opengc-devel

[Opengc-devel] status/Avsim conference update/RFC

[Damion S. <be...@cs...>]

Hi,

I wanted to give everyone a report of the feedback I received after 
"unveiling" OpenGC at the Avsim conference in Reading yesterday (I gave 
a 40 minute presentation). The overall reaction was extremely positive; 
in fact, amazingly so. It appears that there is more of an interest in 
glass cockpit software than I had thought.

I spent several hours today talking with both Enrico Schiratti 
(developer of Project Magenta) and Austin Meyer (developer of X-Plane). 
Having met both of them in person, I would like to put to rest negative 
claims that have been made about both; they're extremely cordial guys, 
and quite supportive of OpenGC. Both Enrico and Austin (Enrico in 
particular) had interesting thoughts on developing sim add-on software, 
and some of their opinions have caused me to rethink the place of 
OpenGC within the simming community somewhat.

As the result of conversations with them and a number of conference 
attendees, I would like to summarize the feedback I received regarding 
OpenGC:

1) People are impressed with its multi-platform, multi-sim capabilities.
2) The open-source nature of the project is not, in and of itself, a 
selling point of the project (keeping in mind that this is largely a 
FS200x, and to some extent X-Plane, oriented crowd).

I have personally believed that the main thing that differentiates 
OpenGC from Project Magenta (and indeed other cockpit efforts) is it's 
multi sim/platform support. While I do not wish to make product 
announcements for other projects, it is worth noting that this will 
likely _not_ be true for the indefinite future. Take that for what you 
will...

This means that OpenGC will soon need to offer a compelling feature 
set, rather than resting on the status of supporting everything under 
the sun. To that end, I have started to rethink the nature of OpenGC. I 
would caution everyone that these are _only_ preliminary thoughts and 
I'm primarily interested in soliciting feedback. In a nutshell, during 
the drive home today, I have debated adding a payware/closed source 
aspect to OpenGC. My reasons are as follows:

1) Closed source development creates a viable business model that would 
allow development of gauges for which there is a non-trivial 
development & distribution cost. For instance, gauges based around 
commercial terrain and map databases (for which there are no freeware 
alternatives).

2) The homebuilt aircraft/EGyro component of OpenGC developed this 
summer will need to be accompanied by physical hardware development, 
which will require purchase of hardware in the multi thousands of 
dollars range. I am unable to float the cost of this myself with no 
expectation of return (for obvious reasons), which would essentially 
block development of what I think is a potentially very valuable 
addition to general aviation.

If commercialization were to occur, and I emphasize IF, I propose the 
following:

1) All of the current OpenGC code would remain publicly available and 
licensed under the GPL.

2) Any code contributed by authors other than myself, and any code not 
part of but used by OpenGC that is GPL'd (i.e. Simgear) would not be 
included in any commercial product, to satisfy the GPL. Code which is 
not authored by me includes the 777 EICAS, 737 PFD and EICAS, and the 
original version of the FlightGear data source. According to CVS logs I 
am the only developer who has touched the majority of the other files; 
I am happy to provide documentation of this to anyone who asks.

3a) A full executable version of the commercial OpenGC that functions 
only with FlightGear (or any future open source simulators) would be 
available, free of charge, in perpetuity. This is to encourage 
development of open source simulators.

3b) A "lite" executable version of the commercial OpenGC, supporting a 
selection of generic gauges, would be available to the commercial 
simming community free of charge, to promote the hobby among younger 
simmers who are working on limited budgets.

4) The source code of OpenGC (the commercial version, perhaps with 
specific gauges removed), would be available to academic institutions 
and personnel free of charge, for use in teaching or research.

I believe that these changes would result in very little impact to the 
use of OpenGC within the freeware community. Very little gauge 
development has occurred within the past year by persons other than 
myself, leading me to believe that people are largely interested in 
_using_ OpenGC rather than designing within it. My personal goal of 
providing OpenGC as a teaching tool would continue with essentially 
zero modifications.

This leaves the following as potential targets for full 
commercialization:

1) Development of more polished gauges, more in line with the Project 
Magenta work, and any gauges which require data unavailable in free 
format.

2) The homebuilt aircraft market.

3) Custom gauge development, either on a contractual basis or by 
licensing a copy of the OpenGC source for commercial development.

Again, I caution everyone that this is not an announcement, merely a 
request for comments. Feedback is of course welcome, and I hope this 
does not come across as a criticism of those who HAVE contributed a 
great deal of their time to aspects of OpenGC's development (you know 
who you are).

Thanks in advance for any feedback, and remember, regardless of what 
happens:

1) OpenGC will remain free for use with free simulators, with the added 
benefit of the free version being supported by commercial development.

2) The source will remain available to students and educators, for 
non-commercial use.

3) I'm not out to price-gouge anyone (neither is Enrico, for that 
matter - really).

Cheers,
-Damion-
<donning flame retardant suit>

---------
Damion Shelton
The Open Source Glass Cockpit Project (OpenGC)
Carnegie Mellon University, Robotics Institute
http://www.opengc.org
da...@op...

