BSD License for future LotGD Development

[CavemanJoe]

Oh, forgot about the license, which is best... No idea, to fully understand any of them you probably need a high IQ or a degree in law. I've got neither.

I do like the following.
1) You can't make changes to the core files and keep them to yourself. You must share any changes.
2) You can't remove the copyright info.
3) Module author's could sell a module more than once.
4) ...

Okay, as I understand the BSD license, (1) wouldn't apply, but (2) and (3) would.

[CavemanJoe]

What will be seeing though in terms of core releases..? Two different core engines and the in-between forked version?

Well, I really don't know.  But what I'd like to see happen is DragonPrime taking the bits of DragonScales that would be useful to them - the performance improvements, integrated inventory system and improved chat engine to begin with, and then perhaps later on the Stamina system, integrated map, World Community Grid integration, and so on and so forth.

I'd say take it all at once, but the thing is, Improbable Island is already a vastly different game just because of the Stamina system replacing Turns.  The next iteration of the Island (based on DragonBones) won't even have levels (beyond what's supplied by the Stamina system), BluSpring's, or even a Dragon.  Going from Turns to Stamina overnight would be very hard for most admins to deal with, so I think we should work our way up to that one with some smaller changes first.  But the core-baked Inventory system and performance improvements would be first on the list.

EDIT: That is to say, yes, DragonScales and Legend of the Green Dragon would be very much in line and compatible with each other right up until DragonBones is released.

Once DragonBones is out and ready, if LotGD is BSD-based then we could even recreate LotGD (in modular form) from inside DragonBones.  DragonBones is, after all, intended to be an engine inside which you can build any role-playing game that uses an inventory, a map, and some basic skill sets.

EDIT AGAIN:
Actually, we could always put in a game setting in the early DragonScales releases to decide whether to use Stamina or Turns - that'd be a good way of easing the transition.

[KaosKaizer]

I have been thinking a bit before actually posting. If the BSD license would actually be better for customization and forks, then I'd say let's go for it. I mean, it would make changing things a lot easier as far as licensing goes, and as far as big companies ripping LotGD off, they would have already done it if they thought they could make some money off of it. I mean, look at how long it has been out.

However, for admins wanting to fork like CMJ with DragonScales/DragonBones, I can see where it would be a lot easier to do what he wanted to do with a lot less time.

Well, I really don't know.  But what I'd like to see happen is DragonPrime taking the bits of DragonScales that would be useful to them - the performance improvements, integrated inventory system and improved chat engine to begin with, and then perhaps later on the Stamina system, integrated map, World Community Grid integration, and so on and so forth.

I'd say take it all at once, but the thing is, Improbable Island is already a vastly different game just because of the Stamina system replacing Turns.  The next iteration of the Island (based on DragonBones) won't even have levels (beyond what's supplied by the Stamina system), BluSpring's, or even a Dragon.  Going from Turns to Stamina overnight would be very hard for most admins to deal with, so I think we should work our way up to that one with some smaller changes first.  But the core-baked Inventory system and performance improvements would be first on the list.

EDIT: That is to say, yes, DragonScales and Legend of the Green Dragon would be very much in line and compatible with each other right up until DragonBones is released.

Once DragonBones is out and ready, if LotGD is BSD-based then we could even recreate LotGD (in modular form) from inside DragonBones.  DragonBones is, after all, intended to be an engine inside which you can build any role-playing game that uses an inventory, a map, and some basic skill sets.

EDIT AGAIN:
Actually, we could always put in a game setting in the early DragonScales releases to decide whether to use Stamina or Turns - that'd be a good way of easing the transition.

CMJ, I do like your idea of DragonBones, and I would like to see it out. I mean, there are some things in the core that a lot of admins would rather not, or even don't, use at all. Some of these things are things that could easily be done via module. I mean, the clans system. I know I like it, but some admins don't. So why even have the code in there at all if they don't want to use it?

And the battle system. I know that some people would like to have their own ways of doing battle. This would enhance it, too.

