Freehand drawing annotations #4685

Was curious if anyone else has thought about adding the ability to make annotations using freehand drawing. Simple line drawings could be rendered as SVG elements just like highlights are done presently, but the challenge is that you don’t have text to anchor the locations of the elements to. If you make a freehand circle around an area in an HTML or PDF document, you’d want that circle to bound the same area for all users with potentially different screen sizes.

I think you could use relative coordinates adjusted for window size so that the backend stored x and y of a drawn SVG element are between zero and one. Then when you draw them on the client you transform to account for the user’s window size.

Curious to hear if there are other ideas on how to provide for this functionality. It seems like PDFs may be easier to handle since there are several tools out there for annotating them with hand drawings. The HTML case seems harder, but if hypothesis could do it and sync the updates in real-time like it already does for text annotations, it’d be super useful for remote collaboration on docs with lots of images and plots that don’t have text to anchor an annotation to.

Also interested in hearing about any other existing open source tools that provide text and freehand annotations of HTML.