Dev meeting minutes 2010-05-23

[Argh]

I would also like to point out that AI != engine. We can use our own licenses, and with the exemption gone, we need only make sure they are GPL compatible.

Thus my point still stands. AIs use a different licensing model, precisely because they are their own separate entities, separate code base, separate projects.

It also appears Argh has yet to even look at Shard as demonstrated by his lack of knowledge, nor has he thought of the consequences of what he says ( If you follow the arguement through, you reach the conclusion that Internet explorer violates the linux kernel GPL license because it links with the linux code via networking should any linux users or server administrators decide to log onto the internet)

<sigh, again>

NO, no, no and again, NO.

There is no "separate license". Quit allowing aegis's arguments, which are largely based on suppositions about things that do not actually exist, to draw you into a totally illegal position.

There is no "it is not part of the engine". My Lua scripts aren't "part of the engine", but are GPL, because that is required, based on our understanding of the law.

There is no, "I have not considered the consequences", either. I darn well know exactly what I am saying here, and have been quite nice about allowing for a certain period to take corrective actions on your part, and have proposed a solution that would allow you to do what you need to do (gather data) whilst not having legal problems. Quit shooting the messenger here.

Lastly, there is no exemption.

Nowhere in the post cited does SJ say, "there is an exemption". If you can find evidence to the contrary, by all means do so... but in fact, SJ's posts there indicated very strongly that not only is there no exemption, but that he was opposed to the concept.

The GPL explicitly states that things dynamically linking (like dlls) to GPLed code must also be released under the GPL licence (otherwise you use the LGPL or some other variant). I guess its technically Spring linking to the AI dlls in this case though so im not sure how that is interpreted. In addition to that though the AIs reuse some spring files (the interface ones) which are GPLed. So I would say that yes you have to provide source code to anyone that asks for it and you have distributed an binary to.

But you are still using the .h files that I created (and you have modified) and released under the GPL. And the GPL only allows you to use GPLed files in a project if you release the whole project under GPL. Probably you could create some reverse enginered .h files I suppose (well not you since you have obviously seen the original files and thus cant do a clean reverse enginering but someone that has never looked on the spring source).

In short, there's absolutely no foundation for that argument, and it's not supportable by the current license at all in any case.

In the meantime, should something that compiles and behaves vaguely like an AI show up in July that doesn't have patented technology in it, who's to know?

Stop making this a showdown, AF. We're mainly sympathetic to your plight, despite you being silly about it

[Das Bruce]

GPL is YTR, why didn't the license change last time this came around?

[Argh]

GPL is YTR, why didn't the license change last time this came around?

Because although the vast majority of us agree with that, it only takes one obstructionist to prevent changes... and the Devs apparently didn't keep good records when it came to contributors, so we have no legal way to do so even if everybody could be convinced

[==Troy==]

A simple way to understand the point of GPL is:

You binaries have no worth.

The only way you can pay back to a free project you are using, is to give back the code.



In the AF case, given the above statements, nothing is being given back to the project. He gets his results, he gets his degree (which is worth something) he gets the reviews, and a testing platform, where he can modify whatever aspects of the game he likes, from ground up.

But.. he does not give anything worthwhile back to the project. GPL simply means that devs want the project to grow, and not be a base for parasitic developments, where the person in question uses the code to create a binary blob, and assume a complete ownership of it, while exploiting the open project in whichever way he likes.

That is the exact reason why WINE turned to LGPL (Cedega took WINE code under MIT license, and started to sell it as binary blob, while assuming all of the WINE work as their own).

It is unhealthy for a GPL project to allow long-term exemptions, and exploitation (basically what AF does in this case), because it prevents further improvement, and relies on a single developer for all the possible issues that the binary blob might have (look at TASClient, although open source, is still maintained by single developer with all of the consequent issues).


GPL (and not LGPL), in its nature, attempts to be as invasive as possible, to remove exploitation of an open source project, and force an open, and free development. Some of the more selfish developers will be against it, some developers, who prefer the open model, only contribute to such projects, because they know, that their code is NOT going to be exploited.


