<html>
<head />
<body>
one small addition from my side: it's not only about the motion of avatars, but it's additionally his interaction with the (virtual) reality to care about.<br /><br /> - sit down on/stand up from chairs<br /> - sit down on a chair lift<br /> - enter a vehicle<br /> - etc.<br />......<br /><blockquote type="cite">
<p>-------- Original-Nachricht --------<br />Datum: Fri, 4 Mar 2011 01:15:49 -0500<br />Von: "GLG" <info@3dnetproductions.com><br />An: "'Joe D Williams'" <joedwil@earthlink.net>, "'Sven-Erik Tiberg'" <Sven-Erik.Tiberg@ltu.se>, npolys@vt.edu, x3d-public@web3d.org<br />CC: luca.chittaro@uniud.it<br />Betreff: Re: [X3D-Public] remote motion invocation (RMI)<br /><br /></p>
<br /><br />ATM all of our avatars are from Avatar Studio 1 & 2. I<br />completely agree that they are simple and there is nothing<br />we want more than implementing H-Anim. But we need to<br />figure out the best way to do it first. That is why I<br />thought John was onto something in his May 2010<br />thread/questions. I think Joe you're the best person I<br />know that can answer this for us. Let me quote him back<br />here:<br /><br />"I agree, if I want my avatar to dance as an end user, I<br />shouldn't have to deal with transforms, points etc. I<br />should be able to download the dancing behavior I want and<br />apply it to my avatar. Let's make a market out of selling<br />motion. Cars are a wonderful example of this. Here's<br />another: http://playstation.joystiq.com/2009/06/19/quantic<br />-d<br />ream-selling-motion-capture-libraries/ Let's convert free<br />mocap library data to stuff that X3D and H-Anim can<br />use...see: http://mocap.cs.cmu.edu/ I'm such a newbie, I<br />don't even know what kind of mocap data H-Anim, X3D or<br />VRML takes! Can someone fill me in? I see 123 .bvh files<br />on the web. Can they be converted to h-anim? It looks<br />like you can go from .asf/.amc to .bvh to something<br />standard (from NIST) Here's all the asf and amc files on<br />the CMU site: http://mocap.cs.cmu.edu:8080/allasfamc.zip"<br /><br /><br /><br />>-----Original Message-----<br />>From: Joe D Williams [mailto:joedwil@earthlink.net]<br />>Sent: Thursday, March 03, 2011 11:04 PM<br />>To: info@3dnetproductions.com; 'Sven-Erik Tiberg';<br />>npolys@vt.edu; x3d-public@web3d.org<br />>Cc: luca.chittaro@uniud.it<br />>Subject: Re: [X3D-Public] remote motion invocation (RMI)<br />><br />><br />>I've got to start back at the beginning and ask how are<br />>you<br />>constructing you avatar?<br />>Is it just a skin or bones and skin? Do you animate by<br />>chaning the<br />>joints or just moving skin around?<br />><br />>> but having said the above, I hope<br />>> you'll see why I would be reluctant to implement the<br />>> previously suggested approach.<br />><br />>I haven't done much on this, just hearing some ideas, but<br />>many avators<br />>I have seen are just simple skins where the movement is<br />>made just by<br />>directly moving sets of vertices around. Of course this<br />>leads to<br />>relativelly simple avatars and low client processing<br />>requirements, but<br />>seems not sufficient to me.<br />><br />>Good Luck,<br />>Joe<br />><br />><br />>><br />>>>How much bandwidth is needed to send rotation and<br />>>>maybe<br />>>>position data to 20 joints? 20 SFVec3f points is Not<br />>that<br />>>>much.<br />>><br />>><br />>> It may not seem much data in itself for one avatar but,<br />>> even with efficient coding, you'd still have 20 times<br />>as<br />>> much data at the very least... No? Probably a lot more.<br />>> Concentrating only on gestures here, I'd rather have<br />>for<br />>> example 'action=run', or 'a=r' (3 bytes) in a stream,<br />>than<br />>> say 'translation=20 1 203&rotation=1 0 0 4.712', or<br />>'t=20<br />>> 1 203&r=1 0 0 4.712' (24 bytes X 20 = 480 bytes).<br />>That's a<br />>> factor of 16,000%. Considering that most decently<br />>featured<br />>> stand-alone MU servers can optimistically handle up to<br />>> just about 100 simultaneous avatars and/or shared<br />>objects,<br />>> I think that would translate in a huge waste of<br />>resources.<br />>> In other words, with that setup, you might only be able<br />>to<br />>> run say 3 to 5 avatars instead of 100. Those are<br />>> admittedly very gross calculations, but MU servers have<br />>to<br />>> be able to send data to clients in rapid-fire fashion,<br />>and<br />>> being able to keep as much of that data in low level<br />>> memory is crucial. It is difficult to know what exactly<br />>is<br />>> happening down there, but don't forget, other apps will<br />>be<br />>> running too, consuming much of the memory, perhaps<br />>forcing<br />>> more data from L1 cache to L2 and back (or L2 to L3,<br />>etc.<br />>> depending your setup, circumstances, running apps,<br />>etc.),<br />>> all of which contributing to slow down the entire<br />>process.<br />>> And we haven't yet really touched on the data<br />>processing<br />>> itself to support it. I believe the server should be as<br />>> bare and as close to the metal as possible. That's the<br />>> point I was making. The simpler we can keep it on the<br />>> server side, the more efficient it's going to be,<br />>> obviously. I don't think there is much disagreement,<br />>but<br />>> no matter how much parallel hardware you have, the<br />>memory<br />>> bandwidth issue never really goes away due to all the<br />>> inefficiencies that implies. The way we are planning to<br />>> solve this with X3Daemon at Office Towers, is to<br />>dedicate<br />>> servers to specific zones that are part of the same<br />>world.<br />>> So for example, if you go to this or that room, or<br />>> building, or area, then this or that server will take<br />>> over. Since it is unlikely that 100 avatars will be in<br />>the<br />>> same room/building/area all at once, that should work<br />>for<br />>> us. In other words, we conceive that, let say, 10<br />>servers<br />>> or more could in theory process and supply the MU data<br />>for<br />>> a single world as required, for say 1000 simultaneous<br />>> users or whatever the case may be. Like on the web, you<br />>> can seamlessly go from one server to the other, but, in<br />>> our case, while staying in the same world. We see here<br />>> real potential for scalability. Instead of piling data<br />>in<br />>> parallel hardware (which we did consider but would<br />>rather<br />>> go with symmetric multiprocessing) however connected,<br />>each<br />>> machine would run independently of each other. Trying<br />>not<br />>> to get into details, but having said the above, I hope<br />>> you'll see why I would be reluctant to implement the<br />>> previously suggested approach.<br />>><br />>> Cheers,<br />>> Lauren<br />>><br />>><br />>><br />>>>-----Original Message-----<br />>>>From: Joe D Williams [mailto:joedwil@earthlink.net]<br />>>>Sent: Thursday, March 03, 2011 1:10 PM<br />>>>To: info@3dnetproductions.com; 'Sven-Erik Tiberg';<br />>>>npolys@vt.edu; x3d-public@web3d.org<br />>>>Cc: luca.chittaro@uniud.it<br />>>>Subject: Re: [X3D-Public] remote motion invocation<br />>(RMI)<br />>>><br />>>><br />>>>Hi Lauren,<br />>>>I was thinking about it in terms of essentially<br />>>>substituting<br />>>>'external' 'streams' of data instead of using<br />>'internal'<br />>>>data<br />>>>generated by scripts, timers, and interpolators. The<br />>>>point is, when<br />>>>the joint receives a joint translation or rotation,<br />>then<br />>>>it is up to<br />>>>the client player to move the child segments and joints<br />>>>and deform the<br />>>>skin to achieve the movement.. So, the data needed to<br />>>>animate the<br />>>>humanoid is really not that much; the client does all<br />>the<br />>>>actual<br />>>>moving. How much bandwidth is needed to send rotation<br />>and<br />>>>maybe<br />>>>position data to 20 joints? 20 SFVec3f points is Not<br />>that<br />>>>much.<br />>>>However, still probably more reliable and maybe<br />>efficient<br />>>>and even<br />>>>faster to just use 'internal' optimized timers and<br />>>>interpolators. The<br />>>>secret of course seems to be getting as much of that<br />>>>computation as<br />>>>possible done in the parallel hardware.<br />>>><br />>>>Note that I am not even considering the idea of just<br />>>>sending new<br />>>>fields of vertices to replace the ones currently in<br />>>>memory. I think<br />>>>this is essentially done in fixed frame per second<br />>>>applications where<br />>>>the backroom app does all the computation to achieve<br />>the<br />>>>mesh for the<br />>>>next timed frame, then just 'streams' the mesh to the<br />>>>renderer where<br />>>>the picture is made.<br />>>>Thanks and Best Regards,<br />>>>Joe<br />>>><br />>>><br />>>><br />>>><br />>>>>On the other hand, I could see a setup where live<br />>joint<br />>>>>rotation and<br />>>>>displacement data was streamed into the scene and<br />>routed<br />>>>>to joints to<br />>>>>affect the humanoid in real time, rather than using<br />>>>>client-side<br />>>>>interpolators and combiners.<br />>>><br />>>><br />>>>Thanks for your reply. Yes, that is better; 'streaming'<br />>>>data does seem closer to something conceivable, even<br />>>>interesting. However, since talking about humanoids<br />>>>likely<br />>>>implies multi-users in most applications, I'd start<br />>>>getting worried about processors' cache/memory<br />>bandwidth;<br />>>>I don't think we are there yet to support a great many<br />>>>users this way. Even with high clock speed, multi-core<br />>>>processors and of course plenty of network bandwidth,<br />>low<br />>>>level memory management will remain an issue as the<br />>>>number<br />>>>of users increases, IMO. It already is with much<br />>simpler<br />>>>systems. That is why I like to have as much done on the<br />>>>client side as possible. Other than helping lowly<br />>>>devices,<br />>>>I'd be curious about the main advantages of the above<br />>>>approach? Open to ideas but, as far as mobile devices,<br />>>>hopefully Moore's law will continue to hold true for<br />>some<br />>>>time...<br />>>><br />>>>Cheers,<br />>>>Lauren<br />>>><br />>>><br />>>> _____ ___ __ __ ___ ______ ___ ___<br />>___<br />>>> /__ / / _ \ / | / // _ \/_ __/ / _ \ / _ \ / _<br />>\<br />>>> _/_ / / // / / /||/ // __/ / / / ___// _// //<br />>/<br />>>>/____/ /____/ /_/ |__/ \___/ /_/ /_/ /_/\_\<br />>\___/<br />>>>______________________________________________________<br />>>> * * Interactive Multimedia - Internet Management * *<br />>>> * * Virtual Reality - Application Programming * *<br />>>> * 3D Net Productions www.3dnetproductions.com *<br />>>><br />>>><br />>>>>-----Original Message-----<br />>>>>From: Joe D Williams [mailto:joedwil@earthlink.net]<br />>>>>Sent: Tuesday, March 01, 2011 7:49 PM<br />>>>>To: info@3dnetproductions.com; 'Sven-Erik Tiberg';<br />>>>>npolys@vt.edu; x3d-public@web3d.org<br />>>>>Cc: luca.chittaro@uniud.it<br />>>>>Subject: Re: [X3D-Public] remote motion invocation<br />>(RMI)<br />>>>><br />>>>><br />>>>>><br />>>>>> Hello Joe,<br />>>>>><br />>>>>> While we are on the subject, I have asked you this<br />>>>>before<br />>>>>> but I think you missed it. There is a link at<br />>>>>><br />>>>>> http://www.web3d.org/x3d/workgroups/h-anim/<br />>>>>><br />>>>>> for ISO-IEC-19775 X3D Component 26 Humanoid<br />>Animation<br />>>>>> (H-Anim) which describes H-Anim interfaces in terms<br />>of<br />>>>>the<br />>>>>> X3D scenegraph...<br />>>>>><br />>>>>> That link is broken. Where else can I obtain this?<br />>>>><br />>>>>goto:<br />>>>><br />>>>>http://www.web3d.org/x3d/specifications/ISO-IEC-19775-<br />>>>>1.2-X3D-<br />>>>>AbstractSpecification/Part01/components/hanim.html<br />>>>><br />>>>>the latest spec changed base and we lost the link.<br />>>>>I will tell the webmaster<br />>>>><br />>>>>> If I am<br />>>>>> asking the wrong person then, can someone else help?<br />>>>><br />>>>>I think the most current spec should have the same<br />>link<br />>>>>forever.<br />>>>><br />>>>>><br />>>>>> Apologies for hijacking this thread, maybe I am not<br />>>>>> getting this right, but, I have difficulties getting<br />>>>>> around my head how server side rasterization can<br />>work<br />>>>>with<br />>>>>> X3D other than with the MovieTexture node.<br />>>>><br />>>>>Well, i think server side rasterization could work<br />>when<br />>>>>you have a<br />>>>>simple mesh avatar and all you want to do is just show<br />>>>>apparent<br />>>>>vertices move so you send sort of a set of gifs on a<br />>>>>billboard down to<br />>>>>the client. Don't get it? me either:).<br />>>>>On the other hand, I could see a setup where live<br />>joint<br />>>>>rotation and<br />>>>>displacement data was streamed into the scene and<br />>routed<br />>>>>to joints to<br />>>>>affect the humanoid in real time, rather than using<br />>>>>client-side<br />>>>>interpolators and combiners. I think even mobile will<br />>>>>always have<br />>>>>power to do reasonable h-anim skin and bone characters<br />>>>>but low end<br />>>>>performance devices may need a more simple character<br />>>>with<br />>>>>fewer joints<br />>>>>and more open mesh and different textures.<br />>>>><br />>>>><br />>>>>><br />>>>>> Thanks,<br />>>>>> Lauren<br />>>>>><br />>>>><br />>>>>Thanks and Best Regards,<br />>>>>Joe<br />>>>><br />>>>>><br />>>>>>>-----Original Message-----<br />>>>>>>From: x3d-public-bounces@web3d.org [mailto:x3d-<br />>public-<br />>>>>>>bounces@web3d.org] On Behalf Of Joe D Williams<br />>>>>>>Sent: Tuesday, March 01, 2011 12:09 PM<br />>>>>>>To: Sven-Erik Tiberg; npolys@vt.edu; x3d-<br />>>>>public@web3d.org<br />>>>>>>Cc: luca.chittaro@uniud.it<br />>>>>>>Subject: Re: [X3D-Public] remote motion invocation<br />>>>>(RMI)<br />>>>>>><br />>>>>>><br />>>>>>>----- Original Message -----<br />>>>>>>From: "Sven-Erik Tiberg" <Sven-Erik.Tiberg@ltu.se><br />>>>>>>To: <npolys@vt.edu>; <x3d-public@web3d.org><br />>>>>>>Cc: <luca.chittaro@uniud.it><br />>>>>>>Sent: Tuesday, March 01, 2011 6:14 AM<br />>>>>>>Subject: Re: [X3D-Public] remote motion invocation<br />>>>>(RMI)<br />>>>>>><br />>>>>>><br />>>>>>>> Hi<br />>>>>>>><br />>>>>>>> A note.<br />>>>>>>> Would it be possible to create a object similar to<br />>>>>>>dynamic h-anim,<br />>>>>>>> looking and behaving like a moose.<br />>>>>>><br />>>>>>>THe basis for h-anim comes from general-purpose<br />>>>>>>hierarchical animation<br />>>>>>>needs. That is, the shoulder joint is connected to<br />>the<br />>>>>>>elbow joint,<br />>>>>>>for example, such that when the shoulder joint is<br />>>>>moved,<br />>>>>>>then the<br />>>>>>>elbow joint is translated accordingly as expected.<br />>To<br />>>>>>>follow the same<br />>>>>>>technique for a different character, just position<br />>the<br />>>>>>>joints and<br />>>>>>>segments you chose to use depending upon the level<br />>of<br />>>>>>>articulation you<br />>>>>>>need in skeleton space then build the skin in skin<br />>>>>space<br />>>>>>>(hopefully<br />>>>>>>the same as skeleton space), connect it up and<br />>animate<br />>>>>>>the thing.<br />>>>>>><br />>>>>>>><br />>>>>>>> No I'm not kidding, in the future we hope to<br />>create<br />>>>a<br />>>>>>>X3D-sceen for<br />>>>>>>> driver simulator with "living" moose and reindeers<br />>(<br />>>>>>>they are not so<br />>>>>>>> exotic up here but more real and a hazard on the<br />>>>>roads<br />>>>>>>).<br />>>>>>>> Could be that the same motion pattern can be used<br />>>>for<br />>>>>>>deers and elks<br />>>>>>>> and ..<br />>>>>>><br />>>>>>>possibly. Note that the selection of initial binding<br />>>>>time<br />>>>>>>joint<br />>>>>>>rotation is important.<br />>>>>>><br />>>>>>>Good Luck and Best Regards,<br />>>>>>>Joe<br />>>>>>><br />>>>>>><br />>>>>>>><br />>>>>>>> /Sven-Erik Tiberg<br />>>>>>>><br />>>>>>>><br />>>>>>>> -----Original Message-----<br />>>>>>>> From: x3d-public-bounces@web3d.org<br />>>>>>>> [mailto:x3d-public-bounces@web3d.org] On Behalf Of<br />>>>>>>Sven-Erik Tiberg<br />>>>>>>> Sent: den 1 mars 2011 08:51<br />>>>>>>> To: npolys@vt.edu; x3d-public@web3d.org<br />>>>>>>> Subject: Re: [X3D-Public] remote motion invocation<br />>>>>>>(RMI)<br />>>>>>>><br />>>>>>>><br />>>>>>>> If there are a need for extending the h-anim<br />>>>>kinematic<br />>>>>>>capacity I<br />>>>>>>> would have a look at Modelica<br />>>>>https://www.modelica.org/<br />>>>>>>And (<br />>>>>>>> openmodelica ) https://www.modelica.org/ for<br />>>>>>>calculation of dynamic<br />>>>>>>> behavior.<br />>>>>>>><br />>>>>>>> IMHO it would not be impossible to interface an<br />>>>>>>openmodelica model<br />>>>>>>> that runs in Real Time to a X3D sceen.<br />>>>>>>><br />>>>>>>> On animiation of human I would like to suggest<br />>that<br />>>>>you<br />>>>>>>take a look<br />>>>>>>> at http://hcilab.uniud.it/index.html and it's<br />>>>editor<br />>>>>>>H-Animator<br />>>>>>>> http://hcilab.uniud.it/demos-videos/item11.html<br />>>>>>>><br />>>>>>>> /Sven-Erik Tiberg<br />>>>>>>> Lulea Univ of Technology<br />>>>>>>> Sweden<br />>>>>>>><br />>>>>>>><br />>>>>>>><br />>>>>>>> -----Original Message-----<br />>>>>>>> From: x3d-public-bounces@web3d.org<br />>>>>>>> [mailto:x3d-public-bounces@web3d.org] On Behalf Of<br />>>>>>>Nicholas F. Polys<br />>>>>>>> Sent: den 28 februari 2011 21:34<br />>>>>>>> To: x3d-public@web3d.org<br />>>>>>>> Subject: Re: [X3D-Public] remote motion invocation<br />>>>>>>(RMI)<br />>>>>>>><br />>>>>>>> One of our members was working on this spec some<br />>>>time<br />>>>>>>ago (2001),<br />>>>>>>> perhaps a good integration target with H-Anim for<br />>>>>such<br />>>>>>>a platform or<br />>>>>>>> communication layer: HumanML<br />>>>>>>> http://www.oasis-<br />>>>>>>open.org/committees/tc_home.php?wg_abbrev=humanmarku<br />>p<br />>>>>>>><br />>>>>>>> it does mention kinesthetics in the overview but I<br />>>>am<br />>>>>>>really not<br />>>>>>>> sure where it has been adopted.<br />>>>>>>> ?<br />>>>>>>><br />>>>>>>> br,<br />>>>>>>><br />>>>>>>> _n_polys<br />>>>>>>><br />>>>>>>> On 2/28/2011 1:54 PM, Joe D Williams wrote:<br />>>>>>>>>> And what if the script could recive states (<br />>>>ROUTED<br />>>>>>>events ) from<br />>>>>>>>>> the<br />>>>>>>>>> 3D browser and do a animation on the fly using<br />>>>>>>inverse kinematics,<br />>>>>>>>>> that would be cool.<br />>>>>>>>><br />>>>>>>>> I think the basic needed structure for that is in<br />>>>>the<br />>>>>>>X3D H-Anim<br />>>>>>>>> spec.<br />>>>>>>>><br />>>>>>>>> Best Regards,<br />>>>>>>>> Joe<br />>>>>>>>><br />>>>>>>>> ----- Original Message ----- From: "Sven-Erik<br />>>>>Tiberg"<br />>>>>>>>> <Sven-Erik.Tiberg@ltu.se><br />>>>>>>>> To: "John Carlson" <john.carlson3@sbcglobal.net>;<br />>>>>>>>> <x3d-public@web3d.org><br />>>>>>>>> Sent: Sunday, February 27, 2011 11:50 PM<br />>>>>>>>> Subject: Re: [X3D-Public] remote motion<br />>invocation<br />>>>>>>(RMI)<br />>>>>>>>><br />>>>>>>>><br />>>>>>>>>> What if using an from the browser external<br />>motion<br />>>>>>>engine as you<br />>>>>>>>>> mention, but let the motion engine host in the<br />>>>>client<br />>>>>>>computer.<br />>>>>>>>>> Communicating new states to the 3D-browser (to a<br />>>>>h-<br />>>>>>>amin model )<br />>>>>>>>>> trough a script.<br />>>>>>>>>> And what if the script could recive states (<br />>>>ROUTED<br />>>>>>>events ) from<br />>>>>>>>>> the<br />>>>>>>>>> 3D browser and do a animation on the fly using<br />>>>>>>inverse kinematics,<br />>>>>>>>>> that would be cool.<br />>>>>>>>>><br />>>>>>>>>> Thanks for bringing up this subject.<br />>>>>>>>>><br />>>>>>>>>> /Sven-Erik Tiberg<br />>>>>>>>>><br />>>>>>>>>><br />>>>>>>>>> -----Original Message-----<br />>>>>>>>>> From: x3d-public-bounces@web3d.org<br />>>>>>>>>> [mailto:x3d-public-bounces@web3d.org] On Behalf<br />>Of<br />>>>>>>John Carlson<br />>>>>>>>>> Sent: den 26 februari 2011 05:49<br />>>>>>>>>> To: <x3d-public@web3d.org> mailing list<br />>>>>>>>>> Subject: [X3D-Public] remote motion invocation<br />>>>>(RMI)<br />>>>>>>>>><br />>>>>>>>>> Instead of downloading motion into a 3D scene,<br />>how<br />>>>>>>about uploading<br />>>>>>>>>> motion into a remote social network server or<br />>>>>robot?<br />>>>>>>One could<br />>>>>>>>>> upload motions to a remote server, and then<br />>invoke<br />>>>>>>them with<br />>>>>>>>>> parameters. One could create classes<br />>aggregating<br />>>>>>>motion methods,<br />>>>>>>>>> instantiate classes to create behaviors in<br />>avatars<br />>>>>or<br />>>>>>>robots.<br />>>>>>>>>><br />>>>>>>>>> What if H-Anim was used to upload and control<br />>>>>avatars<br />>>>>>>remotely,<br />>>>>>>>>> instead of downloaded models? What I am<br />>thinking<br />>>>>of<br />>>>>>>is something<br />>>>>>>>>> like Kinect/PrimeSense for primitive motion<br />>input<br />>>>+<br />>>>>a<br />>>>>>>programming<br />>>>>>>>>> language for<br />>>>>>>repetition/parameterization/classes/instantation.<br />>>>>>>>>> What<br />>>>>>>>>> if I chose a protocol/programming such as<br />>>>>JavaScript?<br />>>>>>>>>><br />>>>>>>>>> Does DIS do this? Are there any patents to be<br />>>>>aware<br />>>>>>>of? If we<br />>>>>>>>>> use<br />>>>>>>>>> something like Caja on the server, or sandbox<br />>the<br />>>>>>>motion on the<br />>>>>>>>>> server, could we insure the security of such an<br />>>>>>>endeavor?<br />>>>>>>>>><br />>>>>>>>>> Initially, I am thinking only of video download.<br />>>>>>>>>><br />>>>>>>>>> Thanks,<br />>>>>>>>>><br />>>>>>>>>> John Carlson<br />>>>>>>>>> _______________________________________________<br />>>>>>>>>> X3D-Public mailing list<br />>>>>>>>>> X3D-Public@web3d.org<br />>>>>>>>>> http://web3d.org/mailman/listinfo/x3d-<br />>>>>>>public_web3d.org<br />>>>>>>>>><br />>>>>>>>>> _______________________________________________<br />>>>>>>>>> X3D-Public mailing list<br />>>>>>>>>> X3D-Public@web3d.org<br />>>>>>>>>> http://web3d.org/mailman/listinfo/x3d-<br />>>>>>>public_web3d.org<br />>>>>>>>><br />>>>>>>>><br />>>>>>>>> _______________________________________________<br />>>>>>>>> X3D-Public mailing list<br />>>>>>>>> X3D-Public@web3d.org<br />>>>>>>>> http://web3d.org/mailman/listinfo/x3d-<br />>>>>public_web3d.org<br />>>>>>>><br />>>>>>>> --<br />>>>>>>> Nicholas F. Polys Ph.D.<br />>>>>>>><br />>>>>>>> Director of Visual Computing<br />>>>>>>> Virginia Tech Information Technology<br />>>>>>>><br />>>>>>>> Affiliate Research Professor<br />>>>>>>> Virginia Tech Computer Science<br />>>>>>>><br />>>>>>>><br />>>>>>>> _______________________________________________<br />>>>>>>> X3D-Public mailing list<br />>>>>>>> X3D-Public@web3d.org<br />>>>>>>> http://web3d.org/mailman/listinfo/x3d-<br />>>>>public_web3d.org<br />>>>>>>><br />>>>>>>> _______________________________________________<br />>>>>>>> X3D-Public mailing list<br />>>>>>>> X3D-Public@web3d.org<br />>>>>>>> http://web3d.org/mailman/listinfo/x3d-<br />>>>>public_web3d.org<br />>>>>>>><br />>>>>>>> _______________________________________________<br />>>>>>>> X3D-Public mailing list<br />>>>>>>> X3D-Public@web3d.org<br />>>>>>>> http://web3d.org/mailman/listinfo/x3d-<br />>>>>public_web3d.org<br />>>>>>><br />>>>>>><br />>>>>>>_______________________________________________<br />>>>>>>X3D-Public mailing list<br />>>>>>>X3D-Public@web3d.org<br />>>>>>>http://web3d.org/mailman/listinfo/x3d-<br />>public_web3d.org<br />>>>>><br />>>><br />>><br />>><br /><br /><br /><br />_______________________________________________<br />X3D-Public mailing list<br />X3D-Public@web3d.org<br />http://web3d.org/mailman/listinfo/x3d-public_web3d.org<br /></blockquote>
<div class="signature"><br /><br /><br />-- <br />Schon gehört? GMX hat einen genialen Phishing-Filter in die<br />Toolbar eingebaut! http://www.gmx.net/de/go/toolbar</div></body>
</html>