NMRAnet

[CaptOblivious]

For those who care or know about this particular topic, the NMRA board has finally approved the first round of standards defining NMRAnet. For those who don't know, NMRAnet is an emerging standard, an open and unencumbered standard, defining a stack of network protocols that allow for various layout elements-turnouts, signals, panels, trains, PCs and tablets, anything that can be electrically controlled- to talk to each other on a layout. Although forward looking, we're working hard to ensure compatibility with current technologies as well-as someone heavily involved in its development, I'm happy to talk about what it is at length.

 

Anyway, I though some here would be interested in the news.

[cteno4]

Capt,

 

hey sounds good! would love to hear more about it when you get the chance, great to hear about it from someone deeply involved!

 

jeff

[The_Ghan]

Let me guess .... they're based on RailCom ... right?

 

TrainController happily works with Digitrax, Lenz, and a range of other products, so the actual protocol can't be such a big hurdle.

 

Cheers

 

The_Ghan

[Guest Closed Account 1]

Yes, keep us informed.

[Martijn Meerts]

There's not that many commands available for most command stations, so the protocols are usually fairly easy. Once you've got basic serial or ethernet (depending on the command station) working, adding support for a new command station is fairly quick. There are some companies who don't freely give out the protocol specs though, which is annoying (for example, ESU doesn't give you the protocol unless you've registered on their forum, and added a valid ECoS serial number)

 

A lot of the protocols aren't very optimal though, there's lots of room for improvement on what's currently available.

[CaptOblivious]

Let me guess .... they're based on RailCom ... right?

 

TrainController happily works with Digitrax, Lenz, and a range of other products, so the actual protocol can't be such a big hurdle.

 

Cheers

 

The_Ghan

 

Nothing to do with RailCom, nothing to do with DCC. This is all about coordinating the rest of the layout: turnouts, signals, block occupancy, cabs, etc. And while TrainController does happily work with those things, I'm sure, we quickly discovered and demonstrated over the course of an evening show-and-tell that getting JMRI to talk to NMRAnet devices was waaaaay easier than getting it to talk happily with LocoNet devices or DCC accessory decoders.

 

But again, more information is coming. I'm posting my two clinics introducing NMRAnet later today; wait to see those before jumping to any conclusions.

[Guest Closed Account 1]

nothing to do with DCC. This is all about coordinating the rest of the layout: turnouts, signals, block occupancy, cabs, etc. And while TrainController does happily work with those things, I'm sure, we quickly discovered and demonstrated over the course of an evening show-and-tell that getting JMRI to talk to NMRAnet devices was waaaaay easier than getting it to talk happily with LocoNet devices or DCC accessory decoders.

 

Are DCC decoders still used?

 

A lot of us have most of our collections DCC.  It would really hurt to invest again in our toy trains.

[CaptOblivious]

nothing to do with DCC. This is all about coordinating the rest of the layout: turnouts, signals, block occupancy, cabs, etc. And while TrainController does happily work with those things, I'm sure, we quickly discovered and demonstrated over the course of an evening show-and-tell that getting JMRI to talk to NMRAnet devices was waaaaay easier than getting it to talk happily with LocoNet devices or DCC accessory decoders.

 

Are DCC decoders still used?

 

A lot of us have most of our collections DCC.  It would really hurt to invest again in our toy trains.

 

Again, let me stress, this has nothing to do with DCC. The two are compatible, or you can use NMRAnet with any other control technology. NMRAnet is completely divorced from whatever technology you are using to drive your trains.

[CaptOblivious]

https://vimeo.com/channels/nmranet

 

This has the clinics, and will have much more, describing NMRAnet.

 

 

This one is an overview of the demonstration layout referenced in the clinics. I had originally intended to show that video at the clinics, until I realized that the layout would actually fit in my checked luggage! Let me add that this last video was originally meant as a bit of an advertisement for Railstars's products. I post it here in the hopes that it is informative.

[cteno4]

Capt, 

 

Nice presentations! Very informative. Unfortunately Vimeo pooped out half way through the tech pesentation, so I'll finish it later.

 

The future looks so bright with this I gotta wear shades! I can see how this will be quite useful and do a lot of the communication heavy lifting.

 

Cheers,

 

Jeff

[The_Ghan]

Cap'n,

 

Having watched the two videos, it appears to me that NMRAnet is a competitive product to Digitrax LocoNet.

 

With LocoNet, I can connect a number of DCC devices that all talk to each other directly - not through the DCC rail bus.  A number of manufacturers, such as CML Electronics, are producing products that work with LocoNet now.

 

Software, such as Train Controller, will talk to CML, Digitrax, and other devices through LocoNet.

 

This raises the question: what is an independent organisation like NMRA doing developing commercial products that compete with established manufacturers and how is this in the best interest of Members?

 

If I'm incorrect then please provide an example of how I might use NMRAnet to achieve something that cannot be done with LocoNet today.

 

Cheers

 

The_Ghan

[Guest Dpharris]

Cap'n,

 

Having watched the two videos, it appears to me that NMRAnet is a competitive product to Digitrax LocoNet.

