Dec 11 2008

A time for OpenSim reflection – standalone Linden servers on the horizon

Silent though these pages be, much has been thought.  I’ve had some quality time with inquisitive Lindens and learned to expect some sort of standalone Linden server product along about 2009.  For me, that’s a game-changer as it’s hard enough to suggest (at work) creating content without also keeping up with an open source thread to stand that content up upon.

This past week I’ve made a real-life geographic shift for a family event, and learned that I’ve got a relation involved in the study of architecture.  That insight has reinvigorated my interest in Jon Brouchoud and some of his writing here.  The notion of architecture as it is currently an academic subject, versus architecture as a current professional practice, and the disruptive possibility of widespread virtual world deployment—this is a notion not so different from geographic science as an academic subject, GIS as a professional practice, and the possibility of immersive 3D disruption of the status quo.

Others in academic circles, including University College London, Centre for Advanced Spatial Analysis (CASA), published a Working Paper about the time this past summer when I was so focused on my 1:1 immersive build.  It was gratifying to see the CASA acknowledge Second Life technology’s place in the world of neogeography and geospatial informatics.

Sitting side by side yet somehow abstracted from mapping, gaming and digital earths
is Second Life and other similar virtual environments. Second Life and their like are
easy to dismiss as pure distraction and entertainment. Yet look under the lid of
Second Life and it contains one of the most powerful geographical data visualisation
kits available

And the fine writing and attention to detail of Jon / Keystone was spotlighted in NY Times’ Style magazine this past weekend.  It was a pleasure to share that link with architecture students!

It’s a big world, and immersive 3D systems must balance the tradeoff between quality and performant physics, and an economically practical level of large land areas served up to relatively sparse users, if we are to identify applications that consume vast tracts of GIS data.  Spanning that scale will require that standalone Linden servers have the ability to shortcut some of HAVOK’s demands to pile in many more than four regions per physical server.  After all, if I can get 100 regions stood up on a 1 GHz Celeron using OpenSim, then a four-threaded dual 64-bit Xeon server really ought to do the same for standalone Linden regions, right?  I surely hope so.

One response so far

One Response to “A time for OpenSim reflection – standalone Linden servers on the horizon”

  1. […] is still so much wondrous creation to behold that some speculate it may even affect the cultural expression of architecture in all realms. I certainly has changed […]

Trackback URI | Comments RSS

Leave a Reply

You must be logged in to post a comment.