Summoning Wars Forum

Development => General => Topic started by: Lastmerlin on October 19, 2011, 08:57:30 PM

Title: License Policy
Post by: Lastmerlin on October 19, 2011, 08:57:30 PM
Please note a very important change to the license policy:

All source code must be released under GPL V3, as before.
All ressources and content must be released under CC-BY-SA 3.0. There is no option to chose a different license anymore.
Title: Re: License Policy
Post by: mafm on October 30, 2011, 04:50:09 PM
Can CC-BY-SA be restricted to 3.0? Or several at once, but including 3.0?

Because, for example, if under 2.5, the artwork cannot be distributed in Debian:
http://wiki.debian.org/DFSGLicenses

I expect that Fedora or others, including Ubuntu, might have similar restrictions.
Title: Re: License Policy
Post by: dezGusty on October 31, 2011, 06:40:55 AM
There is always the option of dual-licensing: http://en.wikipedia.org/wiki/Dual_licensing (http://en.wikipedia.org/wiki/Dual_licensing) which can help with incompatibilities... The only issue now is that adding a license, or changing the existing one requires all the existing authors to give their permission for the change.
Title: Re: License Policy
Post by: loic on November 02, 2011, 07:48:24 AM
Hi,

Because the decision to enforce CC-BY-SA on data is a recent one, it seems logical that the latest version (i.e. 3.0 or later) is designated. However, it would not hurt to change the policy stated above to change the recommendation and explain that 3.0 should be chosen otherwise the game will not make it to the Debian GNU/Linux, Ubuntu etc.

My 2ct ;-)
Title: Re: License Policy
Post by: Lastmerlin on November 02, 2011, 09:37:17 PM
Following this argumentation I have decided to just update the the license number to 3.0 . As far as I know, no artists specified a version number - so we can just assume that the most recent one was meant.
Title: Re: License Policy
Post by: mafm on November 02, 2011, 11:34:22 PM
Version 3.0 was launched even before this project was launched, if I am not mistaken ( 23rd of February 2007, http://creativecommons.org/weblog/entry/7249 ).  So it seems natural that the version is at least 3.0, or better, 3.0 or later versions (so you cover future "upgrades").

But as dezGusty says, you should make sure that the people who created the artwork have agreed for the work to be published under this license, or dual/tripled/n-licensed it, or published under another more liberal license which can be changed to this one.

I am going to update License.txt.  Please let me know in the future if something important changes in this regard (e.g., artists refuse to give their consent).
Title: Re: License Policy
Post by: Kulik on November 09, 2011, 10:34:01 PM
Regarding today's discussion about GPLv3, I think sumwars is better off with GPLv2+ (GPLv2 or, at your discretion, any later version). It doesn't limit the project whilst GPLv3 does (no deployment on iOS devices and various other devices, cutting off lots of potential users).

I think it's not too late to switch back.
Title: Re: License Policy
Post by: mafm on November 10, 2011, 12:26:21 AM
The question of whether GPLv3 restricts the project is quite idiotic from my POV.  GPLv3 protects users from TIVOisation (which sneakily work around laws to get away with using GPLd software while striping benefits for users); fights against DRM, which is one of the most stupid things ever invented (unless you're a corporation interested in milking the consumers); and also fights and protects users from software patents, which are going to be one of the technology cancers of this century (they already are at this point, see all the fighting over Java, Android et al.).  Since I'm not on the corporation side of the fence I have zero interest in going backwards and downgrading the license to GPLv2.

But for me, it's quite telling the notion that people is worried about this and not, for example, the non-commercial restrictions of the fonts that I raised recently, and it was only me who did some tangible effort to avoid this problem.  It is quite disappointing, really.  This non-commercial restriction even forbids you to ask for donations on your website, if you put a link to a download with the fonts inside, so it is a far more immediate and tangible damage for the project overall.

Lastly, my contributions are made with GPLv3, so if you want to switch to a backwards you don't have my permission and have to remove all of them.


PS: I'm not going to discuss this, so if anybody tries to convince me, please don't.  I already rejecting switching licenses in projects far more important and far-reaching than this one.
Title: Re: License Policy
Post by: trapdoor on November 10, 2011, 04:57:12 AM
I am strongly against GPLv3 since it only has negatives, and not a single positive reason to use it. Whilst I would rather the GPL was not used at all (I thought at one point we agreed to switch to MIT like Ogre), GPLv2 is definitely better than GPLv3.