Although I am personally not a dev of this project, in my opinion (which consequently has little worth), spring developers should allow the unie projects to stay closed source until a strict date is given for the code release (And not too far in the future). Or, at least, advise such developers to use the loophole I proposed (i.e. run the AI code networked without releasing binaries). I am sure,for one, do not want spring to turn into what OGRE is now, where you have barebones crappy engine, with all of the bells and whistles costing a hefty sum.


tl;dr

This is GPL project, you signed up for it, and you have to deal with the license. Devs can close their eyes on some issues, but that does not save you from it, because effectively ANY contributor to the project assumes right to sue the hell out of you, if you do not comply. It can seem damaging to some, but it is also protecting others, as any other copyright license out there, not everyone will get satisfied.

[Tobi]

Forb, can you refrain from posting if you didn't read the thread and/or have nothing useful to contribute to the discussion, please?

Anyway, I'm sorry I didn't censor the meeting minutes.

I'm glad though that some things are cleared up now, like the [lack of] exemption for AI code, and I think it's clear that we should encourage and make AI devs aware of what's expected as soon after their inception as possible, to prevent later disappointment.

I want to say though that in my opinion ultimately a competition between open AIs will result in faster innovation than a competition between closed AIs, as the latter inevitably results in people doing double work (competing by slowing the opponent down), while in the first case any competition is solely on added value (competing by innovating). In the case of a true competition (i.e. with rules, a set start and end date and possibly some kind of prize) we can talk about exceptions during the course of the competition.

However convincing it may seem, I can not accept the history argument that was put forward, as that only tells us how it went in this exact case. It doesn't tell us anything about how many/good AIs there would have been if they had gone open source earlier/later. Besides, back at the time Spring was just barely running on platforms other than Windows, while now a binary distribution of an AI is effectively saying 'fuck you' to all users on other platforms.

Besides this, Argh makes a good point that it's reasonable that if a game's Lua code should be open, AIs should be open too; not in the least because of the gray area in between, of Lua AIs or implementing game/gaia code using the AI interface.

That's why I think we should encourage even more and faster than before that AIs must release code, and I think we should start with that in August 2010.

[Argh]

I'm sorry I didn't censor the meeting minutes.

I'm glad you didn't, frankly. This is precisely the kind of thing that needed public discussion, despite the controversy.

The last thing on Earth we need is for such things to be arrived at in secret, then discovered later. That would cause a lot of distrust of the Devs, who have the same incentives in re: code development as AI developers, in the sense that their work may enhance their careers.

I have always thought that that was a perfectly-fair tradeoff, so long as we're open and honest about it; if jK uses his work elsewhere, for example, it doesn't hurt me or anybody else using the Engine, while we get the full advantage of his time and generosity- we aren't merely lab-rats, and it's not a zero-sum game. That is one of the primary reasons why I decided to stay here in the first place- I didn't really understand the GPL at the time, but knew that it meant that coders couldn't screw everybody over by suddenly deciding to take the engine private or otherwise behaving badly, which has happened all-too-frequently with projects that aren't Open Source.

[AF]

Argh, lua code and AI code under a GPL engine does not have to be GPL. It has to be GPL compatible yes, but it doesnt have to be GPL, and they are indeed separate from the engine, they have their own project leads, their own repositories, their own development paths and decision making structures.

There is nothing stopping you from release as LGPL, or PD, or a handful of the creative commons licenses. For example, there would be nothing stopping an LGPL shard being used in a commercial engine so long as they provided the modifications necessary to Shard. It wouldn't be unless spring is linked into that engine that the GPL license of spring comes into play. Since Shard at the moment natively links into spring, that is a moot scenario.

Whereas this, this is just libel, below the belt, I'm shocked you have the nerve to accuse me of being unreasonable then follow up with this:

You disregard the order from your prof that says "do not release binaries" because you want the results for your career, but you do not disrespect the order "do not release source", because others could steel form you or .. whatever (the only reason you gave does not apply, as it is not based on NTai, as you explained yourself). as was noted by someone earlier, you were specifically against the exception in the past, because back then, it was working against your selfish wants.

You disregard the order from your prof that says "do not release binaries" because you want the results for your career,

I want a good mark for my dissertation. Any student would want that. It was my understanding that releasing binaries would have been okay, especially since I made the point of promising to open source at the first available chance. I thought it would have been fine under the exemption, which has now been shown not to be the case. If I was aware this was not possible I would not have done so, but I was not aware that it was a problem at the time, and nobody kicked up a fuss upon release, or the second or third or fourth releases. I have stated this already in this thread yet you have taken what I have said out of context and made a logical leap.

You have accused me of profiteering and deception. This is both innacurate and disrespectful of you.

but you do not disrespect the order "do not release source", because others could steel form you or .. whatever

Here you accuse me of being two faced and manipulative.

