openBVE Licence & Associated

[leezer3]

A Summary of the Current Situation

The primary licence for OpenBVE should be considered to be public domain: https://en.wikipedia.org/wiki/Public_domain
Newer code contributions are licenced under the BSD-2 clause licence, as this is the closest recognised legal equivilant to Michelle's original Public Domain intentions.
We also use third party libraries, for which the licences may be found in the About dialog, or here: https://github.com/leezer3/OpenBVE/tree/master/licenses

If you wish to create derivitive works, fork or redistribute OpenBVE, this is entirely your right.
Please however respect the other contributors who have put in many thousands of hours of work to get us to where we are today by crediting appropriately, and retaining the appropriate licence headers in your own source.

Furthermore, whilst there is nothing in the licencing terms requiring you to provide your source code or contribute back any improvements you make, we and the wider community would appreciate it if you do so.

Please use this topic for any queries, suggestions, debate or discussion on the subject of the project licence.

---

Debate And Discussion

Changing the licence to something more widely recognised has again been raised in #304
This is a specific issue to discuss the mechanics and roadblocks involved.

Now, fundamentally I'm by no means opposed to this.
The original driver for openBVE's current public domain licence was Michelle, and indeed she was actively opposed to any licence at all:

   My attitude differs from so called "free software" licenses. I try to
   explain:
 .
   With proprietary software, the developers usually try to restrict what
   you are allowed to do with the software to the uttermost extreme that
   is legally possible. This falls under the category of copyright.
 .
   Then there is something that is sometimes called green copyright. So
   called free software falls in this category. As with proprietary
   software, a license is used to tell users what they are allowed to do
   and what not. The difference to proprietary software is that so called
   free software licenses are usually much more permissive, yet the
   author decides what you are allowed to do and what not. So called free
   software is usually attributed to derive from "freedom" instead of
   "free of charge". Still, if the author employs a license telling other
   people what they can or cannot do, this inherently cannot have
   anything to do with freedom.
 .
   Then there is something called public domain, which I consider to be
   an inherently good thing as everyone has eventually the same rights as
   the author. However, releasing a work into the public domain is not
   legally meaningful in many countries. Also, releasing a work into the
   public domain (if possible) depends on the generosity of the author,
   and this decision is made by the author.
 .
   My attitude is even one step further: I am opposed to copyright, thus
   I marked the game with the legally not meaningful phrase of
   "anti-copyright". It is legally meaningless, because in every country
   that has some form of copyright, it would require these laws to be
   abandoned. Still, why do I have this attitude?
 .
   With a license, I would be telling other people what they can or
   cannot do. I am not such a person. I will not make any silly
   restrictions on how you can use this program, I will not tell you to
   put my name on any derived work, I will never ever give "permission"
   for you modify or redistribute the software, because I don't think
   that it is up to me telling you what you can or can't do. You should
   make this decision for yourself.
 .
   I hope this briefly explained the situation.

( https://metadata.ftp-master.debian.org/changelogs/main/o/openbve-data/openbve-data_1.4.0.5+dfsg-3_copyright )

Obviously, at this point, she's been gone for ~10 years, and I have had no contact.
The other contributors to the original openBVE code were:

• Anthony Bowden
• Odaykufan (Jens Rügenhagen)
• Paul Sladen

At the time I launched this rebuild / development effort (3.5 years ago), I was in contact with all of these three.

Anthony Bowden and Paul Sladen were both broadly supportive of the direction that things were taking, but TBQH neither was a long conversation and both have not responded to further contact.
Licencing was touched on briefly although not specifically, and I don't believe either would have had a specific problem with re-licencing to something like MIT / BSD. (N.B. The Debian link above notes that they / Ubuntu attempted to get Michelle to change to a recognized licence.
Paul Sladen was the original Debian packager and fixed some Linux bugs and IIRC was part of this discussion)

Odaykufan is potentially more problematic.
He is of the opinion that he, Michelle and Anthony are the only three 'official' owners of the name / project openBVE, and that this project should have been re-named.
Having said that the following was in the same message:

Of course, being released into the public domain, you can do with the
program whatever you want, including passing it off as your own or
passing you off as an official appointee selected by us to continue
our work, which you are not.

I've also had at one stage or another I think a licencing conversation with all the people who've contributed any major sort of code, usually with this sort of mess in mind.

My stance has always essentially been that the current situation, whilst not ideal at all is what we've been left with.
Intellectually, I'm not sure that's changed, but the practicalities of the situation when wanting to access third party CI resources may be catching up with us.....

@mgavioli and @cwfitzgerald were definitely happy at the time I last spoke to either to go to BSD-2 (my preference), and @s520 is suggesting MIT.
Both of these are fundamentally identical, with just different phrasing on the legal fluff.