GPLv3 will only hurt the project in the long run, considering its incompatibility with most popular platforms and the fact more and more developers are understandably wanting to avoid it because of the fact you have to release changes even if you don't plan on distributing the changed binary. The only reason for using GPLv3 is for political reasons.

All my contributions are under the BSD license which is compatible with the GPL, so none would need to be rewritten (another advantage of a less restrictive license).
Title: Re: License Policy
Post by: dezGusty on November 10, 2011, 08:33:40 AM
Guys, if license switching is ever going to happen, it will require the (written) permission of each contributor. If someone will no longer be part of the project, we'll need to track him down and ask for permission. One of the things that would help would be multi-licensing... but that would complicate the way the headers would look, as currently we have the license in each source file. I'm not sure whether we can say we're using this main license (GPLv2/3), but we also recommend licensing under MIT.

As a matter of personal taste, I would like to have the game released under a very permissive license, like MIT. If someone wants to take the game, add a multi-player match-making server and charge money for it, I have no issue with them using my code. But others may have...

Edited: like seen in the posts above; others may have political reasons for not releasing their code under a different license. This would mean that no parts of the game can be reused by anyone else, if someone doesn't give his permission. Because the code has little mentioning of the author of each file/class/function. I would suggest updating the code comments to also include the authors of various sections. (doxy: \author or @author)
Title: Re: License Policy
Post by: Kulik on November 10, 2011, 10:14:54 AM
MIT doesn't make any sense for this project. It would basically allow anyone to close it down, rename it and sell it as his own (yeah, attribution and all that but still).

GPLvX is the only sane choice IMO.

I find mafm's post quite offensive TBH. I understand his motives but it still leaves a bitter taste in my mouth. IMO the logical path is to go GPLv2+ if you want to deploy somewhere with DRM and GPLv3 otherwise.
Title: Re: License Policy
Post by: Lastmerlin on November 10, 2011, 10:26:42 AM
Yeah I love this topic ;)

As soon as I have finally understood the differences between GPL, CC-*, BSD etc, you start to fight about version numbers - yeha x(
You are surely right that this is an important issue, but  apparently I am too naive and idealistic to get a grasp of it.

Can you please just a present a list of the options and concrete implications at the end ? All I can say right now is that any option which has *Important contributor X leaves and we have to revert all his work* on its list has a rather low chance to be chosen. However, I hope that discussion wont go this direction anyway - this is already the highest level of rhetorical escalation and therefore not really helpful. One other thing: Having a wild mix of licenses, in worst case per code line seems a very bad idea to me as well.

Title: Re: License Policy
Post by: dezGusty on November 10, 2011, 10:42:41 AM
Keeping the author written in the code is simply meant to know which parts would need to be rewritten in case someone is no longer in the project, he did not leave the option to relicense (or is not reachable) and a license switch is decided by the remaining team.

Having a multi-license means that the used licenses should still be clear and applicable for all the code. So if we add 2 licenses, we use the exact same 2 licenses in all files, allowing the users to redistribute the code under one of the 2 licenses (whichever they preffer).
Title: Re: License Policy
Post by: Kulik on November 10, 2011, 11:56:49 AM
Lastmerlin: Basically GPLv3 prevents sumwars of ever being on platforms like iOS (ipad, ...), steam, ... (all platforms with inbuilt DRM). You potentially lose a lot of users. GPLv2+ is effectively a multi license (GPLv2, GPLv3, ...). But mafm has a point: GPLv3 prevents Tivoization and various other exploits and loopholes of GPLv2. I would strongly discourage MIT, BSD and other such licenses, they avoid rebranding/exploitation for profit and don't enforce others to contribute back. Multi licensing would only make matters far more complicated (IMO).

Since all contributors contributed code as GPLv2+ previously you only have to contact people who contributed in the time between the GPLv2+->GPLv3 change and now. Shouldn't be that hard. In the end I don't see why someone would make such a strong stand and blackmail the other developers over GPLv2+ vs GPLv3. I hope mafm will reconsider...
Title: Re: License Policy
Post by: mafm on November 12, 2011, 03:49:23 AM
I am strongly against GPLv3 since it only has negatives, and not a single positive reason to use it.