Re: [Opengc-devel] status/Avsim conference update/RFC

[John W. <ca...@mm...>]

> If commercialization were to occur, and I emphasize IF, I propose the
> following:
>
> 1) All of the current OpenGC code would remain publicly available and
> licensed under the GPL.
>
> 2) Any code contributed by authors other than myself, and any code not
> part of but used by OpenGC that is GPL'd (i.e. Simgear) would not be
> included in any commercial product, to satisfy the GPL. Code which is
> not authored by me includes the 777 EICAS, 737 PFD and EICAS, and the
> original version of the FlightGear data source. According to CVS logs I
> am the only developer who has touched the majority of the other files;
> I am happy to provide documentation of this to anyone who asks.
>
Whoaa.. I seem to have a copy of an email where you expressed " I like what
you've done with the VVI"  so before you go making any claims of exclusivity
suggest you tread very lightly.  And there are several shots of a Nav
Display circa Feb 2001 produced by myself. And there may have been others
that made incremental changes and contributions.  I don't think CVS logs
hold ground truth.

Have not read all the details and finer points of the GPL but I think you
need the concurrence of ALL the developers and contributors (past and
present) to make any changes to the license.

Regards
John W.

Re: [Opengc-devel] status/Avsim conference update/RFC

[Damion S. <be...@cs...>]

Hi,

Please be assured that I _am not_ making any claims of exclusivity 
regarding code within OpenGC. As I said, this is merely a request for 
comments, NOT a plan of action.

> Whoaa.. I seem to have a copy of an email where you expressed " I like 
> what
> you've done with the VVI"  so before you go making any claims of 
> exclusivity
> suggest you tread very lightly.  And there are several shots of a Nav
> Display circa Feb 2001 produced by myself. And there may have been 
> others
> that made incremental changes and contributions.  I don't think CVS 
> logs
> hold ground truth.

The current nav display, insofar as the map portion is concerned, was 
written entirely by myself, completely from scratch. It does not use 
any code from Simgear, nor is it based on the nav display contributed 
12 months ago by Jurgen and Michael. This can be easily verified by 
looking at the code in the CVS repository.

The overlay - VorCntr - is, as you point out, code you contributed some 
time ago. Again, I AM NOT claiming sole authorship of OpenGC; I would 
like to make this point extremely clear. In the presentation I gave at 
Avsim 3 days ago, I included a slide with the following credits:

John Wojnaroski (777 EICAS & FlightGr.)
Michael DeFeyter (737NG)
Jurgen Roeland (737NG)
Jeff Ray & PCFlightSystems (EGyro)
The many authors of the software libraries used by OpenGC

Also note that there are many files within OpenGC for which I am not 
even a contributing author, much less the original author, for instance 
ogc747EICAS.cpp, all of the 737 code, the FMC, MCP, and John's original 
navigation code (this is not an exhaustive list).

> Have not read all the details and finer points of the GPL but I think 
> you
> need the concurrence of ALL the developers and contributors (past and
> present) to make any changes to the license.

I agree. I am not _at all_ suggesting that the current license be 
changed for the code as it exists in the CVS repository. All of the 
current codebase would remain GPL'd forever. However, as I said, a good 
portion of OpenGC has _never_ been modified by a third party. The GPL 
merely states that if a license is changed on the file it must be done 
with the full agreement of all of the copyright holders (contributors) 
to that code. In many cases in OpenGC, particularly in the base portion 
of the code (the ogcGauge and ogcFontManager classes to name two), I am 
the sole copyright holder of the code. As I mentioned above, there are 
other cases where different authors are the sole copyright holder of 
the file in question.

This brings up a few issues I would like to raise for the community as 
to why I am considering close-sourcing this project at all. Contrary to 
what it might seem like, I am not discussing this for the sole purpose 
of pissing people off.

1) There have been _no_ CVS submissions by persons other than myself 
since mid-September of last year; the most recent submissions were by 
Jurgen and Michael, with their 737 code. They have since left OpenGC 
and started their own project. This means that for the past 12 months, 
I have been for all practical purposes the sole developer of OpenGC. As 
you might imagine, this is a bit frustrating, since the original intent 
of the project was to encourage submission of gauges from the 
community. I can only conclude that the community is (with a few 
notable exceptions, John included), not interested in contributing.

2) Despite the open source nature of FlightGear, no one has stepped 
forward to maintain the data source object. There has been only one 
significant change to that in previous 18 months - a rewrite I did of 
John's original data source using Plib instead of Unix code - so I can 
only conclude that the actual level of interest in home cockpit 
construction using FlightGear and OpenGC is similarly low. A good deal 
of the code that originally worked with FlightGear - the FMC and MCP - 
has been non functional for nearly a year because of lack of support.

I raise these points not as a criticism, merely as an observation. I am 
not attempting to belittle anyone's contributions to OpenGC, but the 
fact remains that for the past 12 months very little has been 
contributed by the sim community at large. If there is not enough 
interest in this sort of project to create a viable product using open 
source methods, then it may be time to move the project in a different 
direction.

A few final points that bear repeating:
1) I have been careful to credit other developers both publicly and in 
the source code.
2) I have NO INTENTION of ever using code contributed or modified by 
persons other than myself in a commercial product.
3 These are discussions only; I have no interest in running off with 
OpenGC in the cover of night. If this were my goal, I would not be 
soliciting feedback.

Cheers,
-Damion-

---------
Damion Shelton
The Open Source Glass Cockpit Project (OpenGC)
Carnegie Mellon University, Robotics Institute
http://www.opengc.org
da...@op...

Re: [Opengc-devel] status/Avsim conference update/RFC

[John W. <ca...@mm...>]

Hi Damion,

Like you, I have been both disappointed and puzzled by the singular lack of
support and interest within the sim community (hardware and software) for
projects like OpenGC. Yet there are any number of websites about projects
and individuals building flightdecks, selling software, and offering all
forms of products and services.

