[noduerme]

How is what that's doing any different from figuring out element A's desired position in the coordinate space of the body, or in the local coordinate space of element B, and then appending it to one or the other in an absolute positioned wrapper? I've written very large single page webapps that have dropdown menus with sub-menus that are almost entirely CSS with just small amounts of vanilla JS to adjust their screen positions. Sometimes a really lightweight library is useful, but requiring the whole React framework seems like total overkill. On the other hand, adding a new HTML tag for the purpose of popup modals seems unnecessary.