Smess iswas an idea for a chat app inspired by conversations to discuss ideas. These conversations often occur deliberately, so someone could - in principle - open some Smess chat for discussion of idea(s), no different from setting up a meeting, and record a multimedia conversation between two people over some discrete length of time. This could be hacked together with systems like [[matrix.org]], [[https:meet.jit.si][Jitsi Meet]], and other such systems with rudimentary voice recognition.
A canvas-based conversation is worth exploring, but we deliberately aboid adding complex group infrastructure or management features - this is purely for hashing out an idea synchronously. It's a space to focus. [cabal] seems to provide similar ideas and facilities.
After this meeting is over, we automatically archive the chat; no more links can be created in order to prevent endless idea drift and growth. (Looking back, I think this is much more of a self-discipline issue than one that can be solved with technology - we can create the right space for thought, but we can't necessarily enforce discipline).
After this archival process, all of the links to ideas in the chat still work and new links can be created between existing thoughts and ideas on the page. The archived chat may have a different UI that helps browse history in a better way. Revisit [Chapter 1 | Chatting with Glue] for more inspiration here...
Ideally, hashing out ideas feels instant and spontaneous - like you would with a video call or meeting - but everything is in a controlled environment that can be later traced and documented. Inspired by synchronous meetings and mailing list archives - we want the longetivity and traceability of a mailing list archive with the real time impact of synchronous meetings and the sponteneity of a viral tweet (share idea from a chat on twitter - > implicitly invites others to chat - > 1 click signup to get into the chat room - no sign up concerns).
This is probably still worth exploring as an experiment, but I don't think there is a viable product here.