Skip to content
#edge computing Open access

How Do We Form a Safe Swarm House? From Matter to House, from House to Private Property, from Property to Distributed Intelligence — Twelve Steps of Formation, Four Tenure Invariants, Seven Safety Invariants, Forty Counterfactuals and a Runnable House

Sep 2026 · Zenodo (CERN European Organization for Nuclear Research)

Abstract

A dwelling is the smallest object that holds territory, boundary, access, exclusion, ownership, privacy, shelter, safety, services, activity, memory, identity, value, inheritance and permanence at the same moment. This article asks how such an object forms from many simple physical and computational elements, and what remains of ownership once its geometry is free to change. The answer is built rather than described. One structured file declares a house as modules, edge slots, capabilities, sensors, actuators, activity contracts, a ledger of holders and a set of safety invariants; four layers read that same file. A relational layer answers every question about the house as a query over its graph — zones, travel distance to a way out, load paths to ground, and what a zone offers in area, span, illuminance, air change rate, thermal band, acoustic separation, water and surface. A swarm layer gives every cell and every slot an agent that holds its own state, talks where it touches and reads nothing wider: a zone label, a distance to a way out, a set of capability holders and a proposal with its votes all converge by local agreement, and a quorum commits. A safety kernel sits below both, with seven invariants answered at proposal and again at commit. Property is written as four invariants over a ledger — identity, capability envelope as a product of ratios, access, and severability — so that a reconfiguration which changes the plan entirely leaves tenure intact, while one that takes away a holder's means of acting is refused. Twenty-two logged rows report what the house does when its busiest agent leaves, when a third of its actuators stop, when its network splits, when every identifier is scrambled, when the power goes, when the occupant starts using a wheelchair, when five columns are lost and when the exits are locked. The same rules run on three plans of different shape, cell size, module count and topology, measured against a planner that sees everything. Activity conditions carry published figures: 300 lux of daylight provision, four air changes an hour of purge ventilation, an operative band of 20 to 24 degrees, 45 decibels between dwellings, 850 millimetres of clear door width and a 1500 millimetre turning circle, the last of which sets the cell at 1.5 metres. A parameterised cost model then asks whether a house like this is cheaper and faster, and answers in the unit a household lives in: twenty-six contract-hours a day cost 0.79, 0.54 and 0.27 dollars a contract-hour across a room-per-contract plan, a conventional plan and a reconfigurable one, while the reconfigurable plan is the dearest of the three per square metre. The answer turns on one assumption — a household accepting about 108 per cent of the conventional plan's area or less — rather than on the price of a module, which would have to reach forty-eight times the figure used here before the ranking changed. On speed the claim is about a change: one panel moves in 0.4 minutes against eleven hours of trade work. The article closes with a room that can be built from named parts, and with the thirteen ways the whole proposal could be shown wrong. Melo de Magalhães, Planta Smart Homes, 2026 1 Key points · A dwelling is treated as a temporarily solidified configuration of relationships that makes human activities passible, decomposed as matter, relationships, boundary, activities, safety and time. The decomposition is conceptual; every quantity beneath it is computed. · Formation runs in twelve steps from matter to a held claim, and each step is a relation two elements can form where they touch rather than a plan held above them. · Property is reframed from coordinates to a ledger over modules, capabilities and access, conserved by four invariants that a reconfiguration must preserve and that a capability-taking change fails. · A swarm house is defined by what its elements expose — state, relations, capabilities, constraints, perception and passibility — and by the absence of any element that must see the whole. · Activities enter as contracts on conditions rather than as layouts or device commands, so the configuration stays for the house to find: possibility, then passibility, then accessibility, then activity. · Safety sits below intelligence in four layers, with seven invariants that outrank every proposal and answer twice. · Twenty-two logged rows, three plans and forty counterfactuals report what holds and what gives, including the contract this house de Author ORCID: 0009-0008-6255-7724. License: Creative Commons Attribution 4.0 International.

View source

Similar papers

#computer vision Review Sep 2017

Agile Software Development Methods: Review and Analysis

This publication proposes a definition and a classification of agile software development approaches and analyses ten software development methods that can be characterized as being "agile" against the defined criterion.

P. Abrahamsson, O. Salo, Jussi Ronkainen et al. · 727 citations · ⚡54
#computer vision Jun 2008

The impact of agile practices on communication in software development

The study shows that agile practices improve both informal and formal communication, but indicates that, in larger development situations involving multiple external stakeholders, a mismatch of adequate communication mechanisms can sometimes even hinder the communication.

M. Pikkarainen, Jukka Haikara, O. Salo et al. · 401 citations · ⚡48
#machine learning Review Open access Oct 2014

Software development in startup companies: A systematic mapping study

The results indicate that software engineering work practices are chosen opportunistically, adapted and configured to provide value under the constrains imposed by the startup context.

Nicolò Paternoster, Carmine Giardino, M. Unterkalmsteiner et al. · 394 citations · ⚡54

Related blog posts

Microsoft Research Blog Sep 29, 2026

Introducing Quine: An AI research system designed for the complexity of biology

Biology doesn't operate in silos, and neither should the AI representation of it. Quine is an early-stage research effort to create a multimodal world model of biology. By connecting insights across biological scales and modalities, Quine helps scientists computationally search a space far larger than intuition allows and prioritize hypotheses before they reach the lab. Experimental results provide important feedback, helping researchers sharpen future research directions. The post Introducing Q…

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.