Over the course of the past year I been focused on expanding the FG and
OpenGC interaction and network design. The result is that my system has
evolved into a highly customized simulation with FlightGear and OpenGC.
Throw in recent hardware designs ( and software drivers ) that provide for
throttle/control input, momentary and static switches, rotary encoders,
light and LED displays, etc and one has pretty much all the technical means
to build a world class simulator. Just a question of high deep are your
pockets for cockpit hardware and visual systems and how much time to devote
to construction.

I am still troubled by your claim regards the source and copyright. The code
was moved from SourceForge so there is a question of continuity and again I
don't think CVS logs (which can be altered and edited) are sufficient proof.
I can recall making and committing changes to all segments of the OpenGC
code from the make files to basic modules such as ogcAppObject,
ogcRenderWindow, and ogcDataSources. So again, I would challenge your claim
of copyright and lay claim to those portions and modifications within the
larger document(s).

Again, if you wonder why there is a lack of interest, perhaps this is a good
reason.

Regards
John W.

Re: [Opengc-devel] status/Avsim conference update/RFC

[Damion S. <be...@cs...>]

Hi,

> Like you, I have been both disappointed and puzzled by the singular 
> lack of
> support and interest within the sim community (hardware and software) 
> for
> projects like OpenGC. Yet there are any number of websites about 
> projects
> and individuals building flightdecks, selling software, and offering 
> all
> forms of products and services.

My impression, after talking to the folks at Avsim (who I would say are 
among the most fanatic members of the simming world) is that their 
hobby is flying, or perhaps to a smaller degree, constructing home 
cockpits, and not writing code. I think this is similar to the case of 
image editing; a lot of people will edit photos as a hobby, very few 
are interested in writing Gimp-like tools to do so. Moreover, the fact 
that people are interested in paying money to let other people do so is 
(IMHO) merely an indication that they're willing to pay for a 
professional quality product.

> Over the course of the past year I been focused on expanding the FG and
> OpenGC interaction and network design. The result is that my system has
> evolved into a highly customized simulation with FlightGear and OpenGC.

Ok, but this has not been contributed back to OpenGC as a whole. You 
attribute this to (from a recent email):

"since Damion went off on a different direction with the OpenGC project 
it's become incompatible and I just don't have the time, talent or 
resources to set up a website and CVS server or try to sync up with 
X-Plane or FSxxx."

If the only reason the project is useful to you is as the starting 
point for a private development branch (with the split occurring a year 
back), I don't know if that's a particularly convincing argument to 
continue development of the public version of the code. While there is 
nothing in the GPL that prohibits private development, it does little 
to improve the project as a whole.

> Throw in recent hardware designs ( and software drivers ) that provide 
> for
> throttle/control input, momentary and static switches, rotary encoders,
> light and LED displays, etc and one has pretty much all the technical 
> means
> to build a world class simulator. Just a question of high deep are your
> pockets for cockpit hardware and visual systems and how much time to 
> devote
> to construction.

Agreed - I think this is the view that the cockpit crowd is taking. In 
terms of the price to pay for a polished software product (which I 
think we can all agree OpenGC is not), it's really quite minimal 
compared to, for example, a single flat panel screen for an interior 
display.

> I am still troubled by your claim regards the source and copyright. 
> The code
> was moved from SourceForge so there is a question of continuity and 
> again I
> don't think CVS logs (which can be altered and edited) are sufficient 
> proof.

<sigh> If anyone wants to accuse me altering CVS logs in a 
conspiratorial grab for power... well, I consider myself an honest 
person and you're right, there really isn't any indisputable proof I 
can provide that would convince someone beyond a shadow of a doubt that 
I'm not lying.

> I can recall making and committing changes to all segments of the 
> OpenGC
> code from the make files to basic modules such as ogcAppObject,
> ogcRenderWindow, and ogcDataSources.So again, I would challenge your 
> claim
> of copyright and lay claim to those portions and modifications within 
> the
> larger document(s).

Fair enough. Look, if this is a serious issue (and it apparently is) it 
would be a fairly minimal amount of work to start completely from 
scratch. Although, my ability to convince people that I have started 
from scratch is a different question...

> Again, if you wonder why there is a lack of interest, perhaps this is 
> a good
> reason.

Is the suggestion that the lack of interest is due to my "claim of 
ownership" on OpenGC? The first time I've ever expressed a commercial 
interest in glass cockpit software was three days ago, and that was 
largely BECAUSE of the apparent lack of interest in contributing code 
to the project. It has never been suggested to me by anyone that I'm 
maintaining an inappropriate level of control over the project.

As a question to everyone, my main goal at this point is the following: 
I have a personal interest in starting a small company to develop glass 
cockpit hardware and software for homebuilt aircraft. OpenGC was shown 
at the EAA airshow in Oshkosh and, though it was quite popular, there 
was zero interest in the open source aspect of the project. None. 
Pilots want a complete drop-in solution and that's it.

To provide this: terrain databases (the good ones) cost money. Highway 
map databases cost money. Hardware costs money.

If someone can suggest a model whereby the open source and private 
development aspects of OpenGC can coexist, I'm all ears. One option 
would be moving the license to something like BSD or LGPL, although my 
guess would be that that would not be a particularly popular move 
either. My problem is not with having the bulk of the code of OpenGC in 
the open source arena; if I were that paranoid, I would not have GPL'd 
it in the first place.

