Line, Column, Row, Rectangle, Grid

All primitives are just a way of indicating a group of term.Pixel. A Line can be a collection of term.Pixel which can be horizontal, making it a Row or vertical, making it a Column. Using the LineFromPositions function from geom package, you can have non typical lines, like a diagonal or other arbitrary positioned line. Now, thinking about Rect : it's just a thicker Line, isn't it? The only particularity of a Rect is the orientation : a Rect can be organized vertically, making it an array of Rows - or horizontally, making it an array of Columns. A Grid is a collection of `Rect', right ? It is only special as it is "flexible", in terms that it should resize it's children to the desired sizes.

Center of a rectangle

Page and it's context.Context

When a Page gets created, a cancellable context.Context should be passed to it, so it can free the collection of term.Pixel it orchestrates. Stepping back a little, a Page it's just a Rect which just fills the entire screen. Another particularity of a Page it's that will listen for term.ResizeEvent in order to fill the entire screen. Another important one is inheritance of BackgroundColor from the Page beneath itself. Last but least, a Page should be able to keep a stack of owners of a term.Pixel (think Dropdown or Autocomplete needs to "borrow" some pixels from a neighbour for a while).
Page listens for mouse events and dispatches them only to Rectangles that are in that position.