This is false, as me and other people have said in other messages.  Whether you pretend to be that the differences are only negatives and not positives is your point of view, not an universal truth.

In fact, if one likes the original GPL (and the side that GPL chooses: users' freedom), GPLv3 has several objective advantages, since it goes further into trying to make sure to protect user freedom, even by sophisticate ways that some people/companies use to subvert it (the aforementioned TiVO, or software patents).

GPLv3 will only hurt the project in the long run, considering its incompatibility with most popular platforms

If this was true, you'd think that Battle of Wesnoth for iPhones would be doing very well, right?  Maybe not so ([1] and [2]).  It seems that there are no releases for more than 1 year and a half, and in fact it lived actively for less than 6 months, apparently.

[1] http://forums.wesnoth.org/viewtopic.php?f=6&t=35395&p=511806&hilit=iphone#p511806
[2] http://wesnoth.repositoryhosting.com/trac/wesnoth_wesnoth/

Now, I don't think bad about Sumwars, but I would contend that Wesnoth is more close to what people considers complete and polished game, with many years of experince and a strong community of users and developers.  And they have a couple more contributors/workforce than Sumwars, also [ http://wiki.wesnoth.org/Credits ].  If they [seem to have] have abandoned the iPhone path, I would be more cautious than you when saying that by using GPL license (or GPLv3) in particular Sumwars is getting hurt and missing many opportunities.

And it is well known that with this move to try to get Wesnoth into iphones, they created a fracture in the community [ http://lwn.net/Articles/396535/ ], so overall was probably more harmful than good.

So sorry if I offend you again, but this assertion of yours is quite stupid and not substantiated in anything other than your prejudice towards the GPL license.

[...] and the fact more and more developers are understandably wanting to avoid it because of the fact you have to release changes even if you don't plan on distributing the changed binary.

This is simply false.  You can modify GPL software and not publish it, as long as you don't distribute it, it's one of the basic facts being there since the beginning:
http://www.gnu.org/licenses/gpl-faq.html#GPLRequireSourcePostedPublic

I thought that you would be better informed about this issue... the fact that you don't understand this vital part of the license while you say to oppose it fiercely, it's quite amusing.

The only reason for using GPLv3 is for political reasons.

All my contributions are under the BSD license which is compatible with the GPL, so none would need to be rewritten (another advantage of a less restrictive license).

I would expect that 2000+ years after the greek philosophers, we would be already be settled in the notion that everything that humans do is political, in one way or another [ http://en.wikipedia.org/wiki/Z?on_politikon ].

Your choice to not GPL is also quite political, too.  Saying that BSD license is "less restrictive" in the sense of not restricting what developers or companies can do with the product; but permitting them to kill the product and thus restricting freedoms and options in users.

And apart from that, many times products with "more restrictive" licenses win, like Linux with GPL and BSD with their license, as explained many times by Linus Torvalds linking the success of it's pet project because of (not in spite of) being GPL and not any other license.  In fact, the whole history of failed Unix systems before Linux (and GNU) is a mixed history of "too permissive" licenses and companies short-sightingly wanting to make big bucks quick, and killing the product early.


Executive summary: If you (plural: trapdoor or others) want BSD, that's your choice, but don't pretend to treat people as fools and put the burden on people who did an informed choice when choosing to contribute to this project -- like I did.  Had this project be licensed under BSD I wouldn't have contributed to it, and if it would be GPLv2 only (and not GPLv3), maybe not either.  And if you are so concerned about the badness of the GPLv3 license, put your money where your mouth is and rewrite my contributions.  They are not that many, really.

But this said, I am quite pissed off that I've been the only one who, after waiting for a long time to see if somebody else did, finally got to implemente installing with "make install" in a more-or-less sensible way (even if some of you are still unsatisfied), including conditional compilation of enet/tinyxml; and I put very serious effort in trying to widen the channel of distribution of Sumwars by creating Debian/Ubuntu packages.  And lastly, it's seems that I am the only one who takes seriously the problem of licensing (instead of just spitting prejudices in a forum) when it comes to real and current issues, like with the non-commercially-distributable fonts which could potentially get Sumwars into some trouble -- that everybody ignored until now, and continues to ignore the last bit (Commonwealth font).