Things get complicated in trying to figure out a way of releasing a 
"value added" version of OpenGC as payware. Apple has done this fairly 
successfully by releasing the Darwin core of OSX as open source and 
close sourcing their desktop. If something similar could be done with 
OpenGC, great!

Any suggestions?

-Damion-

---------
Damion Shelton
Carnegie Mellon University, Robotics Institute
A408-o Newell Simon Hall
412.268.3866 (office)
412.818.8829 (cell)
412.268.6436 (fax)
http://www.cs.cmu.edu/~beowulf
---------
I hope that after I die, people will say of me: "That guy sure owed me 
a lot of money."

Re: [Opengc-devel] status/Avsim conference update/RFC

[Manuel B. <li...@va...>]

Hi Damion,

just a couple of thoughts from me...

On Sat, Sep 27, 2003 at 11:25:04PM -0400, Damion Shelton wrote:
> 1) People are impressed with its multi-platform, multi-sim capabilities.
> 2) The open-source nature of the project is not, in and of itself, a 
> selling point of the project (keeping in mind that this is largely a 
> FS200x, and to some extent X-Plane, oriented crowd).
> 
> I have personally believed that the main thing that differentiates 
> OpenGC from Project Magenta (and indeed other cockpit efforts) is it's 
> multi sim/platform support. While I do not wish to make product 

I think, the "multi sim/platform support" should remain the primary
"selling point" for OpenGC. ("Selling point" not meant literally here)


> 1) Closed source development creates a viable business model that would 
> allow development of gauges for which there is a non-trivial 
> development & distribution cost. For instance, gauges based around 
> commercial terrain and map databases (for which there are no freeware 
> alternatives).



> 2) The homebuilt aircraft/EGyro component of OpenGC developed this 
> summer will need to be accompanied by physical hardware development, 
> which will require purchase of hardware in the multi thousands of 
> dollars range. I am unable to float the cost of this myself with no 
> expectation of return (for obvious reasons), which would essentially 
> block development of what I think is a potentially very valuable 
> addition to general aviation.

Have you considered 'modularizing' OpenGC (not necessarily on the binary
level, eg. shared objects/dlls/plugins, maybe on the source level...)

Why not leave the core of OpenGC with all the current Gauges under GPL
and continue to use this as the basis for free and commerical
distributions, while creating the possibility for commercial gauges to
just "drop" in their source in a subdir before building.

The resulting binary would be a payware program (which could be done
totatlly independent from OpenGC by a company who wants to use it that
way with their own special gauges)

As far as I undestand, this would mean the core needs to be
relicensed/dual licensed under LGPL.
If such a company wanted a change in the OpenGC core, it could add code
to the opensource CVS core OpenGC (that piece of code haveing the same license
as the core)

That way, every user/developer of OpenGC could profit from those
changes. I think such a development model is much more viable than
having a(nother) closed source Glass Cockpit Software.

My ideas go along the lines of how Mozilla/Netscape or
OpenOffice/StarOffice are orginized and developed.
Both have a free version and also a commerical version with a few more
things added to it that could not go into the free/opensource version.


For me the most important about OpenGC is the fact its free software and
the code is open source.


What do you think ?

> This leaves the following as potential targets for full 
> commercialization:
> 
> 1) Development of more polished gauges, more in line with the Project 
> Magenta work, and any gauges which require data unavailable in free 
> format.

I don't believe that there is a big market for this. Most of it is
already taken by PM. 

> 2) The homebuilt aircraft market.
> 
> 3) Custom gauge development, either on a contractual basis or by 
> licensing a copy of the OpenGC source for commercial development.

Why not set up a model (eg on "contractual basis") that all gauges that
are developed (even for money) to be made free/opensource after its
finished. What I mean is: some company wants a gauge for eg. their
homebuilt aircraft systems. They pay you to write it. When its finished,
they load OpenGC and the gauge they paid for onto their hardware. They
sell this hardware with OpenGC and that gauge in it to their homebuilt
customers. You put that code into CVS. They got what they wanted. OpenGC
has another gauge. You earned some money. 

 
Just some ideas... :-)

Regards,
Manuel

Re: [Opengc-devel] status/Avsim conference update/RFC

[Damion S. <be...@cs...>]

Thanks for the feedback!

> I think, the "multi sim/platform support" should remain the primary
> "selling point" for OpenGC. ("Selling point" not meant literally here)

Agreed. We're still unique in that respect.

> Have you considered 'modularizing' OpenGC (not necessarily on the 
> binary
> level, eg. shared objects/dlls/plugins, maybe on the source level...)

<snip>

> That way, every user/developer of OpenGC could profit from those
> changes. I think such a development model is much more viable than
> having a(nother) closed source Glass Cockpit Software.

Ok. Do you think it would be possible to reach an agreement to 
relicense the "base code" portion of OpenGC under the LGPL or BSD?

> My ideas go along the lines of how Mozilla/Netscape or
> OpenOffice/StarOffice are orginized and developed.
> Both have a free version and also a commerical version with a few more
> things added to it that could not go into the free/opensource version.

There are a few technical issues (handling multiple GUI interfaces for 
example) that would have to be addressed, but they're probably not 
insurmountable.

> For me the most important about OpenGC is the fact its free software 
> and
> the code is open source.

Well, that's what I had hoped for as well. Any ideas about how to drum 
up more interest in contributing to the code base?

> I don't believe that there is a big market for this. Most of it is
> already taken by PM.