I have already promised to opensource, and the vast majority of the codebase is already distributed in clear text lua, how does this point hold? Theft of code and plagiarism are not something I'm bothered about, I only follow those rules because I agreed to as part of my the requirements of my university when starting my degree. Its having my mark struck off and accused of cheating, not personal pride or ignorance that made me follow the rules.

(the only reason you gave does not apply, as it is not based on NTai, as you explained yourself). as was noted by someone earlier, you were specifically against the exception in the past, because back then, it was working against your selfish wants.

Theft and deception

I have never been against the exemption, I agree that open source development is essential, but I do not agree with enforcing it as the only way of developing. If I really wanted it for my selfish wants, why did I open source NTai when I could have kept it closed source and stolen other peoples code? Nobody would have known! Of course the same accusation of theft could be levelled at Shard, but I have not taken other AIs code, as will be shown when it is released, and just in case I needed to, I sent messages to several AI developers explaining and asking if I could, despite their license, just so they were aware of the possibility, and to make sure there were no issues.

[cranphin]

Hee this is fun ^_^ Well, not entirely I guess, but it's always fun to have a big old GPL discussion, I learn something new every time :D

So, ignoring all the other stuff going on, on the part of GPL compatible licencing:

If a program released under the GPL uses plug-ins, what are the requirements for the licenses of a plug-in?

It depends on how the program invokes its plug-ins. If the program uses fork and exec to invoke plug-ins, then the plug-ins are separate programs, so the license for the main program makes no requirements for them.

If the program dynamically links plug-ins, and they make function calls to each other and share data structures, we believe they form a single program, which must be treated as an extension of both the main program and the plug-ins. This means the plug-ins must be released under the GPL or a GPL-compatible free software license, and that the terms of the GPL must be followed when those plug-ins are distributed.

If the program dynamically links plug-ins, but the communication between them is limited to invoking the ÔÇÿmainÔÇÖ function of the plug-in with some options and waiting for it to return, that is a borderline case.

Pay attention to the part where it says:
- This means the plug-ins must be released under the GPL or a GPL-compatible free software license
Looks good right? :)
- And that the terms of the GPL must be followed when those plug-ins are distributed.
Oh, ah.. I think this means source needs to be distributed. BUT I'm not 100% sure, any lawyers in the room? :)
You'd think a FAQ 'd be clearer :)

Usefull btw. :
http://www.gnu.org/licenses/gpl-faq.html

[hoijui]

(cranphin, you should label this quote, eg: "from GPL V3" )

[Forboding Angel]

Forb, can you refrain from posting if you didn't read the thread and/or have nothing useful to contribute to the discussion, please?

I did read the discussion, I realize that you're holding off on bringing the hammer down (Edit: And by the way I appreciate it, so thank you). What I don't appreciate are the hands of the Gods as it were, flippantly discussing something without taking into account the serious reprecussions it may have on game developers.

Yes I'm aware that circumventing this would be simple, but I don't like doing things like that just out of spite. To quote Michael Weston, "Creating bodies generally creates more problems than it solves".

Anyway, I'm sorry I didn't censor the meeting minutes.

Why would you censor them? That would be counterproductive. I might be pissed off but I do think it's something that needs to be discussed, and it has come to a peaceful resolution. Moreover I agree, however there are extenuating circumstances and simply put, without Shard I am seriously screwed.

The fact that this discussion took place does not upset me at all. What upset me were hoijui and kloot's flippant comments. I can't explain how much that upset me.

[AF]

I think its more a case of the defamation point I raised, and I would have to admit in the same position I would not have censored it, not because I disagree, but because it would not have occurred to me, and its not what I disliked in the whole affair.

It shouldn't be too long before I can push out releases again, I graduate July 12th, and if I dont have my marks soon then Ill have to start asking questions and prodding people to get a move on

[zwzsg]

Spring was not released as open sourced originally, nor was it originally GPL. This came after the release.

I suppose "since its release to the community" wasn't unambiguous.

Spring source was released as GPL several monthes after the first binaries were made public and got played. What do you mean, "since its release to the community"? Because if you mean the open source community, then your statment is tautologic. If you mean the TA fans community, or even the internet community, your statement is wrong.

I am wrong. The first public release of Spring binaries was accompagnied by the source under GPL license: http://springrts.com/phpbb/viewtopic.php?f=2&t=680