So that's it, I'm quite tired of all of this.  Get serious, folks.
Title: Re: License Policy
Post by: Lastmerlin on November 14, 2011, 05:08:14 PM
 Well it seems that this topic leads conflicts once again :(

First of all, it seems that you do not feel appropriately valued for the work you did. If this is the case, its my fault. I always said that the should somehow get into these repos for Ubuntu and other popular distros to get a big boost in attention. Therefore, your work was a decisive step into the right direction. I should have expressed this more directly - sadly I always missed you in IRC for instance due to my own limited activity lately. And these complaints should not be taken personal - this just happens if you force any developers to change their workflow.

However, you must accept, that not everyone is proficient with these licensing issues. This is normal and acceptable. I am not good at graphics either (and I wont learn it properly if I tried it due to lack of talent) and I still realize that graphics are important. So if there is some decision in this department, I rely on qualified opinions of those who know better. This is not ignorance - this is just realizing you own strengths and weaknesses, delegating and aquring filtered information.

This works well for most topics - except licensing x(

On the topic: Who threw BSD into the discussion ? I believe that I informed myself about this in the past and rejected it because of any grave disadvantages. But I have already forgot again what the problem was. However, discussion about GPL V2 <-> V3 should be enough, right ?
Apparently both have their issues as I see it. So what I want to ask is: Can we please make a sensible discussion about the risks instead of discussing about general ideals and philosophy and prejudices ? Its not that I have no ideals or philosophical thoughts at all - this is exactly what prevents me from understanding these license issues properly. But here there is no point discussion which license (or company or world view) is *good* and which is *evil*. Here - we should consider what works best for this project.

So the question is: What is the complete list of issues with GPLv+2 (it always seems that only two examples were named but the list might be longer)? How grave and realistic are these issues ? Is there really a realistic chance to get the program running on one of these Apple products (Ipad...)? I somehow doubt it considering our demands in graphics etc. .

By the way: The example of Wesnoth is really disturbing. Wesnoth is always what I consider as the perfect example of open source game developement. Of course they are vastly superior considering completeness, professionality, organization, amount of content, contributors and so on. In fact they state they have is what I dream of achieving one day. Its alarming if they have massive issues with an IPhone port.
Title: Re: License Policy
Post by: mafm on November 17, 2011, 01:04:45 AM
First of all, it seems that you do not feel appropriately valued for the work you did.  [...] I always said that the should somehow get into these repos for Ubuntu and other popular distros to get a big boost in attention. Therefore, your work was a decisive step into the right direction.

Actually, it's not something so straightforward, the question about how I feel comes from different aspects:

1) I somewhat have the impression that I was pissing off people by changing old workflows and what not, when creating a "make install" target and trying to get everything working properly for multi-user systems (personal settings vs system-wide data, etc).  Despite the complaints and some people volunteering to fix things in a better way, I think that nobody made any serious improvement to my implementation, yet.  But, at any rate, I did that after waiting for about ~1 year to see if anybody got the strength to do it, because I had told you that I would try to do it.

I was not waiting sitting with my arms crossed, and overtook Ogre package mantainance in Debian (and Ubuntu) to bring it up to date, and prodded CEGUI's maintainer to a point where he got a bit angry with me, with the main purpose of getting Sumwars into Debian/Ubuntu.  And I got a very difficult year, involving defending thesis, holidays for 1 month in a country without internet, unemployed and new job in a new country, and some personal nasty problems; but that's another story.

2) AFAIK only me cared enough to warn you about serious licensing issues (like with the fonts).  Are you familiar with the running story that Microsoft gets more revenue from Android phones than Windows phones?  I don't know if that's actually true, but it's not beyond possibility that, if you wanted to sell Sumwars as iphone app and got enough popularity/money, they would be suing your ass and could have to pay them royalties for using the fonts, or libraries, or what not.

So it's not that I feel that everybody should bow at me when I join the room.  But having people babbling non-sense about GPL being harmful for the project when they didn't even read the license, and are not doing anything at all to remove the more serious licensing issues of the remaining Commonwealth font (which I brought several times to attention, and still at this point nobody removed) nor did anything previously to remove other really harmful things, is not funny.

On the topic: Who threw BSD into the discussion ? I believe that I informed myself about this in the past and rejected it because of any grave disadvantages.  But I have already forgot again what the problem was.