So my vote is this: Switch to BSD, and come out with DragonBones. If people still want to continue with the development of the regular LotGD, then that is all fine and well. But for those that don't, DragonBones would be an option for them. It would make everything easier, in my opinion. It might also bring in more players, as well.

:: EDIT ::

EDIT AGAIN:
Actually, we could always put in a game setting in the early DragonScales releases to decide whether to use Stamina or Turns - that'd be a good way of easing the transition.

I'd say to transition players and admins over, this would be a fair way to go. I personally have never used the Stamina system. I think the whole idea would be great, but I have not gotten around to using it. So, yes, an option for DragonScales for easy transition to DragonBones would be good.

[Paradox_pm]

I approve the BSD License. Any news?

[CavemanJoe]

As far as I'm aware, we seem to have a consensus, but without word from Upstairs we're still operating on Creative Commons until further notice.

[Paradox_pm]

Great.
Caveman, is your fork released under the BSD license?

[CavemanJoe]

Nope - don't have permission to do so, yet, and there's probably still too much LotGD code in there for me to properly call it a new release.

[Paradox_pm]

So we're waiting for MightyE's word, right?
You see, my LoGD installation is heavily customized and translated, so I'm stuck with 0.9.7.
I tried, but there's actually no way to translate it the way I want it to be without editing the core files.

[Afkamm]

You're talking about not being able to translate v1.1.(0/1/2) the way you want without editing core files?

[Paradox_pm]

Exactly. However, it is not just a plain translation, it's a conversion.
Even a simple translation would be unfeasible without editing the core files.

[Boris735]

So we're waiting for MightyE's word, right?
You see, my LoGD installation is heavily customized and translated, so I'm stuck with 0.9.7.
I tried, but there's actually no way to translate it the way I want it to be without editing the core files.

You do realise, I hope, that there is nothing preventing you modifying core files on a 1.1 installation.  You may have difficulties upgrading to a higher version, in that case, but you're already in that situation.

Now, it would mean that you were bound by the license to make available to anyone who asks any modified core files, or modules that depend on those.  Maybe that is the sticking point for you... I just make this note in case you were somehow under the impression that the license prevented you from editing core files; it does not.

[Paradox_pm]

Exactly.
While the BSD would let me choose whether to release my modules or not, the CC (in my case) does not.

It's a matter of choice.
Thank you for your explanation, though.

[MightyE]

So this topic does come up periodically.  Let me summarize CMJ's position since he's the lead for this movement:

The existing license has no weight because anyone sufficiently large could choose to violate the license, and there's nothing you can do because the legal system is broken.  The license only really applies to people without lawyers on retainer.

So forgive me the hyperbole Joe, I don't really intend to make fun of this line of reasoning because I do believe it has some merit, but I imagine this being said in a haze-filled room with a group of friends sitting in a circle: "The law isn't like the law, man, MONEY is the law."  "Whoa, deep."  Difficulty with enforcing licensing isn't really a reason to just give up on it.

The consequences of a BSD license Joe mentions are concerns of mine, but philosophically I believe in sharing with those who have shared with you.  I also believe there are enough selfish people in the world who would take without giving if they believe they can get away with it that I feel it creates an undue burden. 

The biggest threat I perceive to lack of code sharing is a gangrenous community - one where people consume time & resources of the sharing developers (who hasn't spent time supporting their code no matter how good or well documented it is?) while those developers get little to nothing in return.  It's a parasitic relationship, and as long as sharing is in one direction only, it means that the core could never be as compelling as people's private implementations since the private implementation can have the best of both sides, while the core can only have its own code.

So Paradox_pm provides a good example:

You see, my LoGD installation is heavily customized and translated, so I'm stuck with 0.9.7.
I tried, but there's actually no way to translate it the way I want it to be without editing the core files.

The implied concern is that she'd like to upgrade to the post-GPL versions of the game, but doing so requires a release of her source code since newer versions have a license with greater teeth than the GPL.  She would like to upgrade, but would like to keep her code changes to herself.

