Dynamicland/Realtalk seems to be a really cool exploration of building immersive computing environments for scientific advancement and community building. The key idea seems to be about how to bring computing power out of the isolating nature of single-user screens/keyboards and into a flexible 3-D environment that makes use of real tangible materials to build dynamic interactive experiences. A key phrase I've head from their talks is the idea that they want to bring "computing to where the people are".

What really sells this for me is the fact that RealTalk OS is implemented in itself, as a collection of programs pinned to a wall, that can be edited and experienced in real time. This makes the whole system more convincing than a single demo of using projectors and cameras to implement a specific application. Programmability is often a huge obstacle for systems like this, where applications are complex beasts that require special knowledge and capability to build. It seems like the real value in realtalk is that the interface seems well designed enough to allow new users to envision and help create new applications.

Communal computing is a really inspiring idea to me. I think there's an interesting tension between the benefits of using tangible, physical materials - which are easier to share and collaborate on, and using abstract materials (e.g. simulating things inside of an AR world) which often add barriers to human interaction (e.g. not being able to easily see real human expressions). I do think that advancements in AI/LLMs could help augment systems like dynamicland greatly, by providing a more flexible computing substrate - simplifying the communication between human and computer, but also not getting in the way of human and human interaction.

I'd love to try some of these ideas out. A lot of my work is a bit lower level, so I'm not immediately sure how to bridge the gap between something like the ideas in dynamicland and something much lower level, but I'm at least convinced that this is a cool idea for enhancing communication between scientists. Maybe I can set aside some space in the lab next year to deploy a system like this...maybe if I build some applications to help read and digest new papers I could convince the powers that be to give me the equipment and space needed...