Re: [vwrap] Comments on http://tools.ietf.org/html/draft-ietf-vwrap-intro-00

["Hurliman, John" <[e-mail]>]

> the hypergrid solution is to define a single virtual world. regions
> live in ONE place on a single global map. there's a single adjacency
> map and everyone gets the same map. authentication services do not
> direct clients to different space servers, there's only one.
> 

It could be another terminology mismatch but I don't think that's accurate. In the Hypergrid model, a collection of simulations share an adjacency graph by connecting to a single space service to form a "world" or "grid". But that space service can also point to simulations in other adjacency graphs or worlds/grids that have a different space service (and live in another administrative domain). So a region lives in one place in the local map and in a different place on other foreign maps. It's vaguely similar to how web hyperlinks work which is where the Hypergrid name came from. You could also change a few protocols bits if you wanted to always communicate with the space service in the same administrative domain as your authentication service and you would get the same behavior as what you describe for VWRAP.

The rest of your e-mail cleared things up; just a terminology issue as I suspected. I have some concerns that we're all in basic agreement but disagreeing over terminology still. No one can point to an actual use case where a participant on this list is saying "I don't think that should be allowed or included in our standardization effort", but every e-mail going back and forth is a warning about ignoring the tourist model or dropping support for the walled garden model.

John