The collective braindump (editing)

The braindump is the editing mode of a map — a canvas where you add nodes, draw connections between them, and arrange everything by hand. It’s well suited to workshop-scale maps (roughly 20–200 nodes).

The editor

Opening a map shows the canvas with a sidebar on the left, a toolbar in the header, and a details panel on the right:

The sidebar has three tabs — Nodes, Connections, and Import/Export. Nodes and Connections always list everything on the map — selecting a node or connection on the canvas doesn’t change which tab you’re on or hide the list; its detail view opens in the details panel on the right instead. Each of the two collection tabs also manages the field definitions for its own kind of thing, so a field’s definition sits next to what it’s used on (see Managing fields below). Import/Export holds the CSV/GraphML import flow (see CSV import). Collapse the sidebar with the panel toggle at the bottom of the icon rail.
The Nodes tab splits into Placed and Unplaced sections, each collapsible and paginated (200 per page — see placed and unplaced nodes below), with a Node fields section pinned below them. Unplaced starts collapsed — an import can leave thousands of nodes in it, and it’s a staging area you dip into deliberately; click its header to open it.
The toolbar, next to the mode switcher, has Add Node (click, then click the canvas to place it) and Neaten, which snaps every node to a grid to tidy up a messy layout.
The details panel on the right opens the moment you click a node or connection, prompting you to select one before that. Close it with its ×, the Escape key, or by clicking empty canvas — it collapses to a slim labelled edge on the right; clicking that edge, or selecting something new, reopens it.

Selecting a node highlights its direct neighbours and their connections so you can trace a relationship at a glance; everything else fades to a plain dashed line.

Placed and unplaced nodes

Every node on a map is either placed (has a position on the canvas) or unplaced (belongs to the map but hasn’t been given a position yet). Nodes you create by hand are placed immediately; nodes brought in by import land in the Unplaced list instead, so a bulk import of hundreds or thousands of nodes doesn’t dump them all on top of each other — you can work through them at your own pace, or leave them unplaced and use visualisation instead.

Place a node by dragging it from the Unplaced list onto the canvas, or by clicking it — the tool picks a sensible spot near your existing nodes automatically.
Place all places every unplaced node at once, laid out in a tidy grid. It’s disabled once the unplaced list (or the resulting total once placed) would exceed the 200-node edit limit — place or remove some nodes first.
Remove a node from the map without deleting it (unplace it): right-click it — on the canvas or in the Placed list — and choose Remove from map. It reappears in the Unplaced list; its connections and data are kept.
Remove all unplaces every placed node at once — the reverse of Place all.
Clicking a node in the Placed list opens it in the details panel, exactly like clicking it on the canvas — it never deletes or unplaces the node, and the Placed list stays visible. Opening a node this way also pans and zooms the canvas to bring it into view.