 

With LocoNet, I can connect a number of DCC devices that all talk to each other directly - not through the DCC rail bus.  A number of manufacturers, such as CML Electronics, are producing products that work with LocoNet now.

 

Software, such as Train Controller, will talk to CML, Digitrax, and other devices through LocoNet.

 

This raises the question: what is an independent organisation like NMRA doing developing commercial products that compete with established manufacturers and how is this in the best interest of Members?

 

If I'm incorrect then please provide an example of how I might use NMRAnet to achieve something that cannot be done with LocoNet today.

 

Cheers

 

The_Ghan

 

The standards are being developed in public by the OpenLCB group, a group of modellers designing for modellers, and can be seen at OpenLCB.org.  The NMRA will consider and adopt these after due process as NMRAnet.  Neither group is developing commercial products, and the standard are open, so can be adopted by any manufacturer.  Reference hardware and software is available. 

 

The standards are meant to be general, flexible, and extensible, with support for small and large layouts.  They are independent of transport and traction control systems, and are meant to work with legacy systems. 

 

As to what one might achieve that Loconet cannot, OpenLCB is designed to be faster and have more capacity than Loconet, with multiple bus segments.  It has unique node identifiers, so there will not be conflicts between nodes, including in large modular layouts. 

 

Loconet in many cases is sufficient, my oNeTrak club and I use it.  OpenLCB is to be the next generation network, open to all. 

 

Regards,

David

[The_Ghan]

Yeah yeah .... and Iraq has weapons of mass destruction.

 

So, who is Dpharris?  NMRA?  OpenLCB perhaps?  Just how open is OpenLCB when stakeholders aren't even identifying themselves here?  Seriously!

 

The NMRA will consider and adopt these after due process as NMRAnet.

 

WTF?  What is due process?  It sounds inevitable to me.  So, just in case I was making an incorrect assumption, I checked OpebLCB.org and found this ...

 

The OpenLCB™ project is developing an easy but sophisticated network for model railroad control using today's technology.

 

It is the basis for NMRAnet / S-9.7 Standard bus.

 

... and further down on the same page ...

 

 

There is prototype hardware is based on:

 

* the AVR-based Arduino platform at Hobbyists, including:

* LEDunino, a ATMega328-based compatible from Silicon Railway;

* Railroad Shield from SPCoast;

* Io Developer's Board, an AT90CAN -based compatible from Railstars,

* Io:duino, an Arduino-compatible AT90CAN-based board also from Railstars

used as a platform for a throttle and a command station, and its associated

* CMDArduino, a library for a Command Station, from Railstars;

* the new DevKit from Railstars / NMRA.

 

Now seriously!  You expect me to believe that a commercial product is not being developed when it is abundantly clear that there is collaboration between Railstars, NMRA, and OpenLCB?  Yes?  Well, I still disagree.  I think the licensing page on OpenLCB sums things up pretty clearly:

 

OpenLCB: Licensing

This web site contains the work of a number of people. All the contributions to this site, including code and documentation, are the property of their individual authors. All contributing authors make their work available here under specific licenses. You must comply with the license terms to use these works.

 

All software, in any form, is made available subject to either the GPL version 2.0 license or successor, or the LGPL version 2.0 license or successor. If you are uncertain about the terms of these licenses or how they apply to what you want to do, you must consult OpenLCB or the original author before copying, modifying, distributing or using the software.

 

All documentation and other writings are made available subject to the Creative Commons Attribution-Share Alike 3.0 United States License. If you are uncertain about the terms of these licenses or how they apply to what you want to do, you must consult OpenLCB or the original author before copying, modifying or distributing the works.

 

All hardware designs are made available subject to the CERN Open Hardware Licence version 1.1. If you are uncertain about the terms of these licenses or how they apply to what you want to do, you must consult OpenLCB or the original author before manufacturing, modifying, distributing or using the designs.

 

For other arrangements, contact the specific author directly.

 

The “OpenLCB” name and associated logos are trademarks of the OpenLCB association. They may only be used in association with the documentation and software presented here. We are continuously working improving the documentation and software, including writing compatibility tests, and we reserve the right to publish our observations on compatibility of devices that use the “OpenLCB” name and associated logos.

 

The NMRA says that the “NMRAnet” name and associated logo are their trademarks. The NMRA determines the licensing status for its Standards and/or Recommended Practices.

 

Other terms appearing on this web site may be trademarks belonging to others.

 

I think NMRA members should consider carefully what is going on here.

 

I also think that covert negotiations need to cease.  JNSForum members should be forced to disclose their interests and OpenLCB.org should have an appropriate disclosure page.

[Martijn Meerts]

As far as I see it:

 

OpenLCB is a bus system similar to LocoNet, ExpressNet, Sx Bus etc., except it's free for anyone to adopt without having to license it.

 

NMRAnet is the same as OpenLCB, just with a different name after NMRA sets it as a standard.

 

RailStars the company is building (commercial) hardware based on the new standard.

 

RailStars the person (otherwise known as Capt.) is involved in defining the standard.

