Re: [ogpx] VWRAP Draft Charter: 2009 08 28 revision

[Meadhbh Siobhan <[e-mail]>]

hey kari. thanks' for chiming in.

On Sun, Aug 30, 2009 at 5:05 AM, Kari Lippert<[e-mail]> wrote:
> My two cents....
>
> The first paragraph (where the purpose is being laid out) says:
>
> Conforming  client  applications use  the protocol  to manipulate and
> move the  user's avatar, create  objects in  a virtual world, interact
>  with other users  and their surroundings  and consume and create
> media and information from sources inside and outside their virtual
> world.
>
> I normally lurk in this group but I have to say this surprised me.
> This statement says that I will use this protocol within my virtual
> world, not that I will use this protocol to interface with different
> virtual worlds. This doesn't speak to what I thought the thrust was -
> interoperability of worlds for transportability of avatars. While once
> could argue that the use of the same protocol intra-world would help
> in the inter-world communication, this is not the case and should not
> be assumed to be so. I'm actually saddened that you're all thinking of
> interoperability as achievable only if everyone uses the same
> intra-world protocol.

we do not define the term "virtual world." there is considerable
disagreement as to what the term means. in the previous effort (MMOX)
we attempted to start talking about the term and discovered there was
considerable disagreement as to what it meant. part of the reason the
OGPX BoF was formed was to illuminate a smaller section of the problem
domain. in the part of the problem domain we are choosing to work on,
we do not discuss interoperability between "virtual worlds" as this
term is not used normatively. instead, we define the terms "region
domain" and "simulator" and "agent domain" and a few others and define
interoperability expectations and protocol between _them_.

morgaine uses the term "virtual world" to mean what we call a "region
domain." so if this is your definition, then we are talking about an
inter-virtual world protocol.

we are shying away from using it in this effort because it's use was
so contentious in previous standardization efforts (VWIF, AWG, MMOX.)

interoperability of avatar formats is one aspect of this protocol, but
not the only one.

it should also be noted that we are not attempting to say that our
model is the _ONLY_ valid models for virtual worlds. this is one of
the reasons that OGPX is a distinct effort from MMOX. MMOX retains the
wider focus, and as i understand it, there's no reason someone else
can't continue it's work, or define a separate part of the problem
domain and form another working group.

>
> "Enforcing" the use of a standard intra-world for every world will be
> impossible and quite possibly viewed as some by an intrusion into
> their IP, not to mention that it would kill innovation. Transfer
> between worlds will be lossy: existing worlds are not the same, nor do
> they have the same types of virtual property associated with them.
> This standard is going to either have to address the bare minimum, or
> become obsolete prior to completion with the introduction of a novel
> virtual world into the virtual universe.

unlike MMOX, we are not attempting, in this group, to define a generic
protocol to be used by every virtual world. OGPX is intended to
provide interoperability, not between worlds, but between hosts that
work together to simulate a virtual world.

nor are we trying to mandate a virtual world protocol that everyone
MUST implement. in the same way that POP3 did nothing to supplant the
use of Microsoft's proprietary Exchange protocol, or from the effort
to define IMAP, we think that the OGPX/VWRAP effort will cover one
part of the problem domain.

>
> I concur with many of the discussions that have been put forth
> regarding this draft and have the following suggestions:
>
> Infinity Linden - good rewording to include OGP history; too confusing
> to leave in the main body (a reader will think it a typo)
>
> Morgaine - agree to remove sentence that begins "To support the
> exegesis  of the specifications..."
>
> meadhbh - deployment patterns are very different from models of
> protocols; deployment patterns may be useful but should not replace a
> good model

are we talking about a processing model? we really want to define a
processing model for conforming implementations?

>
> Morgaine - agree with the addition of the Foundation Component but
> argue that it should also be the focus in the first paragraph of the
> description
>
> meadhbh/Morgaine/Carlo - redefinition of virtual world is a bad idea
> but what you are really talking about is the virtual universe as
> composed by a variety of virtual worlds - and I like the plain English
> of Region 1 in VW 1 to Region 2 in VW 2 expression put forth as it is
> very clear, easily understood, not easily misunderstood,  and I
> believe captures the intent of the standard to be developed

again. this is one of the reasons we're trying to focus on terms we
have agreement on.

>
> Once we figure out the focus (which I take to be inter-world, or
> cross-world, transfer) then we can begin capturing the requirements
> for a "successful transfer". From that the model, the deployment
> pattern(s), the conformance guidelines, and other associated parts of
> the standard will come. So I ask, what is the focus (purpose) of the
> working group? Why are you making all this effort? Is it to bring the
> multitudes of virtual worlds to one protocol or to make transfer
> between them possible? I posit those are two different standards and
> only the latter is worth standardization.

we've actually been working on a focus for the past six months. first
we were broad with MMOX, but couldn't get agreement. then with OGPX we
stated a focus on what's come to be called "second life-like" worlds.
the focus of the group is to define an interoperability profile and
protocol, deployment patterns and processing expectations for these
worlds.

>
> <step off soapbox>
> My two cents anyway....
>
> Kari
> _______________________________________________
> ogpx mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ogpx
>