I would not support GPL, and fundamentally anything similar / more restrictive doesn't sit well with me either.

[s520]

I'm not stuck to the MIT license. I think BSD is good as well.

I propose to make this program dual license with current license and open source license approved by OSI.

[ginga81]

At the #304 , I think that this license discussion is about documentaition's translating.

But, if Mr.leeser3 want to discussing about OpenBVE's main program and relating program's licensing, I will say my opinion.

In my opinion, I think that the OpenBVE have to keep the public domain.
But, if Mr.leeser3 want to change to BSD, I obey your opinion.
But at the document of forking of guide-line, it says that have to change the name of OpenBVE.
In this regard, I think that it is a bit difficult to change the license.

Why I always choose GPL, this license is keep copyleft, and protect from proprietary.
In this regard, I know that difference from completely freedom.

In my opinion, to keep the completely freedom, the program that should not to change to proprietary and non-disclosure of source code.
But the original OpenBVE's creator says public domain, so I obey this opinion.

BSD, MIT or any like license can modify code and change from copyleft to copyright one, and use for proprietary programs.
Some proprietary program is modify and using BSD license's program, but not disclosure of source code.
At this regard, I think too that it can say 'completely freedom', and near opinion of the public domain.

At the GPLV3, this change is for mainly protect from software patent.
But unfortunately, from the GPLV3, '0. freedom to run the program' has removed.
So I am a bit not support to GPLV3's opinion.

[leezer3]

The documentation and the program are I think somewhat interwoven, and we can't really discuss the implications of re-licencing one without the other.

The forking guidelines are another thing that came directly from Michelle / the original team:
https://openbve-project.net/documentation/HTML/information_forking.html

I suppose this then pulls us back into the argument as to whether this is a fork or more properly as I suspect a continuation, and whether the name etc. should have been changed-
At the time this was launched, no meaningful work had been done for over 7 years, and Odakyufan's site had vanished. From conversations had with him by both myself and others, he had no intentions of doing any further work, or maintaining a website etc.
The decision was taken after a community discussion that whilst retaining the name openBVE was in part controversial, due to the small community size and lack of interest from any of the original parties, retaining the name and the original project aims were better than the alternatives.

No idea whether this was the right decision or not, but there we go.

[ginga81]

This comment's opinion is our unified view.

I(ginga81) ,S520 and F81_tec200 was discussed about to change licensing from public domain to BSD.
If the lisence changes to BSD from public domain, the difference from public domain, 'The Original Three((Michelle), Anthony Bowden, Odakyufan, Paul Sladen)' 's copyright and the author's moral rights will protect by BSD forever.
As a result, Odakyufan's name and moral rights will protect forever.
And Odakyufan, he said that 'The Original Three is the only author that holding copyright and moral right. And he said that the another anyone must not hold copyright. '.
We thought that if change to BSD, this is consistent with the his opinion.
Because, If change, the original OpenBVE's copyright and moral right is going to protect by BSD.

Next, we discussed about that the public domain is end to the 1.4 or old, and from 1.5 or newer, we should change to the BSD.
In our opinion, from the 1.5, change to BSD, and original copyright is to keep 'The Original Three' ((Michelle),Anthony Bowden, Odakyufan, Paul Sladen).

We do not still hearing abot Mr.leeser3's opinion of copyright from 1.5.
If you will have copyright from 1.5 by BSD, because of The Original Three's opinion, we are doubt about that you will having OpenBVE only for you by BSD.

[leezer3]

My opinion?
I've been around since before BVE was introduced, and developing in one way or another for the whole if it's life.
Unfortunately, copyright tends to get used as a weapon whenever there's a disagreement, not as a tool. The GPL (and especially the later versions) takes that to a new level, attempting to force everybody to their ideology with the linking provisions etc.

My personal opinion would probably end up somewhere between Michelle's complete rejection of copyright and BSD.
I would hope that anyone using the works of another would have the good manners to credit as appropriate, and I use BSD as a gentle reminder to do that. If that doesn't happen, I'm personally not really bothered. A little disappointed yes, but not much more than that.

Having a recognised copyright / licence would therefore seem to me be a necessary evil.

[cwfitzgerald]

I agree in full with @leezer3.

For the record, I am completely in support of a switch to any of the BSD's.

I don't necessarily have the same views on copyright and licensing, but I completely agree that for this project with the history is has had, BSD should be in a nice and open license. GPL goes against that idea.

[sladen]

It would be great to move to a well-known licences (such as BSD-2), one cannot rewrite the copyright or licences of other authors without a statement from that author allowing it. That basically means that unless Michelle is contactable and willing to make such a statement, then it probably has to stay as-is.

The copyright on the code, openbve-data/translations and Anthony's example plugins, routes, trains that were also shipped in Debian/Ubuntu are all slightly different. This is another reason why each package was packaged separately—there are/were multiple upstreams.