 

Dpharris (David) is also involved in defining the standard.

 

 

I know next to nothing about all that copyright and licensing stuff though, it's all way over my head. This whole free and open source and whatnot is nice, but their licensing documents are bloody impossible to understand :)

[Guest Dpharris]

I am David Harris, and am part of the OpenLCB team, also member of MERG, NMRA, and subscriber to numerous other egroups.  I have been working on layout control bus development for years, pre-dating CBUs from MERG. 

 

OpenLCB is open, but even open projects need licences.  If you don't believe that, then I suggest you investigate JMRI and the ensuing lawsuit that eventually set precedence to protect open source projects from predatory practices by commercial interests.  JMRI is probably the largest and best known open source MRR project.  The proceeded licences are standard licences designed for open software and hardware.  I agree they are complicated, they make my head hurt, however, the lawyer talk needs to be precise if they are to hold up in court. 

 

Please read NMRAnet.org which has been set up to distribute news about what is happening with NMRAnet, outlines process etc. 

 

Manufacturers have begun development of commercial products, otherwise there would be no products for us to buy.  DIY hardware is possible, and the reference designs are published.  The early development was on Arduino-type hardware, but other platforms are being used, including CBUS hardware, etc.  There is an OpenLCB / NMRAnet DevKit for developers and early adopters, which contains early products from Railstars and TCH Technologies. 

 

I suggest people check out the available information before forming their opinions. 

 

David

[CaptOblivious]

Geez, I take a few days' vacation, and come back to this? Nigel, sit back and relax a bit. Tomorrow, when I get home, I will be writing a lengthy paper that speaks to your concerns. I think I understand where you are coming from, but I also think you're missing quite a lot of the bigger picture. Give it a few days, let me get back to my office, and I think you will come to understand how these various projects slot together and are very much in the interests of the NMRA, its members, and model railroading generally. We are entering a new era of open standards, where the barrier to entry is as low as its ever been, a very exciting time indeed.

 

let me hasten to add that David here has zero commercial or financial interest in any of these goings on. Rather, he wants to help move model railroading forward because it is a good and useful project. Let me also add what an all around good guy he is, too, having finally met him in person last week.

[cteno4]

I actually find that having someone like capt willing to throw himself behind working on getting an OPEN standard done and willing to put commercial support behind the concept to allow folks to actually play with the stuff and help finalize the standard a HUGE benefit to the hobby. With open standards like this you have to have someone willing to throw themselves out there in the pre stages to get something working to prove the concept and let other developers and hobbyists have something to start playing with.

 

in a good standards process there is a good working group of some sort of org that tries to look out for the industry (NMRA in this case) and companies that will produce product and users of those products. Usually there is someone that brings a significant technology to the table that trumps most everyone there and standards are set around it with maybe some mods, then standard is set and then the company that owned the technology gets a fat royalty license. In this format railstars are not getting a royalty. their skin in the game is to get a good standard set that is easily doable on the design and production side while being as forward looking and as open as possible so this can be used far and wide as possible. thats pretty altruistic in this day of business!

 

The alternative is to have large companies just come out with closed standards that dont interoperate well or at all, they decide how it will all be done w/o input from anyone else, and you are their mercy. even if they will license the standard but control it, anyone that does get on that bandwagon is at their mercy as well. single companies tend to be slow at change, can lack technical expertise sometimes needed, and can not have good long term plans or vision with their business or the technologies Open standards are a way to try to get around this. with open standards the base stuff is developed and supported by a broad group that all have an interest in a good standard being adopted and supported. They can share expertise and vision outside of their own company's, possibly limited, business plan.

 

I've read a lot on dcc and automation but stayed out of it so far as its been so closed and once you choose a system thats it. I love the idea of something using a open standard like this that both allows everyone to use the same bus and at the same time takes a lot of the communication load off the users. I also get the feeling the fracturing of DCC and control technologies has slowed up that market from keeping up with both the rapidly evolving world of software, hardware, and networking, but also with how ideas get turned into product these days. i find that the unOpenness of a lot of dcc so far as lead hobbyists to be very protective of the manufacturer they have chosen and just set up more us against them mentality which i find very detrimental to the hobby, so having open standards is a good way to stop this.

 

Ghan, give them a chance here and let this all come out and be explained. Its not something sinister. I think the world of closed systems has also made folks suspicious when someone comes along and says they will pony up a huge amount of effort to make something jump ahead a generation or two in technology and make this open to everyone to use. But in this case i they are trying to do some good here and maybe eventually they can make a living at it, but things like this will never make you rich fast, if even get you a decent living!

 

Something like this does not happen in secret, its been in the mix with nmra for a while and im sure if you were a member and interested you could be a part of the group. capt was just letting us know of the early stuff railstars is doing to show what nmranet could do as well as showing off the work of the working group at the nmra conference, so it not hush hush stuff, hes just sharing it with us here.

 

jeff

[Guest Dpharris]

Thanks.  I think you got it in one. 

 

I can understand the_Ghan's suspicions --I too am cynical about hidden agendas. 