This is what I spent over an hour talking to Enrico about at the Avsim 
conference. He doesn't agree that PM will always dominate the market, 
and has in fact expressed frustration that he's a bit overwhelmed by 
requests for new features. He would like "competition", if only to 
relieve some of the development pressure.

> Why not set up a model (eg on "contractual basis") that all gauges that
> are developed (even for money) to be made free/opensource after its
> finished. What I mean is: some company wants a gauge for eg. their
> homebuilt aircraft systems. They pay you to write it. When its 
> finished,
> they load OpenGC and the gauge they paid for onto their hardware. They
> sell this hardware with OpenGC and that gauge in it to their homebuilt
> customers. You put that code into CVS. They got what they wanted. 
> OpenGC
> has another gauge. You earned some money.

This is what I spent a bunch of time talking to Austin about. He has 
similar relationships with several companies that use X-Plane in their 
businesses; they pay him to add features with the understanding that 
whatever is created gets rolled back into the main release of X-Plane.

I think this is a _very_ good model under which to operate. I have 
worked in close cooperation with a company called Kitware at work, and 
one of their main products (VTK) is released under a similar model. If 
you're curious, the copyright under which they release their code can 
be found at:

http://www.kitware.com/Copyright.htm

The difference I see here is that without the ability to "hold back" 
some features, in the case where _I_ am the company doing the homebuilt 
aircraft development, there is really no way to fund the development of 
gauges that require a cash investment. In other words, I'd be perfectly 
happy if the following were true:

1) OpenGC still exists as open source
2) Anyone can contribute code to the open source version
3) I'm allowed to develop closed source add-ons that make OpenGC viable 
in the homebuilt aircraft market.

This requires something along the lines of a BSD/LGPL license. As long 
as everyone can agree to switch to that, I'm perfectly happy. My ONLY 
interest is being able to pursue an interesting new avenue of 
development that I don't think is possible operating purely under the 
GPL. I have absolutely zero interest in stopping development of OpenGC 
if there's a way to keep it open source yet still pursue commercial 
goals.

Thoughts? Opinions? John - since you're the other major developer in 
the history of OpenGC, what do you think about this?

Cheers,
-Damion-

---------
Damion Shelton
Carnegie Mellon University, Robotics Institute
A408-o Newell Simon Hall
412.268.3866 (office)
412.818.8829 (cell)
412.268.6436 (fax)
http://www.cs.cmu.edu/~beowulf
---------
I hope that after I die, people will say of me: "That guy sure owed me 
a lot of money."

Re: [Opengc-devel] status/Avsim conference update/RFC

[Manuel B. <li...@va...>]

Hi Damion,


On Mon, Sep 29, 2003 at 07:30:58PM -0400, Damion Shelton wrote:
> Ok. Do you think it would be possible to reach an agreement to 
> relicense the "base code" portion of OpenGC under the LGPL or BSD?

LGPL should be the easiest. BSD is too different (you know, there are
the GNU people and there are the BSD people)

With BSD, anyone can take the code and use it. Without telling you,
without contributing back changes... 
Like M$ uses the BSD TCP/IP stack in Win*. 

It might be worth looking at licenses of Mozilla and OpenOffice (MPL and
?) since they use the development model that I was using as an
illustration.

Why not propose to all the other contributors of OpenGC these ideas and
the choices of licenses and see what they say.

> There are a few technical issues (handling multiple GUI interfaces for 
> example) that would have to be addressed, but they're probably not 
> insurmountable.

Yeah, esp. when you have binary modules. Thats why I would prefer the
"source -module" way. 

> > For me the most important about OpenGC is the fact its free software 
> > and
> > the code is open source.
> 
> Well, that's what I had hoped for as well. Any ideas about how to drum 
> up more interest in contributing to the code base?

Hmmm... how about some simple "Hello World" -type tutorials and some
docs for both installation (source and binary) AND programming.
Try to appeal to the windows/MSFS "gauge programmer" type of people.

The docs should include how to get the sources and dependencies and
build tools, how to compile, how to add something "custom" like a 
"Hello World" gauge and how to set the thing up to talk to the sim.

Get those MSFS junkies to actually do something more than just work
within the confines of MSFS (and its hacks)

Also put up a short explanation of the (future) license so they know
what they and others can do and what they can't.

Most of the "freeware" for MSFS is very "unfree" in GPL/BSD sense.

> This is what I spent over an hour talking to Enrico about at the Avsim 
> conference. He doesn't agree that PM will always dominate the market, 
> and has in fact expressed frustration that he's a bit overwhelmed by 
> requests for new features. He would like "competition", if only to 
> relieve some of the development pressure.

Interesting. 

> This is what I spent a bunch of time talking to Austin about. He has 
> similar relationships with several companies that use X-Plane in their 
> businesses; they pay him to add features with the understanding that 
> whatever is created gets rolled back into the main release of X-Plane.
 
> I think this is a _very_ good model under which to operate. I have 

Yes. I try to work that way if possible. I think its is the BEST way to
develop software.

> worked in close cooperation with a company called Kitware at work, and 
> one of their main products (VTK) is released under a similar model. If 
> you're curious, the copyright under which they release their code can 
> be found at:
> 
> http://www.kitware.com/Copyright.htm
> 
> The difference I see here is that without the ability to "hold back" 

Holding back certain parts wouldn't be too much of a problem as long as
it is "special" stuff that wouldn't make much use to other people.
If "open source" is the "preferred" method (and maybe a little cheaper
than if not open sourced), that seems to be a good approach in my eyes.

