We are playing with how to position context menus.

Here is how that looks like:
- we start with an example that creates a menu object
- the Playground inspector view gives us a draggable point related to which we open the menu.
- by dragging the point we can see how the menu positioning will react.

What if visual anchors were delicious?

Anything in a software system can be seen in multiple ways. Actively choosing the perspective can accelerate how you answer questions about the system.

For example: how did a file evolve? Looking at commits tells you when/who changed it.

But: did the file grow? A new perspective answers better.

When we execute a piece of code we might want to inspect the result at a later time.

A user interface solution is to remember the execution result in a visual element next to the code. And the cherry on top would be to highlight the code when hovering over that attribute.

In our quest to make creating custom views more and more inexpensive, we now got to the point in which we can create a view based on the object obtained from executing a snippet in the contextual playground from an inspector.

Sounds complicated? It's really not. Just take a look.

1/

How is a post with an embedded link described in the AT Protocol data?

Let's inspect it.