 

I have nil profit in this game.  In fact, I am several thousands of dollars to the negative: flights to NMRA meetings, sponsoring development, etc - in short *putting my money where my mouth is*.  I have a full time job with little spare time.  I am only continuing this project because if I didn't, it would have failed.  My wife will tell you that I have spent far too much time on this.  My layout is suffering.  So, being accused of nefarious dealings is a little thick, and gets my dander up. 

 

I am doing this because it needs doing, and almost no one else is stepping up to the plate.  The small group that deserves kudos, not brick-bats. 

 

That said, I am dedicated to seeing this through.  I am proud of what has been developed.  I have tried to make it the best system possible, which means complete, usable, and elegant.  The 'elegant' part is particularly important to me --- and that means as simple as possible without limiting the system, making each component as general as possible, and keeping as many doors open as possible. 

 

So much for philosophy :-).

 

My hope is that others will join us to make this a truly community effort. 

 

David Harris PhD MD <-- not as a status symbol, but rather to show that I have other rather serious pursuits. 

[The_Ghan]

Gentlemen,

 

I'll reply to each responder since my last post:

 

@Martijn: I generally agree with your assessment.

 

@Dpharris: Such a declaration should have been included in your first post.  Even better, in your signature.  This is exactly the point I'm trying to make: qualify your position, mate.  Perhaps it was an honest mistake and I'm prepared to give you the benefit of the doubt.  You mentioned that you're a member of a number of forums, you are also part of OpenLCB and NMRA.  Unlike you, I confine myself to a couple of forums.  I'm outside America.  NMRA site is generally of little interest to me and I'm not a member.  Therefore, any assumption that I might automatically know the situation in detail would be incorrect.  For this reason, my concern is transparency in-forum.  While I accept your methodology, none of it has been disclosed in-forum.  Members here make purchasing decisions based on advice they receive.  They have the right to know the motivations of such advice.

 

@Cap'n: I'm sorry about that.  Yes, we have discussed this privately.  I did want to give you time away, but the first post from Dpharris triggered me into action.  I do truely hope that Railstars is successful, but I'm reminded that you and I have had several private conversations about DCC, and about Digitrax.  I think it is only fair that you should have disclosed your position to me at that time, some 6 months ago, or more.  Until a few days ago, it was my honest perception that you were doing Railstars as a side-line hobby.  It is a reasonable proposition that other JNSForum members might believe the same.

 

@Cap'n again:  I'll accept your best intentions for the benefit of the model railway fraternity if you'll remove your reference to the Digitrax SE8C when describing the alleged difficulties that NMRAnet will overcome.  BTW, the reference comes right on the back of a comment about NMRAnet not being about DCC.  Then comes the king-hit about the number of DCC addresses that the Digitrax SE8C board has and the allegation that it is difficult to set up.  The simple fact is this: setting a single board ID will set all of the required board addresses.  These addresses are laid out in a table in the manual.  The rest is a matter of hard wiring and setting up in software such as TrainController.  If you use the slow-motion turnout controllers on the SE8C it can be configured to set the appropriate signal aspects associated with the position of that turnout automatically. If you use a different stationary decoder for turnouts, ensuring both devices use the same address will put the SE8C signal control in sync with whatever decoder you are using.  I don't believe your representation of the SE8C in-video is unbiased, but that's not to imply it was deliberate either.  Perhaps a demonstration of NMRAnet features that aren't already available would relieve this situation;

 

@cteno4: I agree.  Having Don in the know so-to-speak can be beneficial.  I just wish we were in the know too !!!  You mention how a "good standards practise" works and I'll add that it is good standards practise to evaluate the products already on the market and report accordingly.  I'll plead ignorance here and look forward to someone pointing me to the White Paper on the topic that must have been written some time ago. 

 

@cteno4 again: You mention that Railstars isn't getting any royalty, but Railstars hasn't made any disclosure as yet.  There might be a fee for service arrangement with NMRA, IP might have been sold to NMRA, there might be copyright on product.  I'm not saying any of that is bad, just that we don't know formally what the relationship is.  From the NMRA end, if I was a member, I'd want full disclosure. I'd want to know how, where, and why my membership fees are being spent.  If such a document exists please let me know, I haven't looked for it.  From the JNSForum end we have an Admin who has a commercial interest, the extent of which we are yet to know.  What implications could that have on the independence of JNSForum?  Being a D&D fan, I'm accutely aware of how frank and open discussion about eTools on the Wizards community forum helped bring CodeMonkeyPublishing to its knees about 7 years ago.  I wouldn't like the JNSForum in-forum discussions being used inappropriately either, and I'm not saying they have, but I am saying they could be.

