Re: [vwrap] Call for a vote on interop BETWEEN independent virtual worlds or not

[Meadhbh Hamrick <[e-mail]>]

On Mon, Sep 20, 2010 at 11:11 AM, Morgaine
<[e-mail]> wrote:
> Joshua, you're right, this subject has been recycled more times than we can
> remember.  What's more, EXACTLY as you wrote back in October 2009, on each
> occasion we arrived at the same conclusion, namely that no matter in which
> terms we describe our protocol, it will provide what ordinary users of VWs
> recognize as interop between their worlds.

um. what's this "we" we're talking about?

> In other words, if we have two virtual worlds such as OSgrid and 3rd Rock
> Grid, then if those worlds wish it, they could use VWRAP to enable an agent
> to teleport from one to the other and to retain avatar properties and assets
> after the TP, as one commonly sought example.  The user perspective on this
> is extremely simple, and all-important.  It cannot be ignored just because
> some document writers profess not to understand the term "virtual world".
> (Everyone else does.)

the issue is not whether we "understand" the concept of a virtual
world, but whether it has meaning to people writing software. i know
you don't understand this since you're not a software developer, but
ultimately someone has to write code. and that code will make certain
assumptions about it's execution environment and which network peers
it trusts.

the assumption in this group was originally that we would have "domain
decomposition." we later changed that to "service decomposition." this
allows software developers to build software for "services" which
deployers can configure into "virtual experiences."

> The above kind of interop is either possible with VWRAP or it is not.  The
> phrase "no interop BETWEEN virtual worlds" denies the possibility point
> blank, and while it's nice to try to smooth it over as an artifact of
> terminology, we are faced with diametric opposites here.

no. the phrase "no interop BETWEEN virtual worlds" means two things:

a. we are not doing interop between two distinct virtual worlds that
use different protocols (the WoW vs. SL example.)
b. the protocol does not REQUIRE a participant to implement a complete
virtual world. a participant in the protocol may decide to implement
only a single service.

it has always meant this, despite the fact it's not what  you're interested in.

> While I agree with you entirely that there is (near unanimous) consensus
> that we are doing interop between virtual worlds (even if using different
> language), the drafts do not reflect that, and as Crista detailed, barely
> have any relevance to OpenSimulator-based worlds at all.

repeat after me: VWRAP is not HyperGrid. VWRAP is not OpenSim. VWRAP
is not Linden Lab. VWRAP is not Second Life.

> As many people have said, we need to fix the language of the drafts so that
> it reflects our interop goals clearly, because it fails to do that at the
> moment.

i have requested on several occasions that you provide a list of
issues you have with ANY of the drafts.

when you provide comments of the form: "the document says <this> and i
would prefer it says <that> because ..." then you will find that your
comments are addressed.