Thank you for clarifying your earlier terms. If I am understanding correctly, when you say "Second Life like" you mean that the environment produced lacks guarantees that the situation in one client is identical to the situation in another client.
Our goals ARE different as you have said. The goal of MXP is to enable a metaverse which is the product of many "participants" and which scales by having cooperative "hubs" which are tracking the overlap of objects' physical bounds with their awareness bounds. I believe your goal is to be able to support a true gaming experience - possessing guarantees of identical behaviour across clients. I think the relevant questions now are: - CAN there be one protocol which supports both of these models - SHOULD there be one protocol which supports both of these models (does it matter) Arkowitz On Oct 27, 8:51 am, Kripken <[email protected]> wrote: > You can look here for some specific things: > > http://groups.google.com/group/metaverseexchangeprotocol/browse_threa... > > There may also be an IRC log of my talks with the MXP devs, somewhere. > > Furthermore, consider e.g. the object fragment/movement event data in MXP > 0.5. That is suitable for an SL-type world, but not suitable for either > something like Quake, or a 'scripted movement' (movement is constrained to > particular places/paths) world. Such worlds simply need different > information in their object updates than what MXP provides. Also, this isn't > just different worlds: Different objects in the *same* world may need very > different update messages, and the ability to customize reliable vs. > unreliable, channels, and so forth. > > Bottom line, different protocols are needed. One approach, my personal > choice, is to allow each area to define its own protocol. So when you use > the client to connect to different areas, you will actually be running very > different protocols, to allow for very different environments (of course > this is done over a fixed 'core protocol' that lets you set up the final > protocol, but that core protocol is on the one hand extremely thin, much > more than e.g. MXP, but on the other hand provides features like channels, > etc.). In this way, a single client would be able to connect to both > Quake-type worlds, SL-type worlds, and 'scripted movement'-type worlds, as > well as others. > > - kripken > > > > On Tue, Oct 27, 2009 at 2:19 PM, arkowitz <[email protected]> wrote: > > > Yes, please explain what you mean by "SL-type". MXP was designed with > > the intention of supporting distributed physics, which danx0r found to > > be quite feasible, and which is not at all SL-like, seems to me; when > > I think of SL architecture the big thing that sticks out at me is that > > a monolithic "sim" controls everything within a specific "region"; and > > MXP does not imply this at all. Nor does MXP attempt to specify what > > types of content may be placed into the virtual space. > > > You need to separate the nature of a protocol from the way it may be > > used before you say it implies a certain type of architecture, and you > > certainly need to get specific about your vague claims. Please > > describe this other type of virtual world you are claiming MXP does > > not support. > > > Arkowitz > > > On Oct 27, 2:42 am, Kripken <[email protected]> wrote: > > > Even if the SL-specific assumptions you mention are lifted in MXP, you > > still > > > have what I called an "SL-type virtual world". You can call it something > > > else if you want (I guess the name bothers you). But my point is that, > > for > > > MXP to support the non-Second Life like aspects I mentioned (here and in > > the > > > MXP mailing list), it would need many changes or an entirely different > > > approach. > > > > In other words, what you call "moving beyond the SL assumptions", I see > > as > > > still being an SL-type world (just a better version of it). But again, > > > that's just terminology, the actual details are what matter. > > > > - kripken > > > > On Mon, Oct 26, 2009 at 11:44 PM, arkowitz <[email protected]> wrote: > > > > > "Second-Life-like"? NOOOOOO!!!!!!! If you want a "Second-Life-Like" > > > > virtual world, go with OpenSim... the architecture I contributed to > > > > MXP is not intended to replicate the Second Life set of assumptions. > > > > And I think if MXP is producing a Second Life experience, it is > > > > because it is being used in a certain way. > > > > > Can it be that I architected something specifically to move beyond the > > > > Second Life assumptions (square regions tied to specific hardware, > > > > avatars as special objects, scripting only in the server, and on and > > > > on) and then have it come full circle and be described as producing a > > > > "Second-Life-like" experience? There is a major disconnect here. > > > > > Arkowitz > > > > > On Oct 26, 9:07 am, Kripken <[email protected]> wrote: > > > > > On Mon, Oct 26, 2009 at 2:40 PM, arkowitz <[email protected]> > > wrote: > > > > > > > Kripken - just what do you mean by "for what it does"? > > > > > > MXP is a nice, clean, from-scratch protocol for a *Second Life*-like > > > > virtual > > > > > world. > > > > > > However, some people want other things. For example, if you want a > > more > > > > > fast-paced virtual world (that allows action games like Quake etc.), > > or > > > > at > > > > > the other extreme if you want a very slow-paced virtual world (only > > > > > semi-realtime, and maybe with limited places you can walk to - > > limited > > > > and > > > > > scripted), then you would need and prefer a different protocol than > > MXP. > > > > > > Btw, I have discussed this stuff with the MXP devs, in their mailing > > list > > > > > and on IRC, there might be logs somewhere (the mailing list, for > > sure). > > > > > > - kripken --~--~---------~--~----~------------~-------~--~----~ http://groups.google.com/group/realxtend http://www.realxtend.org -~----------~----~----~----~------~----~------~--~---