 

@cteno4 again: "i find that the unOpenness of a lot of dcc so far as lead hobbyists to be very protective of the manufacturer they have chosen and just set up more us against them mentality which i find very detrimental to the hobby, so having open standards is a good way to stop this."   I disagree.  I'm not against the idea of an open standard, but I'll use a software analogy again:  with OpenOffice being freely available why are there so many pirate copies of MSOffice out there?  With the many free OS out there, such as Linux, BEOS, HAIKU, and PC-BSD we are all still buying Windows and MacOS.  How about a couple of hardware examples: the Tandy TSR computer could be built from scratch at home.  I know a guy who made one here in Australia.  I know other people who went out and bought one, or bought a completely different brand.  A university in England is providing, free, the plans to make your own 3D printer.  They also sell a kit to make one.  There are also a dozen or more commercially available products on the market, none of which are based on the free product.  An open standard with open flexibility is a contradiction in terms and we could argue all day about the pros and cons.  I'll happily accept that it's going to happen and applaud it's public launch, but still promise you that it will have no affect on what big companies are doing.  Don't expect too many of them to come rushing in and sign up, dropping their own IP in the bin.  For time-poor people like me, shelf products are the way to go.  Yet, last week there was a fleeting moment of concern when I thought my product of choice, Digitrax, would be orphaned.  This will not be the case.  Other members who have invested in any existing DCC technology 

 

@cteno4 again: I don't think anyone on this forum is overly protective of the particular DCC technology that they chose to run with.  I'll happily explain the reasons why I chose Digitrax.  I'll happily agree that they don't make a cab decoder that supports headlight dipping and auto reverse.  But I'll then say this can be achieved with a $15 DZ123 and a 3c resistor, or the $18 Kato FL12.  Either way you get transponding, which is a feature I personally like.  I have no problem with anyone who chooses a different product as they might have different priorities, budget, etc.  No brand is without its limitations.  No open standard is going to be the panecia to these limitations either and trying to be would only result in the development of another camel - a horse designed by a committee.

 

@cteno4 again: "Ghan, give them a chance ..."  How long do they want?  The project wasn't started last week.  It's good that Don will reply formally over the next week.  All I want is the same disclosure and transparency that Bob from BTTrains demonstrated.  There was a man with integrity!  This is about transparency and disclosure within JNSForum, not about the success or failure of Railstars, NMRAnet, or OpenLCB.  I sincerely hope all three succeed.

 

@cteno4 again (finally): "Something like this does not happen in secret ..." well Jeff, yes, it does!  At least, to me.  It is the secret thing that has fanned the whole debate.  I'm looking forward to full disclosure, full transparency and a clear understanding of relationships being provided in coming days.  Frankly, I don't care if Railstars is being funded, sold IP, gets royalties or whatever.  Business is business.  Don is an intelligent man.  He has a right to make a living out of the service he can provide and the knowledge he brings.  But as a member of JNSForum, I feel I have the right to be well-informed about the nature of advice given and comments made in forum.  I believe I have a right to know who runs the forum.  I've joined this forum in the belief that it is independent and open yet it appears that may have been a misaprehension on my part.

 

Generally:

 

1. I believe all members, irrespective of the topic covered in this thread, should disclose any commercial or developmental interests and associations they have when posting advice and opinions in forum.

 

2. NMRA standards aren't that important.  Really!  I promise you!  Similar standards exist in other countries.  I doubt any manufacturer complies with them all.  They aren't enforcable.  Kato, for one, is using Digitrax products and won't comply with NMRAnet.  The sky won't fall and we won't stop buying Kato because of this.  It won't be a LocoNet killer, or any other bus system that is out there for that matter.  Digitrax probably couldn't really care less about NMRAnet.  But that shouldn't imply arrogance on the part of Digitrax or any other manufacturer.  Rather, we should all be reassured that these manufacturers maintain confidence in their product.

 

3. LocoNet compatibility is available under license.  It isn't expensive.  CML Electronics products are well priced, especially the Super Tower panel product.  CML signaling products will automate your layout signals without the need for TrainController, if you have incorporated occupancy detection.

 

4. Transponding is also available under licence.  It isn't expensive either, considering the Kato EM13 is widely available for less than $18.

 

5. Similarly, RailCom is available under license.  I know that Hornby have Railcom compatible decoders, for example.

 

6. Despite my comments immediately above, I'm not on a crusade on behalf of Digitrax.  I'm sure they can handle things themselves.  Despite having around $3,000 worth of Digitrax gear I'm blessed to be in the financial position that I can easly, dump the whole lot and start afresh with OpenLCB if it were a benefit.  I am on a mission for truth, justice and the American way, but I think that was Batman's line.  I'm going to steal it here.  In JNSForum let's be truthful, honest, and give credit to those who do good things.