BSD or GPL?
http://www.matusiak.eu/numerodix/blog/index.php/2007/12/15/gpl-vs-bsd-a-matter-of-sustainability/
  (read fully or not at all, you can get the wrong impression if only reading half of it)

Basically, it boils down to if you want to ensure that everybody (including players) should have access to project that you develop remains freely available forever (GPL); or if you want to give developers who want to use your code the power to use it in any way that they please, even if that involves creating a closed-source project from that, and never publish the code (BSD).

However, discussion about GPL V2 <-> V3 should be enough, right ? Apparently both have their issues as I see it. So what I want to ask is: Can we please make a sensible discussion about the risks instead of discussing about general ideals and philosophy and prejudices ?

It's difficult to discuss licensing without discussing ideals and philosophy.  Because, if you are motivated enough to create a vast software project as you did, for sure you all did it for a reason, with a motivation, with a general ideal, a goal.  And, ultimately, licenses are implementations (in law) of the ideals and goals that you have with your new baby-project.

So you should choose the driving ideal, goal for the project first, and then choose the license that you think that suits that ideal/goal better.

Like, if you prefer that your game becomes freely available at all times in the future for everybody, to allow others to continue your effort when you are not interested (or create alternate games), but always having players enjoying your game freely.  Or if you prefer to get to as many users as possible for your game as a whole or only parts, even if that means risking a company using your code and dialogs for a different game in the next XWii PS-Box console, without you getting a dime nor your users being able to exercise the rights that they had when using your code directly.

So the question is: What is the complete list of issues with GPLv+2 (it always seems that only two examples were named but the list might be longer)? How grave and realistic are these issues ?

Reasons to move to GPLv3: http://gplv3.fsf.org/rms-why.html (http://gplv3.fsf.org/rms-why.html) and http://www.gnu.org/licenses/quick-guide-gplv3.html (http://www.gnu.org/licenses/quick-guide-gplv3.html)

Probably not much that affects a project like Sumwars.  "Tivoization" is not a problem for a project Sumwars.  But it will allow companies to not have to pay for Sumwars to Microsoft (see Novell-Microsoft deal in the article).  Or, if distributed on a DRM-enabled platform, with GPLv3 you would be giving your users the freedom to remove DRM restrictions (which is why they don't).

So it depends on your opinion of the matters, I consider this a feature.  I don't want to allow any company to take profit from my project to turn it against my users and take rights away from them.  If they are not going to maintain the rights that I want for my users, I rather not having the company dealing with my project at all.

Apart from that, some company could sue Sumwars as developers, for infringing their patents, even if they distribute it, which is quite crazy.

To remove all of this non-sense, and try to fight agains software patents (http://lxnews.org/2011/05/10/shuttleworth-prefers-gplv3/), GPLv3 was created.

Also, a project with GPLv3 is compatible with Apache 2.0-licensed stuff, which at the moment you're not using it, AFAIK.  But apparently, GPLv2 and Apache 2.0 are incompatible.

So, as a summary, there's no reason practical reason to downgrade your license to GPLv2, other than if you want to allow some of the described practices trying to sneak around the protections that you give to the users of the software.

Is there really a realistic chance to get the program running on one of these Apple products (Ipad...)? I somehow doubt it considering our demands in graphics etc. .

Those topics don't concern me... if you want to try, go ahead.  I just won't change the license of my contributions to allow you to reach that goal, if that's what you, as a project, desire.

I very rarely (or never, probably) get involved and contribute with projects with licenses allowing that, or with exceptions to let the project run in essentially control-freak I'll-sue-your-ass-and-forbid-porn-and-WiFi (http://news.cnet.com/8301-13579_3-10464021-37.html)-and-what-not platforms like Apple's.

Its alarming if [Wesnoth] have massive issues with an IPhone port.

Don't take my word for granted, I only clicked in a couple of links to get to the status of the iPhone port, I didn't follow closely and I don't know what's the reason that they abandoned the port, or if there are similar efforts to port to other places.