> some features, in the case where _I_ am the company doing the homebuilt 
> aircraft development, there is really no way to fund the development of 
> gauges that require a cash investment. In other words, I'd be perfectly 
> happy if the following were true:
> 
> 1) OpenGC still exists as open source
> 2) Anyone can contribute code to the open source version
> 3) I'm allowed to develop closed source add-ons that make OpenGC viable 
> in the homebuilt aircraft market.
> 
> This requires something along the lines of a BSD/LGPL license. As long 
> as everyone can agree to switch to that, I'm perfectly happy. My ONLY 
> interest is being able to pursue an interesting new avenue of 
> development that I don't think is possible operating purely under the 
> GPL. I have absolutely zero interest in stopping development of OpenGC 

True. The GPL wasn't made for that. Some projects are using an amended
version of the LGPL, eg. the WxWindows toolkit (www.wxwindows.org)
This amendment has to do with linking... 
You might wanna take a look at it.


Regards,
Manuel

Re: [Opengc-devel] status/Avsim conference update/RFC

[John W. <ca...@mm...>]

> Thoughts? Opinions? John - since you're the other major developer in
> the history of OpenGC, what do you think about this?
>
As I understand the GPL there is nothing to preclude the use of open-source
software for commercial purposes provided the resultant code is provided
and/or made available along with the binary executable. The problem arises
trying to distinguish between source code that is integral to the program
vis-a-vis programs that link to a library or can be considered unique and
seperate

IMO OpenGC in it's current form and structure is not amenable to the spirit
and intent of the GPL. Quite frankly, starting an open-source project and
then suggesting a change in the license is a bit ingenuous. My suggstion
would be to go for a "clean room" approach, but that will require a third
entity not exposed to the source to create a spec or design document based
on observations of the programs performance.

Trying to produced closed-source and open-source programs under the same
license will most likely result in the open-source side stagnating and
eventually being overwhelmed by the closed-source-portion. It would have a
very chilling effect on new ideas and innovations from the community at
large.

Regards
John W.

Re: [Opengc-devel] status/Avsim conference update/RFC

[Damion S. <be...@cs...>]

Hi,

> As I understand the GPL there is nothing to preclude the use of 
> open-source
> software for commercial purposes provided the resultant code is 
> provided
> and/or made available along with the binary executable. The problem 
> arises
> trying to distinguish between source code that is integral to the 
> program
> vis-a-vis programs that link to a library or can be considered unique 
> and
> seperate

So here's the problem... if the source code and binaries are made 
completely available for a finished product (I'm talking here about the 
homebuilt aircraft market), what's to stop some other hardware company 
- presumably with more resources than me - from releasing a competing 
product derived completely from OpenGC? While it is technically not 
legal to do this without also releasing the source code, I don't know 
that it would be possible to prove that they violated the GPL without a 
lengthy court process.

> IMO OpenGC in it's current form and structure is not amenable to the 
> spirit
> and intent of the GPL

Really, why is that? I've contributed everything I've done back to the 
project, provided a CVS server using my own resources, and paid for the 
web site out of my own pocket. All of the existing code is GPL'd, and 
all of the contributing authors are credited in the header of each 
file. I haven't made any money off of OpenGC to date, nor do I have an 
existing commercial project.

> Quite frankly, starting an open-source project and
> then suggesting a change in the license is a bit ingenuous.

Not really, as long as the new license is also open source. Something 
like the MPL simply says, "Do whatever you want as long as you 
contribute any code changes back to the project". In many ways, this is 
less restrictive than the GPL. Keep in mind that, given that I provided 
the initial codebase for the project and have been maintaining it on a 
solo basis for the past year, I would have the most to "lose" from such 
an approach since I would essentially be opening up my work to 
commercial development by others.

> My suggstion
> would be to go for a "clean room" approach, but that will require a 
> third
> entity not exposed to the source to create a spec or design document 
> based
> on observations of the programs performance.

I don't think so. Such a third party doesn't really exist, nor do I 
think one is necessary. I can always push development back to where I 
was prior to August 26 2001, since I have archive copies on CD. As I 
said, if the closed-source route were taken (and I'm now leaning 
towards trying to find a compromise that would allow the project to 
continue largely as open source) the best I can do is promise that I 
would not use anyone else's code in a closed source product. The GPL 
does not amount to a non-competition agreement.

> Trying to produced closed-source and open-source programs under the 
> same
> license will most likely result in the open-source side stagnating and
> eventually being overwhelmed by the closed-source-portion. It would 
> have a
> very chilling effect on new ideas and innovations from the community at
> large.

I disagree. If the bulk of the code is licensed under something like 
the MPL, then it would be up to developers to provide a value-added 
product. For instance, I could run everything _but_ the EGyro project 
as open source. As an example, I point out Kitware Inc. They have been 
quite successful running VTK as an open source product, yet also 
distributing a payware application called VolView that uses some of the 
components of VTK wrapped into a slick interface.

Also, I believe the Mozilla/Netscape example contradicts the "chilling 
effect" example; if anything, the Mozilla project has hurt Netscape's 
bottom line since their releases are often better than Netscape's. If 
the project really _is_ viable from an open source standpoint, then the 
community at large can create a completely open-source alternative to 
whatever commercial product is released. If not, then the community 
benefits from someone going the extra mile to create a polished 
commercial product.