 

To summarise:

1. I don't have a problem with an open standard;

2. But I do believe that if a standard is to be endorsed by NMRA then the NMRA should have properly evaluated and reported on the commercially available products to member before committing any funding to a new project.  If this has been done then simply point me to that document;

3. An open standard will not influence major players.  Digitrax won't come on board and neither will anyone else who has developed similar IP.  So don't hold your breath there;

4. I do have concerns about JNSForum members not properly disclosing their position and relationships.  The recent post by a new member, Dpharris, only serves to highlight my concern;

5. I have concerns about board Admin, their duties and responsibilities to members and the ability to run an independent board when they have commercial interests in the model railway industry;

6. I'm not wed to Digitrax, LocoNet or Transponding.  Yes, I'm a Digitrax customer, but that might change in the future.  For the record, I think the new RailCom Plus product is great, especially for modellers of live steam, and you can't go wrong with the Lenz Gold series decoders either.  I'd use them myself if they supported Transponding as well;

7. I think Don is a good bloke.  Over the last year or so we've had our fair share of pvt conversations, mostly on the topics of DCC and new forum software.  I'd like to be using Railstar products some day and had even suggested to Don that I'd buy a product that could turn my sidings on and off if he developed such a thing (maybe I will need NMRAnet sooner than I think ...  ).  I'm not beating up Don, Railstars, NMRAnet or OpenLCB.  I want all to succeed.  I run my own practice.  I get how Don feels about Railstars.  I'm reminded also that Martijn and Don recently invited me to be involved with the development of the new forum.  Being time-poor, I want to limit my role to testing and skinning, but I consider it an honour to be involved.  So please don't go thinking I'm out to get anyone here;

8. The time for full disclosure is when someone sitting around the table in a meeting says, "We don't need to tell them about that ... "

 

Cheers

 

The_Ghan

[CaptOblivious]

Because I don't have time for more: I advocate for NMRAnet because I believe in it. What I, and Railstars, get in return for my significant contributions to the project is attribution. Nothing more. I am paid $0.00 for my efforts, and that is scheduled to continue for eternity. Moreover, I have put my work out there to be used by anyone for any purpose. Chew on that for a moment. I didn't disclose that here, because it's all been disclosed somewhere more relevant, namely in the licensing terms for my software and hardware designs.

 

Patience, please, people. This is getting out of hand. Indeed, the tone of this conversation has already exceeded what I normally tolerate on this forum. Don't make me delete this thread.

[The_Ghan]

Fair call Don,

 

I've had my say on the subject.  If I've misinterpreted the situation then I'll make an appropriate resonse later.  Until then, I'll refrain from further comment ... except to say that I do still want a DCC product that can turn my sidings on and off ... and the "motherboard" that will accept a PM42, 2-3 BDL168's and an SE8C

 

Cheers

 

The_Ghan

[CaptOblivious]

Some preliminary comments.

 

First, as I have said before, I get basically nothing for everything that I give to the OpenLCB/NMRAnet project. I sure as hell am not being paid by the NMRA or any other organization, either directly or through Railstars. It is true that I run a business that manufactures NMRAnet-compatible hardware, but let's get the direction of causation right: Railstars uses NMRAnet because I, personally, really believe in the project, and because I have invested so much of my personal time in it. This is just another way to show my support, because without commercial manufacturers actually making things, NMRANet is nothing more than some pieces of paper. As Ghan rightly points out. It has to start somewhere, so why not with me?

 

So my advocacy of NMRAnet on this forum is simply part of an effort to inform and educate; some members have in the past expressed an interest, and I believe this development—as with any new NMRA standard—is relevant and timely. It is true that I sell NMRAnet-compatible products, but so do a few others, and so soon will many more. To this end, I've tried to be ecumenical in the content of my talks, and the content of the posts here, leaving out actual advertisement beyond giving the name of my business and indicating that we make NMRAnet products. I believe anything I've posted here would pass the FCC requirements for non-commercial radio, and thus should satisfy my own self-imposed injunction against commercial postings outside the designated forum.

 

 

 

Second, some may have noticed I'm not particularly fond of Digitrax's products. This is not because they are competitors with Railstars. The opposite is true, in fact. Again, the direction of causation here is something like this: After beginning to invest in a Digitrax system, I grew quickly frustrated with their more interesting products. It was this frustration that led me to teach myself electrical engineering in order to design my own system, one that someone like me who has a hard time grasping complex abstract concepts can use. I also get quickly tired of their attempts at creating a closed ecosystem in which, once you have bought in, it becomes prohibitively expensive to buy out. This attitude is setting the model railroading industry back by years, if not decades. So when I disparage Digitrax products, it is not in order to position myself in the market against them, it is because I genuinely, as a person not as a business, dislike them. They are my personal opinions. The observant will notice that I have nothing bad to say about products from other companies that Railstars competes against; if disparaging the competition were my aim, would I not cast a wide net?

 

Regarding the SEC8 in particular: I have been told by one JMRI developer that there is a large sub-culture within JMRI users dedicated to figuring how to make the SEC8 work with their layout. In most cases, it seems, getting it to work is non-trivial. This not because it is a bad product—far from it—but because it was designed for a very specific purpose in mind, and trying to use it for anything that deviates from this purpose even slightly is like pulling teeth. Why else would RR-Cirkits (a Loconet licensee BTW) feel the need to create the LNCP and other related products? I should note that I rather like Dick, the owner of RR-Cirkits, and his products, even as we compete. Why? Because, unlike Digitrax, Dick has done the hard work of trying to figure out how his customers will use his product, and designing it to be used as painless as possible. I have the greatest respect for this attitude. So, I stand by my comments about the SEC8, and I know I am far from alone in my opinion.

