Re: [vwrap] one question

[Morgaine <[e-mail]>]

The "complex system" behind the URL is OpenSimulator --- you've explained
that your student is using that, and since you've been working on
OpenSimulator for a long time as a primary developer, that is of course what
we expected anyway.

But if it's OpenSimulator and not some other kind of unspecified Web app
world, then the many responses you have been given about your use case being
supported by VWRAP seem well founded.  What's more, since VWRAP has
specifically had OpenSimulator in mind from the start, it is hard to see
where the alleged problems lie.

I'd like to help more, but I'm thwarted by not being shown the actual
problem.

One of the best aspects of OpenSimulator is that it can support multiple
protocol stacks concurrently.  That's a great feature, and it seems
tailor-made for introducing a VWRAP protocol module at that interface,
perhaps configured for a minimal deployment pattern offering only teleport
and remote asset services accessible via region proxy.  Web-browser based
residents of your world could then TP to another VWRAP-enabled world, buy
some junk, and bring it back with them.  A win!

I see no showstoppers just because your native client is a Web browser.  It
would work for travellers in the other direction too.  The reason why all
this can be expected to work is that OpenSimulator runs a data model that is
compatible with the data model used by VWRAP asset services.  This is why we
say that we can't see an *a priori* problem.

If I may suggest a way out, perhaps we could continue with examining other
parts of the puzzle, and when something arises that might not work with
those "web-based virtual worlds" you could raise a red flag for us?  Asset
services need to be an early item of discussion anyway, so we ought to see
quite soon if something is askew in the model.


Morgaine.




======================================

On Sat, Sep 25, 2010 at 1:27 AM, Crista Lopes <[e-mail]> wrote:

>  No one is talking about web pages. I'm talking about web applications:
> what's behind an URL is a complex system -- like Facebook, but with a 3D
> component to it.
>
>
> On 9/24/2010 5:21 PM, Morgaine wrote:
>
> Sure, I've seen that Quake II demo in a browser.  Nice, and very promising.
>
> That doesn't help us here though.
>
> As I wrote, virtual worlds are far more complex than web pages, and can't
> be handled in the same ad hoc fashion as web pages if we expect any useful
> amount of interop to occur.  Browsers becoming more powerful doesn't modify
> this situation at all.  Complexity doesn't decrease just because you can
> process code faster with newer technology.
>
>
> Morgaine.
>
>
>
>
> =================================
>
> On Sat, Sep 25, 2010 at 12:37 AM, Crista Lopes <[e-mail]> wrote:
>
>> On 9/24/2010 2:05 PM, Morgaine wrote:
>>
>>> Virtual worlds are far more complex than web pages, and can't be handled
>>> in the same ad hoc fashion as web pages if we expect any useful amount of
>>> interop to occur.
>>>
>>
>>  Morgaine, just in case you missed this:
>> http://googlewebtoolkit.blogspot.com/2010/04/look-ma-no-plugin.html
>>
>>
>
>