clicking on some parts and then asking the user to identify them, is not a suitable thing. what would be cool is, if a user could just stand in front of the cam and their measurements are automatically taken
it looks like this is already possible in the SDK
other automatic tools - i am not sure, if we need more computing power in the cloud
i think the missing js support means we need to reconsider and chose another project. but: we could still do this project later and if you are interested i can ask cat (when i see her in october) to send you a developer of google tango.