 

Finally, I fail to see how any of this compromises my role as owner and admin of this forum. Nothing I have done has been performed in secret. If I have failed to disclose anything, it's because there has been nothing to disclose. Everything I've written above is just to say: there is nothing going on that isn't already publicly available information in some form or other.

[CaptOblivious]

NMRAnet FAQ. Note: if you didn't see my post directly above, you should probably look at that first. I've also unlocked the topic, so feel free to reply to this or the previous post below. I won't be locking it again.

 

-What is NMRAnet?

NMRAnet is a set of emerging standards governing the communications channels between layout control devices. In other words, it is an effort to standardize the ways in which things like turnout controllers, block occupancy detectors, and so forth, talk to each other. The aim is to create an open-access freely-available set of standards that any manufacturer can use, to ensure the greatest degree of interoperability.

 

-Why is the NMRA getting involved when there are already commercial products to do the same?

The NMRA is, first and foremost, a standards-setting body that governs the model railroad industry on a voluntary basis. No manufacturer must comply with NMRA standards and recommended practices. But to the extent that a manufacturer would like to ensure that their products will be compatible with those from other manufacturers, they would be wise to pay heed to the standards, indeed, they would be wise to seek official sanction for their product (see below). Conceptually, there is no difference between standards for wheel flanges and standards for communications. The NMRA has already provided standards for digital communications to trains over the rails (namely, DCC); NMRAnet is the obvious extension of the NMRA's mandates and goals as an organization.

 

-What does it mean to be an open standard?

First, let me draw a distinction. NMRAnet is a standard, which means a bit of writing on a piece of paper that is given some normative force. NMRAnet also has implementations in hardware and software, but these implementations are not themselves NMRAnet. Consider the difference between the standards for wheel shape (S-4.2), and the physical wheels included in a manufacturer's models. Those wheels are not the standard, the piece of paper is. Those wheels are an implementation of the standard. Likewise, nor is particular piece of hardware or software itself the NMRAnet standard, rather they are implementations of the NMRAnet standard.

 

So, to be an open standard means, in this context, that the text on the piece of paper is available at no cost, and that there are no limitations on the implementation of the standards. Anyone is free to make an NMRAnet board or software package. Anyone. You don't even have to ask permission. Take it and use it. (This is in contrast to LocoNet, for which you must first ask permission to see the piece of paper, and then ask permission again to implement a board that conforms to that standard.) We even offer—and this is where Railstars enters the picture—reference hardware and software implementations that you can use to get started. These reference implementations are not part of the standard, but are guaranteed to conform to it. Take them and use them! That's why they are there.

 

-So if all this is so open, why are "OpenLCB" and "NMRAnet" trademarked?

While the standards are open, NMRA needs some legal mechanism to enforce compatibility. Without some kind of mechanism, it would be easy for one manufacturer to co-opt the standard (observe what Microsoft did to HTML in the 1990's), and effectively create a parallel closed standard. Such a mechanism also gives you the user confidence that a product you purchase really will be compatible.

 

Here's how the mechanism works. OpenLCB and NMRAnet will each provide a (perhaps overlapping) set of conformance tests. Currently, OpenLCB is working on their validation suite which is, I should hasten to add, freely and openly available under a generous open source license. Only boards that the OpenLCB group or the NMRA have certified as passing these tests will be permitted to use the OpenLCB or NMRAnet trademarks. Boards that do not pass, or have not been submitted to pass, will not be granted licenses. Notice that this means you can have an (official) OpenLCB board that is not an NMRAnet board. The precise legal details are still being worked out, but that is the gist of the mechanism: OpenLCB and NMRAnet are using trademark law to provide a mechanism for ensuring compatibility and the future openness of the standard. Full stop. Because the details are being worked out, we cannot say how much certification or trademark license fees will cost, but in the end, certification really only matters to manufacturers of commercial products, because for hobbyists and DIY applications, there's no need to slap a logo on the project.

 

This enforcement mechanism is patterned after the mechanisms the NMRA already uses with their trademarks on the NMRA Conformance Warrant (the "football" logo) and the DCC logo.

 

-Why should I care about open standards?

This question is part of a debate that has been going on for decades. I won't try to rehash the arguments here. Instead I'll let Tim Berners-Lee (look him up on Wikipedia if the name doesn't ring a bell) speak for me: http://www.scientificamerican.com/article.cfm?id=long-live-the-web

 

-But this isn't really going to change the industry, really. It's got no teeth, right?

In some loose sense, all NMRA standards are optional and voluntary. Despite this, the NMRA did a fantastic job of whipping the industry into shape back in the mid-twentieth century. It is likewise true that no manufacturer need implement NMRAnet. But it is the belief of the NMRA (as I understand it) and the various people working on the projects that once a critical mass of commercially available hardware is out there, and once model railroaders come to understand the value of NMRAnet, that market forces will provide a strong incentive for manufacturers to provide NMRAnet-compatible products. After all, this is more or less how it came to pass with their other standards as well. We may be wrong about how this will play out, but we cannot let fear of being wrong stop this grand experiment, not when we have the opportunity at hand.

