This is the roadmap and brainstorm for development
- B scripter GUI (improvements to interface)
- B/E V8 closures (fix storage of functions in v8 to save closure as well as function text)
- E/J CDN support for music (serve music files as well as meshes)
- Backups of OH data
- Repository for world-specific scripts (avatar script, anything we expect to be in there by default).
- Build a nice demo scene, large enough to require some exploration to see it all.
- Scripts for building into package (i.e. all the data required to support Sirikata URIs
- Get a few good, tested avatars on the CDN and get the default demo avatar to use these simple animations well
- Physics improvements -- get avatars on terrain working well
- Generate collision event messages for objects, maybe based on subscription request
- Expose progressive meshes
- Progressive mesh loading
- Improve Emerson storage -- make it much easier to interface and get a persistent object
- Improve object manipulation interface (could be in deployment specific code, i.e. under Demo World)
- Object migration between OHs
- Fix closures/provide full snapshotting (timers/register handlers/etc)
- libmesh filters should not strip animations (eg. check if joints before collapsing two vertices).
- Playing multiple animations at once (blend/interpolation)
- Exposing physics collisions to Emerson.
- Fix Gui isolation
- Better OH connection failure handling
- Reduce Emerson memory usage
- Storage tutorial
- Sandbox tutorial
- Getting started with demo, maybe just update getting started for users since a lot of steps can be avoided when someone provides the configuration for you.
- Simple demo videos showing how to join and interact with the world/other users.
Long-term Goals & Research
- Add audio support (purely OH-based to support, e.g. simple sound effects and local voice chat, then consider adding mixing support to the space server).
- Load balancing objects across object hosts
- Distributed physics -- extend physics to work with objects including boundaries
- Mesh/object aggregation
- New transport abstractions targeted at VWs (e.g. last reliable)