Image, video or text — on the DOTV canvas, selecting, renaming, copying, deleting and wiring all work the same way. If this is your first AI video in DOTV, reading this section once is the biggest time-saver there is: once it's done, the chapters after it only have to cover what each kind of node can do on its own.
Double-click the name to change it in place, then press Enter. The name sits above the card's picture, not inside it.
A node with no name of its own shows the first line of its prompt, and failing that a fallback like "Image node 3" — so one look tells you whether a node has been renamed.
Deleting after Select all asks you first — that one empties the whole canvas. And if you really did delete the wrong thing, there is ⌘Z.
| What you want | How |
|---|---|
| Copy / paste | ⌘C copies the selected nodes, ⌘V pastes. An image copied from somewhere else drops onto the canvas with ⌘V too; paste several files of the same type at once and they stack automatically by file type (see 2.5). |
| Copy the picture itself | Copy image in an image node's right-click menu copies the bitmap (not a link), so it pastes straight into chats or slides. Copy to clipboard copies a text node's text, or a media node's output URL. |
| Copy the wires along with it | ⌘D, and the nodes plus the wires between them all come in a second copy. Make a copy in the right-click menu does the same. |
| Drag off a copy | Hold ⌥ and drag a node to leave a copy behind (without wires); hold ⌘⌥ and the wires come too. |
| Delete | ⌫. Or right-click the node and choose Delete. |
Save to my assets in a node's right-click menu saves that node's output into "My Assets" (the multi-select toolbar has the same button): it opens the Save assets dialog — one row per item, pick the folder, rename, tag, then save. Generated media is recorded automatically anyway, so this one is mainly for uploaded / pasted material.
⌘Z undoes, ⌘⇧Z redoes. Creating nodes, deleting them, wiring, moving and grouping are all on the undo stack.
A few things are not on it, so think before you act: deleting a canvas, clearing a canvas (see 1.3), and restoring a version.
Put the mouse over a card and a dot appears on each side:
You can also select two nodes and press ⌘L to wire them; it runs from the one you clicked first to the one you clicked second.
When one side of a node carries several lines, those dots stagger up and down into a column — the line to the node above leaves from the upper opening, the one to the node below from the lower one, so several lines never pile into a single one end to end.
What can feed what is set — a line means the upstream content gets used downstream, so only combinations that are actually used can be wired:
| This node | Takes |
|---|---|
| Image | Images (as a reference), style samples |
| Video | Images (as a first frame / reference), effect samples |
| Text, script | Text, scripts, images, videos |
| Audio | Text, scripts |
| Video editor, frame study | Videos |
| Previz | Images |
| Style / effect sample | Nothing |
A wire that is not allowed simply is not made, and it tells you which kinds that node does take. The two dots follow the same table: the side that can take nothing does not appear at all (which is why a sample node has no port on its left), and in the menu from the +, the kinds you cannot go to are grayed out.
Double-click a node's picture and the canvas centers it and zooms it to a sensible size — much quicker than pinching your way there when there are a lot of nodes. On a text node the picture is the editor, so double-clicking drops the cursor straight into it.