 

-I've never heard of any of this? Why all the discussion behind closed doors?

If you've never heard of any of this before, it's for a lack of advertising on our part. When you're neck deep in code, there's not a lot of time to proselytize. You're hearing about it now because we now have the resources to really ramp up the PR machine.

 

That said, there are no and never have been any closed doors. This is a new experiment the NMRA is trying out: developing and ratifying a standard completely in the open. We're very excited about the possibilities that the open-source development model has for creating real innovation in model railroading. To that end, you are personally invited to get involved! Interested in getting involved in the technical development of OpenLCB, or maybe just testing out various implementations? Visit http://openlcb.org for instructions on how to get involved. Curious about how the OpenLCB tech gets turned into a standard, or wish to contribute your opinion to the process? Visit http://nmranet.org for instructions on how to get involved in the conversation.

 

-What advantages does NMRAnet offer over LocoNet?

 

Aside from the question of open versus closed standards, NMRAnet has several design goals that give it distinct advantages. First, it is designed to be as general purpose as possible, and so is compatible with a wide range of end devices and usage scenarios. Second, it is designed to be as future-proof as we can make it, so that it can adapt to future technologies.

 

Just for comparison, it is worth understanding how LocoNet works. Normally, a DCC stream must be generated continuously, and offers no method for multiple devices to access it simultaneously. This is why DCC requires one and only one command station—a single device  to produce that stream. LocoNet is, at its roots, a very clever way to permit multiple devices simultaneous access to the DCC stream. This way, you can have, for example, multiple throttles sending DCC packets at once: Throttles send these packets in the form of LocoNet speed commands. The LocoNet command station gathers all these command messages, inspects and prioritizes them, and then injects them into the DCC stream on the behalf of all the throttles. While it is also true that LocoNet provides a method for two devices on the LocoNet to talk directly to each other, their communication is largely restricted to exchanging what are effectively DCC messages.

 

NMRAnet is far more ambitious than this. NMRAnet does not put any kind of restriction on the data being passed, for example, the message contents doesn't have to be a DCC command. Unlike LocoNet, NMRAnet is not focused on providing a contention mechanism to the DCC stream. Indeed, NMRAnet isn't concerned with DCC at all. NMRAnet devices communicate in a more abstract, and hence far more expressive language that permits communication of arbitrary data. The basic building blocks of an NMRAnet are sensors and actors—what we call "Producers" and "Consumers". The sensors—"producers" in NMRAnet-speak—detect events of interest on a layout, and report this event on the network by producing a so-called Event Report. Event Reports are broadcast over the network, and picked up by the actors—"consumers" in NMRAnet-speak. A consumer that receives an Event Report it was programmed to watch for will then take some action on the layout (such as setting a turnout or lighting an indicator). Event Reports are themselves abstract, containing no concrete data about the event; rather, it is up to the consumers to interpret the meaning of a given Event Report. This producer-consumer model permits very easy creation of free-form layout behaviors.

 

But NMRAnet is not just producers and consumers. NMRAnet also permits the use of point-to-point communications to send large chunks of formatted data over the network—for example, to read or write large chunks of a boards's memory directly, to learn about the topology and capabilities of the boards on a layout, to send targeted commands, or to stream audio or video (eventually, we're still working on this last one!).

 

Finally, unlike LocoNet, NMRAnet is designed to have virtually no latency under heavy loads, so that a particular NMRAnet can be used by a large number of boards and users simultaneously without introducing the risk of accidents or the frustration that builds when devices don't appear to respond to commands in a timely manner.

 

-Does NMRAnet require DCC? Does NMRAnet obsolete DCC?

NMRAnet is orthogonal to questions of traction control. While it is true that the OpenLCB group is working on a traction control protocol that would permit the direct control of trains over an NMRAnet, this protocol is designed to be as abstract and general as possible, so that it can be used (with a suitable command station-like device) to control DCC trains, or DC train, or Selectrix trains, or what have you. So, NMRAnet is meant to, first, compliment DCC without requiring DCC, and second, provide a clear and easy upgrade path when newer train control technologies become available.

 

-Does NMRAnet require RailCom? Will NMRAnet work with Transponding?

NMRAnet does not care how particular nodes sense layout events, such as block occupancy. A particular NMRAnet node may sense occupancy using whatever method the designer envisions, including the use of RailCom or Transponding. NMRAnet, in virtue of its abstract design, makes it easy for an NMRAnet node to communicate simple occupancy events, or detailed RailCom or Transponding reports to other nodes on the NMRAnet.

 

I'll update this post with more questions as they arise.

[Guest Closed Account 1]

CPT:

 

Can you show us a picture of a NmraNet decoder next to a common N Scale decoder?

[CaptOblivious]

CPT:

 

Can you show us a picture of a NmraNet decoder next to a common N Scale decoder?

 

I'm afraid I cannot, because no one has yet developed an NMRAnet node that would fit into a loco; nor have the wireless protocols been worked out. Sorry!