The idea is to have a panel that can be dragged and positioned anywhere in the Obsidian interface (e.g. in the sidebar), and function as a customizable command center, allowing the user to add buttons, inputs, menus, etc that can trigger custom actions when clicked.
This would allow creating macro pads, quick command centers, or interactive panels without needing to clutter the ribbon, title bar, status bar, etc, used by the Commander plugin. People could build custom interfaces tailored to their specific workflow needs.
This could improve interface clutter, help with limited space in the interface, persist across window view/layout/theme changes (e.g. hiding status bar, ribbon, fullscreen etc), give more flexibility over positioning, provide a larger interactive area, improve visual prominence.
I picture users implementing it in a similar way to the Photoshop interface to allow them to manipulate their knowledge with custom command panels, and even undock them from the sidebar and create floating pinned windows with commands in them.
What you suggest is already possible using Obsidian Advanced URI combined with html and css. However a plugin would make the customization process more accessible to normal users. In essence your idea is about look since you can already execute general actions very easily using Command palette and note-specific actions using QuickAdd macros or Advanced URI. You can pin commands in Command Palette.
Well the issue with using the command palette is that it’s not persistently visible and only comes up on demand. Going back to the photoshop example, I don’t know if I’d ever manage to master adobe software if everything were accessed only through a command palette, requiring you to know every available command or browse through many many commands. Having it visually available is crucial for reducing feature neglect.
I did not realize you could do this kind of thing manually using Advanced URI. I’m going to experiment with this, but definitely agree that a plugin would make the process a lot easier.
EDIT: so after playing around, it seems that approach won’t work. The problem is the buttons apply their commands to the currently active file, which is the page the button is on itself. I tried both selecting the “Don’t specify a file” and “current file” and set an “export to PDF” button but neither export the currently active pane but instead the button page itself (in this case “untitled 49”).