Anyways, it seems like there are two options:

1) Start from scratch, create a new closed-source product that is 
functionally similar to OpenGC but based on new code (or on code from 
the original project prior to the Sourceforge days). As I would have to 
do this myself, there would be little incentive to continue development 
of the original OpenGC and this would effectively kill the project.

2) Re-license OpenGC under a less restrictive license more amenable to 
commercial development. This would allow the bulk of the code to remain 
publicly available, while still making it feasible to pursue a 
commercial product.

Personally, it seems like option 2 is the best route at this point.

-Damion-

---------
Damion Shelton
Carnegie Mellon University, Robotics Institute
A408-o Newell Simon Hall
412.268.3866 (office)
412.818.8829 (cell)
412.268.6436 (fax)
http://www.cs.cmu.edu/~beowulf
---------
I hope that after I die, people will say of me: "That guy sure owed me 
a lot of money."

[Opengc-devel] opengc "toolkit" proposal

[Damion S. <be...@cs...>]

Hi folks,

I've been doing some thinking over the past few days about how to 
address the future of OpenGC (both in terms of where it's going in 
functionality and development). I think I'll throw the following out as 
a straw man proposal to get feedback on. To summarize, since the 
following may be a bit long-winded: OpenGC will be rewritten into a 
form of which I am the sole author (or have express written permission 
to include the code of others), after which it will be released as open 
source under a license modeled after the VTK license.

1) I will completely rewrite OpenGC based in part on my source code 
from pre Dec. 2001 (the date of the first CVS contributions of others), 
in part on modern source code about which there is no dispute regarding 
sole authorship (the font manager, X-Plane data source, the NEW nav 
database, etc.), and in part on a new design I've been working on based 
on the ITK and VTK toolkits, http://www.itk.org and http://www.vtk.org 
respectively.

2) Under this new design, the appobject would disappear completely, as 
would the gauge and gaugecomponent classes (to be replaced by a more 
general widget class). Objects would also be reference counted, using 
smart pointers, to avoid memory leaks. The eventual goal would be to 
provide something along the lines of VTK; i.e., a finished 
"application" as such is not provided, but the tools to rapidly produce 
one are. Example applications would of course be included, but more 
full-featured applications (with advanced mouse interaction for 
instance) would be left as an exercise for the reader. Under this 
model, an application main() might look something like:

main()
{
	ogcRenderer* pRen = ogcRenderer::New();
	ogcDataSource* pDataSource = ogcFSUIPCDataSource::New();

	ogc777PFD* pPFD = ogc777PFD::New();
	pPFD->SetDataSource(pDataSource);
	pRen->AddWidget(pPFD);
}

The above code may look a bit odd to those of you not familiar with 
VTK, but is intended to show the basic steps involved in creating a 
simple one-gauge display that drives a 777 PFD from FSUIPC. The _major_ 
advantages of switching OpenGC to this modular approach are:

A) It becomes possible to use the CABLE tool ( 
http://www.itk.org/HTML/CableSwig.html ) to generate wrappings in other 
languages. Prefer to code your GUI in Java or Tcl? No problem. This 
would address the common question of "can I use OpenGC in language X".

B) OpenGC is not tied into a particular application framework. As it 
stands, OpenGC doesn't really exist without AppObject and main.cpp - it 
would be nice if this weren't true.

3) This code would be distributed under a modified form of the Kitware 
license/VTK toolkit copyright. This is an extremely open license (do 
whatever you want), with one exception: the version of _contributed_ 
code placed in the repository becomes copyrighted solely by myself. All 
users and developers still retain full rights to use and redistribute 
the code, with or without modifications, with or without included 
source.

I've talked to Kitware in the past, and they have adopted this license 
purely to avoid copyright dispute regarding the code INSIDE the 
repository; as the license states, users are still free to do WHATEVER 
THEY WANT with the code. Please take a look at the license here: 
http://www.vtk.org/copyright.php

This is likely to be the portion of this proposal that creates the most 
angst among others, so feedback is encouraged. Please keep in mind that 
this license has worked extremely well for Kitware; VTK has something 
on the order of 2000 active developers. The copyright clause does not 
mean that you're giving up copyright to code you've written; you're 
free to publish another version elsewhere copyrighted solely by 
yourself.

4) As I produce a "sanitized" version of OpenGC, it will posted for 
everyone to see. I do not want to be accused of using code which 
someone else may legitimately have a claim to, so the entire revision 
process will be conducted in the open. After it has been established 
that I am the sole author of version "X" of OpenGC, it will again be 
opened up for public contributions under the new license terms.

5) Code for which permission cannot be obtained for inclusion in OpenGC 
mark II will have to be removed from the repository. I will try and 
work with the FlightGear folks to produce a non-GPL'd version of the 
FlightGear data source. All of the data sources will be changing 
dramatically, both in function and variable names. The current version 
is very poorly designed, by my own admission, and does not allow any 
form of bi-directional communication.

Given how well things have worked out for the Kitware folks and the 
users of VTK (of which I am one) I think this is probably the best 
direction in which to proceed. I realize that this will effectively 
halt new development of OpenGC for 4-6 weeks during the rewrite, but 
I'd rather get this sorted out sooner rather than later. The current 
version of OpenGC, as licensed under the GPL, will continue to be 
available for download even after the new version is released.

Feedback is encouraged...

Cheers,
-Damion-

---------
Damion Shelton
The Open Source Glass Cockpit Project (OpenGC)
Carnegie Mellon University, Robotics Institute
http://www.opengc.org
da...@op...

