random tought: I'm taking notes on roam research about a talk from a video, I would love the notes to have the timespan of the video when I took them, so I can jump there later

Didn't one of the CDG / Dynamicland folks annotate a video like that?

This is one of the many potential uses that would be easy in a hypothetical 'no apps/silos' world. Draw gesture to connect the 'video' object to the 'note' object and customize the binding.. wait this is #C5U3SEW6A I should stop.

my thought was random, go on... :D

This reminded me of a talk by Alan Kay where he takes one 'app' - an animation - and connects it to another app - a picture editor. By exposing the methods of the animation and live binding it to the editor, he is able to construct an ensemble where one frame of the animation can be live edited in the picture editor, while the animation is playing. The main point was that these 'apps' were not designed with each other in mind, yet were able to work together. These were not apps, really - they were just objects with views. For them to work together, they were both introspectable (could see their methods), built from the same substrate (same smalltalk system) and shared the notion of some types (e.g. a picture itself).

I've been thinking about the same but using another metaphor, fractal applications. Applications all the way down! what if every single part of an application was also a self contained application, whatever view of the application you are using could be just a part of other applications for other people, applications have no root or center. Instead of getting rid of applications we make them reusable.

@Ivan Reese We actually made several prototypes at CDG about watching videos in a group physically together where everyone is watching and writing in the chat. Bret made big posters of these artifacts later. You could laser any part of the poster and it would jump to that part of the video: