<html><head></head><body><div style="font-family: Verdana;font-size: 12.0px;"><div>Hi all,<br/>
<br/>
This is just FYI. It describes a few of the major design ideas behind the SrrTrains v0.01 project......<br/>
<br/>
and some interesting statements from John Richardson about "Network Sensor vs. HLA/DIS"</div>
<div> </div>
<div>I hope it is OK, when I make these statements public.<br/>
<br/>
All the best<br/>
<br/>
Christoph<br/>
<br/>
Gesendet: Samstag, 22. September 2018 um 01:32 Uhr<br/>
Von: "Valentin Christoph" <<a href="http://Christoph.Valentin" target="_blank">Christoph.Valentin</a>@kapsch.net><br/>
An: "Christoph Valentin" <<a href="http://christoph.valentin" target="_blank">christoph.valentin</a>@gmx.at>, "John Richardson" <richards@spawar.navy.mil>, "John R" <**********><br/>
Cc: "Christian Schoebel" <scoe@brg19.at>, "Onic Roman" <<a href="http://Roman.Onic" target="_blank">Roman.Onic</a>@kapsch.net><br/>
Betreff: AW: SrrTrains#3<br/>
<br/>
Maybe it's the fascination of inventing the wheel a gain and again. But ON YOUR OWN :-)))))<br/>
<br/>
<br/>
<br/>
<br/>
<br/>
<br/>
<br/>
Von meinem Samsung Gerät gesendet.<br/>
<br/>
<br/>
<br/>
-------- Ursprüngliche Nachricht --------<br/>
Von: Christoph Valentin <<a href="http://christoph.valentin" target="_blank">christoph.valentin</a>@gmx.at><br/>
Datum: 21.09.18 18:59 (GMT+01:00)<br/>
An: John Richardson <richards@spawar.navy.mil>, John R <**********><br/>
Cc: Valentin Christoph <<a href="http://Christoph.Valentin" target="_blank">Christoph.Valentin</a>@kapsch.net>, Christian Schoebel <scoe@brg19.at>, Onic Roman <<a href="http://Roman.Onic" target="_blank">Roman.Onic</a>@kapsch.net><br/>
Betreff: RE: SrrTrains#3<br/>
<br/>
<br/>
John,<br/>
<br/>
<br/>
<br/>
Thanks again for the answers.<br/>
<br/>
<br/>
<br/>
I'll wait for your answer about Mixed Reality, before I'll probably send SrrTrains#4.<br/>
<br/>
<br/>
<br/>
And yes, I understand the advantages of HLA/DIS over the concept of the Network Sensor, although I'm just having a rough idea about what they are.<br/>
<br/>
<br/>
<br/>
1. DIS is a communication protocol defined by the IEEE, where the PDU structure is INTERNATIONALLY defined down to the bits. This is great.<br/>
<br/>
2. HLA seems to be some object oriented description about how the content of a scene must be structured in order to efficiently use DIS (true?)<br/>
<br/>
<br/>
<br/>
Call me a stubborn old donkey, or call me anything, but I will stick to my approach of using the Network Sensor and BS Contact/BS Collaborate, until the alpha release of SrrTrains v0.01 will be finished ("not before the year 2020").<br/>
<br/>
<br/>
<br/>
Our catholic teacher said: "Omnis sanctus pertinax" and I will stick to this advise :-) :-)<br/>
<br/>
<br/>
<br/>
And I must think about Luke 5:37 - 5:39<br/>
<br/>
37And no one pours new wine into old wineskins. If he does, the new wine will burst the skins, the wine will spill, and the wineskins will be ruined. 38Instead, new wine is poured into new wineskins. 39And no one after drinking old wine wants new, for he says, ‘The old is better.’”<br/>
<br/>
<br/>
Maybe you can understand my obsession about the Network Sensor, can't you?<br/>
<br/>
<br/>
<br/>
Everything would go a different path, if I could work with multiuser X3D/VRML during office hours, but that's nothing I can influence :-) :-)<br/>
<br/>
<br/>
<br/>
Have a great weekend, too.<br/>
<br/>
<br/>
<br/>
Christoph<br/>
<br/>
<br/>
<br/>
Gesendet: Freitag, 21. September 2018 um 18:00 Uhr<br/>
Von: "John Richardson" <richards@spawar.navy.mil><br/>
An: "'Christoph Valentin'" <<a href="http://christoph.valentin" target="_blank">christoph.valentin</a>@gmx.at><br/>
Betreff: RE: SrrTrains#3<br/>
<br/>
Oops,<br/>
<br/>
Forgot you...:-(<br/>
<br/>
In 1994 I was at the Pre VRML conference related to VRML 1.0 [VRML 97 is VRML 2.0].<br/>
<br/>
One of the main speakers mentioned that the shortcoming of VRML 1.0 was that the most interesting goal of VRML was a world with more than one avatar, which VRML 1.0 did not meet.<br/>
<br/>
So, yes multiuser is important and that was why the VRML 97 spec had distributed nodes.<br/>
<br/>
I'll see if I can send the DIS and HLA documents. On your own, try the SISO standards website to locate all such documents if you want them 10 minutes after you get this reply..:-)<br/>
<br/>
I believe that it is <a href="http://www.sisostds.org" target="_blank">www.sisostds.org</a>. You have to troll the site for the working groups. The working groups are probably the maintainers of the DIS and HLA standards. You can also volunteer to help review standard changes...:-)<br/>
<br/>
Railroads are not 1 dimensional. All railroads are on an oblate sphere. It is called the earth. But really it is just the WGS-84 map projection. Now DIS/HLA is used by governments and is acceptable. <strong>But the Network sensor is also acceptable</strong>. The tracks have an elevation at any point. The elevation and azimuth change every unit of measurement. Roads can even have a third rotation [they bank to allow for cars to have more traction]. So you will need all the utilities from OSGeo [Open Geospatial organization] to translate from X,Y,Z to Latitude Longitude and height above surface. But I assume that you can get all that information since the organizations and governments that build the tracks have that information.<br/>
<br/>
Note: Everything that is in DIS can be described in a PROTO. DIS is a protocol but mostly it is a non object oriented predecessor to HLA. Since it is of interest to governments, a lot of the enumerations are related to CHARACTERISTICS of aircraft, weapons, radars, comms [plus lots of other bureaucratic stuff] and when focused on logistics are focused on characteristics of trucks and tankers. Probably more. Also DIS may have a railway enumeration but it is probably useless. What you need is the characteristics of every railway car and locomotive that you will be simulation described in a PROTO. With get and set methods. Then you will need a PROTO for propulsion and some algorithms to simulate railroad propulsion. Plus a Signal type of PROTO for track switching and scheduling control so you don't crash. Probably others but these are minimum.<br/>
<br/>
I probably don't have time today to address mixed reality since I am leaving for a machine learning lecture at SDSU soon.<br/>
<br/>
Have a great weekend.<br/>
<br/>
John<br/>
<br/>
-----Original Message-----<br/>
From: Christoph Valentin [mailto:<a href="http://christoph.valentin" target="_blank">christoph.valentin</a>@gmx.at]<br/>
Sent: Friday, September 21, 2018 2:07 PM<br/>
To: John R <**********>; John Richardson <richards@spawar.navy.mil><br/>
Subject: Fw: SrrTrains#3<br/>
<br/>
John,<br/>
<br/>
May I ask, whether you have not had a possibility yet to answer or whether you will not answer to the e-mail below?<br/>
<br/>
I'll wait with SrrTrains#4, until this question is solved.<br/>
<br/>
Thanks,<br/>
Christoph<br/>
<br/>
Gesendet: Freitag, 14. September 2018 um 19:45 Uhr<br/>
Von: "Christoph Valentin" <<a href="http://christoph.valentin" target="_blank">christoph.valentin</a>@gmx.at><br/>
An: "John Richardson" <richards@spawar.navy.mil><br/>
Cc: "John R" <**********>, "Roman Onic" <<a href="http://roman.onic" target="_blank">roman.onic</a>@kapsch.net>, "Christian Schoebel" <scoe@brg19.at>, <a href="http://christoph.valentin" target="_blank">christoph.valentin</a>@kapsch.net<br/>
Betreff: SrrTrains#3<br/>
<br/>
John,<br/>
<br/>
<br/>
<br/>
I am daring to write this e-mail as a kind of RFC (request for comments), asking you kindly to let me know any thoughts that come to your mind, when you read these lines.<br/>
<br/>
<br/>
<br/>
Summary:<br/>
<br/>
In the present e-mail "SrrTrains#3" I am trying to give an explanation, why the multi user mode of SrrTrains was always important to me. I explain, why I decided to stick to the X3D/VRML Network Sensor node (in contrary to the X3D DIS component) in early 2009. Then I explain, why I think the multi user mode is a necessary pre-condition for mixed reality mode.<br/>
<br/>
<br/>
<br/>
Why is multiuser important?<br/>
<br/>
When I started to write concept papers for the SIMUL-RR project (must have been around 2005 - 2007), then I immediately required some multiuser mode for the virtual railroad. I remembered the days, when my brothers and I had been children and we have had space enough for a *real* model railroad. Model railroading had always been an argument, why Daddy took some time and played with us. Therefore railroading was always something *positive* in my feelings. Playing *together* with the brothers and the father was the major feature of model railroading.<br/>
<br/>
Nowadays, I recognize the multiplayer mode of many games is an important feature for my son (who is 18 years old now), when he plays games over the Internet together with his friends (I am happy that he still goes out, to the movies, to the pub, and so on, not always sitting around in the front of the screen :-) ).<br/>
<br/>
<br/>
<br/>
Why Network Sensor?<br/>
<br/>
When I actually started the SrrTrains project in spring 2009, I had already decided to use the X3D standard. I was still looking for an eventual application by my employer (who was Siemens from 1997 until 2010) in the TELECOM/RAILWAY industry. Therefore I asked at the X3D-public mailing list, whether there was a standardized protocol for the multiuser mode.<br/>
<br/>
They told me, I could use the DIS component or I could use the Network Sensor.<br/>
<br/>
I had a look to the description of the X3D DIS component and found it was quite specific. Each entity had one and only one three-dimensional position and an orientation (which I thought would not fit to railway simulations, because the position is described one-dimensionally "along the tracks").<br/>
<br/>
The Network Sensor appeared to be much more general than the DIS component was. Therefore I thought, I could implement a greater variety of different simulated entities, when I used the Network Sensor.<br/>
<br/>
By the way: I still haven't got a copy of the IEEE DIS standard, nor of any specs about HLA. Just several good slides and links from Don Brutzman. I wouldn't have the time to study them, anyway.<br/>
<br/>
<br/>
<br/>
Mixed Reality Mode<br/>
<br/>
In early 2011 there was an e-mail at the X3D-public mailing list. Somebody suggested to think about robots, drones and so on. This mail turned me on to think about "real reality". I concluded, we would need an (N+1)th scene instance.<br/>
<br/>
Besides the N scene instances for the N "virtual identities" (aka players of a multiplayer game) we would need another - invisible - scene instance that would relay the Network Sensor states and events to and from the "real reality" (which was called Internet of Things, as I learnt later).<br/>
<br/>
A robot could be denoted as a "real life avatar" (from the viewpoint of the virtual reality), objects that would have a representation in the virtual reality would be denoted as "real life objects" and objects that would not have a representation in virtual reality would be denoted as "collateral entities".<br/>
<br/>
Since N >=1 that means we had at least 2 scene instances in Mixed Reality Mode, one visible and one invisible.<br/>
<br/>
Nowadays I am rather thinking of N + M scene instances, but that's a detail I do not want to discuss in all the details now. Probably I am again trying to invent the wheel another time :-)<br/>
<br/>
<br/>
<br/>
Thanks,<br/>
<br/>
Christoph<br/>
<br/>
</div></div></body></html>