The Public Domain statements and background obtained from Michelle in the end were clear enough for Debian.

(For the avoidance of doubt, anything contributed by Paul Sladen is welcome to be distributed under any DFSG-compatible licence: eg. Public Domain, BSD-2 clause, GPL-*, etc.)

[leezer3]

Legally, I'd suspect that's probably not quite true with public domain content, although that's a topic that I'm sure would keep a whole bunch of lawyers in pay for years!

---

My quasi-legal assessment is as follows (I have a good understanding of UK law and copyright, and for the avoidance of doubt, this interpretation was made under the laws of that jurisdiction):
The FSF note that all work is automatically copyrighted under the Berne Convention.
However, they further note that if specific steps are taken to disclaim the copyright, then the work may be taken as not copyrighted at-all.
References:
https://www.gnu.org/philosophy/categories.en.html & http://directory.fsf.org/wiki/License:PublicDomain

Michelle's clear intention from all original communications was to disclaim the copyright.
Most jurisdictions implicitly or explicitly allow copyright to either be given or assigned to another entity.

Disclaiming copyright is a murkier business.

To the best of my knowledge, the primary jurisdiction that explicitly does not recognise the disclaiming of copyright would be Germany.
Reference, see the section on the fundamentals of copyright law & assignment of rights:
https://en.wikipedia.org/wiki/Copyright_law_of_Germany

If in doubt, we should take the most restrictive legal interpretation possible. Whilst it's not clear whether the same situation would apply in the UK, for the avoidance of doubt, I'll assume that it does.
Therefore, you state, and I fully agree that we must assume that there is a 'licence' & following copyright of some degree attached to the code.

The FSF maintains a list of open-source licences here:
https://www.gnu.org/licenses/license-list.html
However, it's equally possible to create your own licence. Whilst in the software world, it's normal for a licence to be formal and written, it's equally possible to create your own implied licence.
https://en.wikipedia.org/wiki/Implied_license

With this in mind, let's look at the statements made by Michelle.
First, the forking guidelines:
https://openbve-project.net/documentation/HTML/information_forking.html
No mention here is made of copyright etc, just a request to change the name, so we can't really do anything with this.
Second, would be her statement, which I've reproduced in the first post. I think the relevant paragraph is this:

With a license, I would be telling other people what they can or
   cannot do. I am not such a person. I will not make any silly
   restrictions on how you can use this program, I will not tell you to
   put my name on any derived work, I will never ever give "permission"
   for you modify or redistribute the software, because I don't think
   that it is up to me telling you what you can or can't do. You should
   make this decision for yourself.

Now, we can read that one of two ways.

Way 1:
It's explicitly stated that permission will never be given to do anything.
With that in mind, a good lawyer could probably argue that any forks are 'illegal' (horrid word, but I haven't got a better one off the top of my head)

Way 2:
She specifically states that she does not require her name to be placed upon the software, or her permission to be gained for it to be redistributed etc.
If we then read it in conjunction with the paragraph above, she specifically states that her intention is to create a situation where everyone has eventually the same rights as the author
This would implicity and for that matter depending on your interpretation / reading of the entire document explicitly IMHO allow for re-licencing as desired.

---

It's entirely possible (and for that matter probable) that there are others out there who disagree with my legal interpretation.

So, therefore to my mind, the question is more whether it's worth the aggro of the change, as opposed to whether it's legally possible.

I have no answer to that. I'm happy to defend my interpretation of the legal situation, but the opposing view may well be equally valid, and without input from Michelle, that puts us back at the impasse.

Unfortunately, the only email address I've ever had for Michelle (reschanger (at) gmail.com) has been inactive since she left.

[leezer3]

Marginally secondary item, but related.
I strongly suspect michelle to be an anglicised version of her name, and that the correct name is likely to be:
Michèle Boucquemont

The only result on Google (an uncommon enough name!) comes back to a single person:
https://www.google.com/search?q=%22Mich%C3%A8le+Boucquemont%22
At the minute though, I don't entirely wish to start poking sleeping dogs, even 10 years later.....

[sladen]

Hopefully people are thankful with Michelle's contributions.

[Had it been known (at the time) that Michelle would switch shortly after to a mononymous name, then I would not have made the git commits using the provided name; and I suspect everyone else would be none-the-wiser…]

[cxong]

Please consider a license like Unlicense (https://choosealicense.com/licenses/unlicense/) or CC0 (https://creativecommons.org/share-your-work/public-domain/cc0/). They are equivalent to public domain in jurisdictions that recognise it, and provide equivalent permissions in those that do not.

[cwfitzgerald]

Both of those licenses have issues because in some places it is very hard to actually put things in the public domain. They are also not OSI approved.