Re: [Opengc-devel] status/Avsim conference update/RFC

[John W. <ca...@mm...>]

Hi
>
> So here's the problem... if the source code and binaries are made
> completely available for a finished product (I'm talking here about the
> homebuilt aircraft market), what's to stop some other hardware company
> - presumably with more resources than me - from releasing a competing
> product derived completely from OpenGC? While it is technically not
> legal to do this without also releasing the source code, I don't know
> that it would be possible to prove that they violated the GPL without a
> lengthy court process.
>
So why should anyone contribute source code to an open-source project that
anyone can use for profit without providing any value added from a derived
product? I cuts both ways. Example, I've got a pretty decent FMC running on
my system ( My masters thesis at Purdue was on control and optimization of
shuttle approach trajectories) and with a little effort and application of
current control theory you could turn it into a reasonable autopilot. That
code will NEVER leave the confines of my hard drive.

Plus there is the added potential for litigation.

The last thing I need is for someone to grab all or part of my code and the
algorithms, put out a commercial product, and the next thing I know my
attorney and I are in court because either the manufacturer goofed in
modifying the code and/or some idiot pilot flew into the side of a mountain.
Yes, I know what the license and warranty state...  try telling that to the
judge and jury. And funny how lawyers seem to gravitate towards "deep
pockets"

I don't know what to suggest. You seem determined to try a commercial
venture that involves a high level of risk ( wonder why they call it
experimental/home built ) with software and derivations based on an
open-source license and project that started out as a learning exercise.  I
simply cannot agree with such a change

Hopefully, you understand my position.

Regards
John W.

Re: [Opengc-devel] status/Avsim conference update/RFC

[Damion S. <be...@cs...>]

Hi,

> So why should anyone contribute source code to an open-source project 
> that
> anyone can use for profit without providing any value added from a 
> derived
> product? I cuts both ways.

It's a personal choice. It depends I suppose on whether or not you 
consider the change so novel as to be revolutionary. E.g., if I notice 
a small bug (let's say a color is a little off) in OpenGC, I'd probably 
be willing to make the change even understanding that someone else may 
profit from it. If I come up with a completely revolutionary gauge 
design, the question is a bit more difficult. All I can say is that 
this model _has_ actually been successful...

> Example, I've got a pretty decent FMC running on
> my system ( My masters thesis at Purdue was on control and 
> optimization of
> shuttle approach trajectories) and with a little effort and 
> application of
> current control theory you could turn it into a reasonable autopilot. 
> That
> code will NEVER leave the confines of my hard drive.

Regarding the FMC, that's certainly your prerogative. Likewise, a bunch 
of the code I've written at work will remain private as well. But, your 
FMC also illustrates my current frustration with OpenGC - lack of code 
being contributed back.

> Plus there is the added potential for litigation.
>
> The last thing I need is for someone to grab all or part of my code 
> and the
> algorithms, put out a commercial product, and the next thing I know my
> attorney and I are in court because either the manufacturer goofed in
> modifying the code and/or some idiot pilot flew into the side of a 
> mountain.
> Yes, I know what the license and warranty state...  try telling that 
> to the
> judge and jury. And funny how lawyers seem to gravitate towards "deep
> pockets"

Well, it seems that this risk applies with the GPL as well. The only 
added requirement with the GPL is that the source be made available. 
I'm a bit skeptical that releasing the source provides any sort of 
legal buffer as far as liability goes. If this was a real concern, open 
source couldn't exist. With both projects I've been involved with (VTK 
and ITK) the liability is placed on the system integrator, who is 
responsible for testing to the satisfaction of the regulatory agencies 
involved, primarily the Food & Drug Administration with what I've done 
so far. This is in fact a main argument for permitting commercial 
development of open source code; getting regulatory approval for these 
sorts of products is insanely expensive.

> I don't know what to suggest. You seem determined to try a commercial
> venture that involves a high level of risk ( wonder why they call it
> experimental/home built ) with software and derivations based on an
> open-source license and project that started out as a learning 
> exercise.  I
> simply cannot agree with such a change

Understood. Hopefully the proposal I mailed out will address this. As I 
mention in it, any revised license version will be available for you 
(and others) to examine and verify that none of the "debated" code is 
included. However, as I've said, the only difference between the 
current and proposed versions are the ability to use the code without 
restriction. Trying "a commercial venture that involves a high level of 
risk" is not prohibited under the current license...

> Hopefully, you understand my position.

Yes, I do. I hope you understand mine is well, even if you don't agree 
with it.

Best,
-Damion-

---------
Damion Shelton
The Open Source Glass Cockpit Project (OpenGC)
Carnegie Mellon University, Robotics Institute
http://www.opengc.org
da...@op...

Re: [Opengc-devel] status/Avsim conference update/RFC

[John W. <ca...@mm...>]

> However, as I've said, the only difference between the
> current and proposed versions are the ability to use the code without
> restriction. Trying "a commercial venture that involves a high level of
> risk" is not prohibited under the current license...
>
Understand the risk I allude to is not the lose of capital or business
failure, but the loss of life and limb. Burt Rutan finally gave up licensing
the long-ez design because of litigation and pressure from his commercial
partner BeechCraft ( I believe) The Starship while a radical design for a
business class aircraft met all FAA standards and certs, but it never was
able to shed that image of "experimental".

My hunch is that OpenGC will die a quiet death and I wish you well in your
new venture.

Regards
John W.