One of the things I've been looking at has been tying up the virtual "worlds" that are imagined, with different ways of visualising them. The current SIMPLE platform has a built-in map visualiser which takes rudimentary Cartesian coordinates, and drops them on top of a "map" graphic.
But there's really rich applications out there, such as Google Earth & Maps, which are perfect for displaying geographic information. I know most SIMPLE simulations will exist in an un-real location (non-real?) but this doesn't have to be the case.
So as an experiment I've built a KML file containing some of the entities from one our simulations, inserted it into a simulation instance so that It can be served from within SIMPLE into Google Earth via a Network Link. And lo...it works!
This is all well and good, but I don't want a static file that is the same for every instance of the simulation, I want the unique client name from each simulation to be in the KML file as well...so a little bit of tweaking of the SIMPLE download code, and we have now SIMPLE variablisation system performing replacements on the SIMPLE variables used in the KML file.
I know its a bit of a teaser, but I'll post up a publicly accessible version of the tech demo sometime.
The mundane thing here is however, that I've built the KML using Google Earth, which I've then simply added to a SIMPLE simulation blueprint as a resource, as opposed to having SIMPLE generate the KML from the entity information stored in the blueprint. Yes it is a bit of a redundancy and using both the built-in viewer and the GE version could end up being contradictory, and broadly speaking the KML data could be generated from information set within the blueprint itself, but we'd have to extend the information stored in the blueprint to hold all of the other meta-data required to locate the entities properly (such as Viewport spec). AND GE offers a fairly intuitive drag'n'drop interface for putting down your place markers.
No comments:
Post a Comment