Just to be clear on this front, the reason I chose GPL for earlier versions of the game was not to provide a loophole where people could keep their code private, but rather I chose this license out of naiveté, not realizing it failed to provide the protections I expected it to.  If I had my way, there never would have been a version of the code where you could keep your edits private, it was an oversight on my part that let you get this far.

I know that sounds overly harsh, so I want to step this back a notch.  It is also not my desire that all installs of the LoGD codebase be cookie-cutter replicas of each other.  I love the idea that people can differentiate their worlds, and I have secretly periodically created an account on other people's games just to see how different things might be.  I love that diversity.  Being as familiar as I was with the codebase, the game held no surprises for me, and had long since stopped being enjoyable to play; being on someone else's unique world would provide surprises, and that was a lot of fun! 

I would love to make a license where you can customize the game world but still require you to share functional changes back.  Unfortunately the current architecture does not support this idea.  Making modules exempt from the "SA" clause of the license was an attempt do to this, but the division between module code and core code isn't sufficient for a truly great experience, only a wholly new codebase would be able to separate the game world from the game engine adequately.

Something I haven't told anyone but those fairly close to me is that I worked for a while on a "LoGD 2.0" game with this design in mind.  It was going to be a 100% rewrite, and was going to take the form of a game engine (not called "LoGD," but whose name had not yet been decided), plus a game implemented with this engine (called "LoGD").  Sidenote: ultimately I decided my architectural approach was wrong, and I scrapped and started again, but my second attempt is so skeletal that it's really only maybe a dozen files overall.

TLDR: So that's a lot of rambling, allow me to summarize.  I want people to be able to create unique games, but I want them to share back with the community as they do.  These two desires are in significant conflict with each other under the current codebase.  I don't see a way around that without a rewrite.  I'm also no longer anywhere near as invested in the development of LOGD except at an emotional level.

At this point, my sense is that despite the disadvantages of a switch to BSD, the advantages are greater.

Because that seems to be the will of the community, I am inclined to agree.  I will ask JT to weigh in here, and if he does as well, it'd be nice to hear suggestions from the current core maintainers on how to make that transition.

[Paradox_pm]

Thank you very much for sharing your thoughts, MightyE.
As I said before, it's a matter of choice. In my position, I'd be forced to share ALL of my modules and changes.
I'd just want to be able to choose the modules I'd like to share and the ones that I'd rather not.

I also understand your concern about "leeches".
However, it is my understanding that in GPL-driven communities (say, Linux), sharing, which is the base of the code evolution, is still of the utmost importance and relevance.
I guess "leeches" are around here even now, just predating the modules our talented coders write and not contributing back. I think the solution you suggested is the way to go: create a "skeleton", a simple game engine (and that's what I'm using 0.9.7 now, as a plain and simple game engine, nothing else) released under GPL or BSD (you name it) and, maybe in parallel, continuing with the CC LoGD experience. In the meantime, we could also release the current "full" LoGD core under BSD, however stating that it's BSD only as a "game engine" and not as a "ready-to-play" game. 

MightyE, thank you for listening to the community, I personally appreciate your interest.

[MightyE]

In my position, I'd be forced to share ALL of my modules and changes.

You realize, of course, that I shared all of mine within the same rules I set up for everyone else, right?  Being the copyright owner, I wouldn't have had to, but I drank my own medicine and abode by the license as though I was any other developer. 

I recommend not saying, "I'd be forced to share," as though it is a weak aspect of the current license.  This is the primary reason I chose that license, and it is the GPL's failure to attach this clause for PHP scripts (like it does for compiled languages) which caused me to switch.  I think this argument weakens your case rather than strengthens it.

It is only several years of detachment, and the very sad realization that I will probably never be able to be a part of a project like this again, which has me entertaining this idea at all.  If I had the ability to remain active in the community, I most likely would reject the idea outright.