intent, rather than the slow accumulation this cluster's page on shared mental models describes for an established pairing. A surgical team assembled fresh for one operation, drawing on staff who rotate through different combinations constantly, is Edmondson's own central case — and it works, when it works, not because the specific people have history together, but because each of them brings a pre-built, standardised model of the procedure and of what a competent colleague in each role should be doing, acquired from training rather than from this particular team's own shared past.

Positioning Yourself When the Structure Won't Hold Still Applied / MethodologicalKnowledge with a 5–10 year half-life — stable practice

AA stable team's members don't need to actively present who they are to each other — the slow route described earlier in this cluster does that work for them, over months or years. A member of a fast-forming or loosely-structured group doesn't have that luxury, and has to do deliberately, and quickly, what a stable team gets for free. This is the practical problem a piece of work closer to home addresses directly: the "This Is Me" project, developed at the University of Reading, treated a person's visible digital trace — what shows up about them online, to people who have no other basis yet for judging who they are — as something worth actively managing rather than leaving to accumulate by accident, particularly for anyone whose work depends on being legible to people encountering them for the first time². Nancy White's later adaptation of the same material for NPO and NGO contexts reframes it explicitly around exactly the groups this page is about: people who show up in a loosely structured, often transient online community — a volunteer network, a project-based coalition, a professional group with high turnover — and need the rest of that community to form a usable judgement of who they are and what they can be trusted with far faster than a stable team ever has to³. A clear, honestly presented digital identity is, in this reading, the transient-community equivalent of the shared mental model a stable team spends months building: a shortcut to a working judgement about someone, substituting deliberate self-presentation for time nobody in the group has.

How Much Memory Does a Temporary Team Actually Need? FoundationalKnowledge that endures for decades — core principles

AA temporary team can perform like an experienced one if its members bring a shared model from somewhere else, rather than needing memory of this specific group's own history at all: standardised training, a common professional background, a protocol everyone in the room has independently learned before ever meeting each other. That's the general answer Edmondson's surgical-team case already demonstrates. What a temporary group with no such common background has to build, it has to build through exactly the fast self-presentation the previous section describes, compressed into the first few minutes or exchanges rather than spread across months. A group becomes a genuine community, in the sense this cluster has been using the word, roughly at the point enough of its members hold a converging, mutually legible model of who the others are and what they're each good for — whether that convergence came from years of shared history, from standardised training everyone brought with them, or from a fast, deliberate act of self-presentation that let strangers skip most of the slow route.

Strongest Objection: Some Swarms Need No Identity or Memory at All FoundationalKnowledge that endures for decades — core principles

AA genuine swarm — an ant colony laying and following pheromone trails, or a multi-agent AI system coordinating through a shared blackboard rather than through any one agent's model of the others — can produce impressively coherent collective behaviour with no individual memory and no identity-tracking whatsoever. This site's page on multi-agent systems covers this kind of coordination directly. This looks like a counterexample to everything above: if a swarm needs neither memory nor self-presentation to coordinate well, maybe human transient teams don't really need it either, and the perceived need is just an artefact of human psychology rather than a real requirement of the coordination problem. The gap is what a swarm is actually solving. Stigmergic coordination — leaving a trace in a shared environment for others to read, rather than building a model of any specific other participant — works precisely because no individual agent's identity or history matters to the outcome; any ant following the same trail does equally well. A human emergency-response team, a project coalition, or a volunteer network is coordinating around decisions where who specifically is making a given call, and how much that person can be trusted with, matters a great deal — which is exactly the condition under which identity-positioning earns its keep, and exactly the condition a true swarm has engineered itself out of needing.

Provisional Conclusion FoundationalKnowledge that endures for decades — core principles

AA temporary team doesn't need its own long shared history to function like a community — it needs a shared model its members can converge on quickly, built from standardised training, a common background, or deliberate, fast self-presentation, substituting for the slow route a stable team takes for granted. Where none of those substitutes is available, a group stays a crowd regardless of how long it's nominally been working together, and where a true swarm's stigmergic coordination removes the need for identity altogether, it's solving a different problem than the one most transient human teams actually face.

Questions for Further Thought

• Is deliberate digital-identity management a genuine substitute for a stable team's slowly-built trust, or only a thinner approximation that works until something goes wrong and the approximation is tested?
• Could a transient team deliberately adopt some of a swarm's stigmergic tricks — leaving traces in a shared environment rather than building models of each other — for the parts of its coordination where individual identity genuinely doesn't matter?
• Does standardised training that lets strangers coordinate well (Edmondson's surgical team) ever fully substitute for a team's own history, or does it only cover the routine cases, leaving a team with no shared past more exposed exactly when something unprecedented happens?

Related Topics

• Shared Mental Models: Why Experienced Teams Need Fewer Words — what a stable team has time to build that teaming substitutes for, and what a transient team can borrow instead.
• Communities of Practice: Learning by Belonging, Not by Being Told — the slow route to legitimate membership this page's transient groups don't have time to travel.
• When Should a Decision Be Delegated, Not Just Automated? — the trust judgement this page's fast self-presentation is trying to support without the usual time to build it.
• Multi-Agent Systems: Coordination, Communication, and Strategic Interaction — the stigmergic, identity-free coordination this page's strongest objection draws the contrast against.

Further Reading

• Edmondson, A. C. (2012). Teaming: How Organizations Learn, Innovate, and Compete in the Knowledge Economy. Jossey-Bass.
• Williams, S., Fleming, S., Lundqvist, K., & Parslow, P. (2012). This is me: Digital identity and reputation on the Internet. In S. Warburton & S. Hatzipanagos (Eds.), Digital Identity and Social Media (pp. 104–117). IGI Global.
• White, N. (2009). Digital Identity Workbook for NPO/NGO Folks. Adapted from Williams, Fleming, Lundqvist & Parslow's "This Is Me" materials. fullcirc.com.

References