[sldev] AWG: Geometric Content Creation

[Dzonatas]

Mark Burhop wrote:
>
>
> On 10/29/07, *Tao Takashi* <tao.takashi at googlemail.com 
> <mailto:tao.takashi at googlemail.com>> wrote:
>
>     2007/10/29, Erik Hill <zortiger at gmail.com
>     <mailto:zortiger at gmail.com>>:
>
>         I think it would be wise to do some graphics upgrading, but
>         the way i see it, no matter what is done its going to be
>         outdated about the time its completed.
>
>
>
>     I'd think that instead of thinking about all possible graphics
>     standards it would make sense to simply add some field in the
>     protocol describing the graphics/object model used eventually with
>     a version number. Then some general blob of data can follow which
>     needs to be interpreted by the individual model engine.
>
>
> In principle, I agree. However, I don't think you can just open the 
> flood gates to all formats as bandwidth is an issue. Unlike the web 
> where an HTML page might reference an AVI or PDF file but you don't 
> actually have to load it, a 3D scene is more complicated.
>
> For example, what do you do if you go to a location with a hugely 
> complicated automobile engine in some special format? I can think of 
> some engineering use cases where this would be great and some casual 
> use cases where, depending on the implementation, this might be 
> terrible ( i.e. sluggish viewer response, missing data, regions which 
> support only certain formats. etc.).
>
> I'm not saying not to do this, just that I think this is a place for 
> the architects to really earn their money (or "kudos" for those not 
> paid :-)

Quite awhile ago I made a simulation where, from the outside, you saw a 
computer, and you could go into it and interactively swap the virtual 
hardware modules around. It went deeper to where you could go into the 
modules and make changes to the processes and software. Of course, it 
went further then that. The simulation was run on an old computer with 
basic graphics. It would put you into a virtual world where the computer 
is found. The neat thing about this is that as changes where made to the 
computer inside, it immediately affected the virtual environment of the 
world outside. In a sense you could say the virtual computer ran the 
virtual simulation of the virtual world.

The most important thing to note: changes to simulated computer/world 
did not require a restart to make major differences in the entire 
virtual world.  Only the objects directly or indirectly affected by the 
changes recompiled as needed.

That is a sample of what paravirtualization can do.

One dream I've had for a long time is to introduce the concept of jobs 
to work inside the virtual world to let real people design, redesign, or 
even fix software and architecture.

Now, here is Second Life, and it now has real world people and 
temp-agencies in-world. What if, the virtual world provides a method to 
find all areas of the virtual world that need to be addressed, and then 
immediately hand any determined issue over to a job queue. What if, the 
in-world temp-agencies offer services for those jobs that pop up in the 
queue.

I noticed that CSI:NY suggested some potential in-world jobs for several 
forensic fields.

Bandwidth is an issue, but we have enough to merge in this kind of 
architecture (I described) and underwrite/transform yesterday's 
mainstream notion of virtual worlds -- to at least start somewhere.

-- 
Power to Change the Void
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.secondlife.com/pipermail/sldev/attachments/20071029/cbd1cac9/attachment.htm