It's only that this discussion brough to my memory the licensing debates of them, and how it harmed the community, and how ultimately the non-clear licensing and wanting to publish in these AppStuff platform harmed them more in the long term than anything else.
Title: Re: License Policy
Post by: dezGusty on November 17, 2011, 09:31:11 AM
Here's the official page from GNU as to the possible reasons to upgrade GPL from v2 to v3 http://www.gnu.org/licenses/rms-why-gplv3.html (http://www.gnu.org/licenses/rms-why-gplv3.html)

It basically says that if you go with GPL, go with the latest, because it's better protecting against patents. (Although I haven't heard of any open source games having a problem with this. And Wesnoth, cited by Lastmerlin has GPLv2 http://wiki.wesnoth.org/Wesnoth:Copyrights (http://wiki.wesnoth.org/Wesnoth:Copyrights)). As to anti-DRM... not sure that helps or hurts in any way. Meaning: if we ever add a multiplayer dedicated server with user/password and some encryption for securing identities, will GPLv3 affect this in any way? I don't want to bother with this yet; I just hope we get to progress far enough that we have this problem one day :-)

EDIT: mafm: please write shorter entries :)
Title: Re: License Policy
Post by: artisticdude on November 17, 2011, 05:01:37 PM
Quote
"Its alarming if [Wesnoth] have massive issues with an IPhone port."

Don't take my word for granted, I only clicked in a couple of links to get to the status of the iPhone port, I didn't follow closely and I don't know what's the reason that they abandoned the port, or if there are similar efforts to port to other places.

It's only that this discussion brough to my memory the licensing debates of them, and how it harmed the community, and how ultimately the non-clear licensing and wanting to publish in these AppStuff platform harmed them more in the long term than anything else.
Actually, what killed the Wesnoth iOS port was that the guy who ported the project to iOS jumped ship and got involved with another project. Since the iOS port is under his iOS dev account, only he has access to it, which means that we have no control over the port at all (although the Wesnoth project still gets a certain percentage of whatever revenue the guy takes in from sales of the port), and efforts to contact the porter have been futile. There was some talk of licensing issues and such, but it was the porter abandoning the project that doomed the iOS port of Wesnoth, not the licensing of the port itself. Now we've got an Android port (done by an official Wesnoth developer), although since I personally don't own an Android device and there has been minimal discussion of the port on the Wesnoth forums, I can't say how that's been going.

But anyway, I'm just here to say that whatever license you want to put my contributions under, I'm fine with it. Like, literally whatever license.

Also, I'm no expert on licensing and legalese, but Wesnoth is licensed under the GPL v2+, which allows for later revisions of the license (including the GPL v3). Would something like that be feasible?
Title: Re: License Policy
Post by: Kulik on November 19, 2011, 12:48:27 PM
This all boils down to philosophy. I believe GPLv3 is the way if you are not going to be distributing the game on DRM platforms. If more developers decide to go GPLv3 the distributors will be forced to support GPLv3 software as well. GPLv3 definitely offers most freedom to the user and that's what matters IMO.

GPLv2+ as I suggested would be a compromise.

GPLv2 and GPLv3 are separate licenses, GPLv2+ means GPLv2 and all future versions (so GPLv2 and GPLv3).

GPLv2+:
+ proven (Linux et al)
+ can be distributed on DRM platforms
- no patent protection
- TiVoization possible

GPLv3+:
+ most user freedom
+ patent protection
+ TiVoization impossible
- de facto forbidden on many DRM platforms (which might be an advantage actually)

BSD/MIT/...:
+ most freedom for developers
- possibility of exploitation (someone closing the game down, rebranding and selling without contributing back)

All in all it's a philosophy battle. In my honest opinion GPLv3 is philosophically the best but limits the project in some ways (which might or might not matter). To get the discussion closer to conclusion, why don't you discuss how would you want sumwars to be distributed and whether GPLv3 poses a problem there?

Apparently you can't even distribute GPLv2 software on apple store: http://www.fsf.org/news/2010-05-app-store-compliance http://www.zdnet.com/blog/hardware/should-apple-ban-gpl-from-the-app-store/11484
Maybe that changed in the meantime though.

I am changing my vote to GPLv3+.
Title: Re: License Policy
Post by: fusion44 on November 19, 2011, 04:27:44 PM
I vote for GPLV3. My guess is that GPL will be banned some time in the future anyway, like Microsoft did already for WP7 store...
Thechnically we can't even distribute it on Android Market because there is no direct way of getting the source...

http://blogs.computerworld.com/17279/gplv2_blocks_vlc_from_apples_app_store