What you can do

Add a node by placing it on the canvas and giving it a label. A new node opens in the details panel with the label field focused.
Move nodes by dragging — positions you set are saved per map.
Connect nodes by dragging from one of a node’s four handles to another node. A node can’t connect to itself; drawing is instant and doesn’t force you to pick a type up front, so a fresh connection is dashed and labelled Undefined until you give it a reason or a field value. Any number of connections can exist between the same pair of nodes — each observing a different field — and they always fan out into separate curved lines so each one stays visible and clickable.
Grow the map from a handle — drag out from a handle and release on empty canvas to create a new node where you let go, joined by an undefined connection, ready to label and define.
Edit a connection by clicking its line or its label — either opens it in the details panel with a Reason / Field switch, each side named after what it writes. Reason is freeform text (“why are these connected?”) — zero setup, no field required. Field assigns a Field and its value (a number field or a true/false toggle depending on the field’s type); the panel remembers the last field you picked this session, so assigning the same field to a batch of connections only takes one pick. The switch is only about which one you’re editing — a connection can carry both a reason and a field. Direction isn’t something you set per connection — it comes from the assigned field itself, and the panel shows the resulting reading in plain language (e.g. “Alice funds Bob”) along with a note to change it on the field definition. Direction is shown as an arrow on the connection line; the connection’s colour and line style reflect whether it’s reason-only, numeric, or true/false. The two nodes it joins are shown as a pair of clickable buttons — click either to jump to that node. Once a directed or mutual field is assigned, a Swap button appears below them — if the connection was drawn the wrong way round, use it to flip which node is which, the only way to correct it now that direction always follows source → target. A small “Show in-use fields” link under the field picker switches the sidebar to the Connections tab, where you can see which fields are already used on this map and edit their definitions. An Add another connection button below draws a second (undefined) connection between the same two nodes and opens it for defining — a shortcut to adding more connections between a pair without drawing a new line on the canvas.
Edit a node’s properties — its label plus any number of freeform name/value properties — from its details panel entry. The same panel has a Fields section for assigning node fields: pick one from the dropdown to add it with a default value, edit its value inline, or remove it; a New field button lets you define one on the spot without leaving the panel. A node’s Connections are also listed there, each split into two clickable parts — the connection’s field on the left (jumps to it in the details panel) and the other node on the right (jumps straight to it).
Delete nodes and connections from the details panel, or with Delete/Backspace once a node or connection is selected. Deleting a node removes its connections on this map.
Manage fields from whichever collection tab the field applies to — see Managing fields below.

Managing fields

A field is quantitative or qualitative data you can compare across the network — a number or a true/false flag. Every field applies either to nodes or to connections, and it’s managed from that collection’s own tab, alongside the things using it. There’s no separate Fields tab.

Defining a field asks for a name, a value type, and an optional category and description. A connection field also has a direction — undirected, directed, or mutual; a directed field shows a live preview (e.g. “[Node A] funds [Node B]”) and takes an optional inverse name (e.g. “funded by”) for the reverse reading. You never have to say whether a field applies to nodes or connections — that’s decided by which tab you create it from.

System-defined fields are shared across organisations. They’re marked system on their detail line and can’t be edited or deleted, though you can still hide or remove the connections using them.

If a field is already in use, changing its value type or deleting it asks for confirmation first. Deleting a field removes the node values that used it and un-defines (rather than removes) any connections that used it — unless they also carry a reason, in which case they stay defined.

Connection fields

The Connections tab is the connection-field list. Each row shows the field’s name, its value type and direction, and how many connections on this map use it — so the count and the definition are in one place. Clicking a row highlights those connections on the canvas.

Each row’s ··· menu (right-clicking the row does the same) covers both scopes, and says which is which:

Edit field… and Delete field… change the definition for your whole organisation.
Hide connections on map and Remove N connections from map affect only this map. Removing connections keeps the field definition itself.

By default the list shows only fields with at least one connection here; Show N unused fields reveals the rest of your organisation’s catalogue as dashed rows with a count of zero, so you can edit a field before you’ve used it.

Connections carrying no field are listed individually below, named after the nodes they join (e.g. “Alice - Bob”): under Reason only if they have a reason, and under Undefined connections if they have neither. The Connections tab’s icon carries an amber badge counting the undefined ones.

Node fields

The Nodes tab has a Node fields section pinned to the bottom of the panel. It stays visible however long the Placed and Unplaced lists get, and it never grows past about 40% of the panel — past that it scrolls on its own — so neither the lists nor the fields can push the other out of view. On a map with more than 25 nodes it starts collapsed to leave the lists more room; expand or collapse it yourself and your choice sticks while you stay on the tab.

Each row shows the field’s name and value type, with edit and delete buttons; fields with no values on this map are drawn with a dashed outline. Note this section defines what node fields exist — assigning a value to a particular node happens in that node’s own details panel on the right.

Undo, redo, and recovery

Your edits are tracked as a session behind the scenes today — every change you make is recorded — but undo/redo, the idle-recovery window, and collapsing a session into a restorable revision are coming with revision history, not yet available.