Oolite Bulletins

First of all, I reckon I really shouldn't go away for even a few days, it's taking forever to catch up with what's been happening while I was gone!

In other words, apologies if I say something really dumb.

Commander McLane mentioned forking OE as a stand-alone project. Even before he posted that, I've been of the idea that such a fork could represent an excellent win-win. And one that could avoid many licence related problems. Please bear with me for a minute.

The Oolite project is all about reinterpreting Elite, a game entirely focused on the activities of a cobra mk3, a good all-round average-ish ship.
As such, uber ships - and massive fleets - are meant to be the exception rather than the rule.
After all - in Elite - the cobbie is meant to be the best (and only) ship to have, full stop.

OE - amongst other things - presents the player with a much more crowded universe, with many ships more powerful than a cobra almost everywhere. Using OE, having a cobra doesn't offer the same relative strengths as in vanilla Oolite, and most player find plenty of incentives to use different ships than the cobbie.

Quite a lot of open source projects have forked in the past, and many more will do so in the future, for what many people would consider entirely trivial matters. For example, GNUzilla IceCat split from Mozilla Firefox mainly because the firefox logo is trademarked (while still freely distributable, and acceptable to the majority of open source distributions, the logo cannot be altered without infringing Mozilla's trademark).

In my mind, the different emphases in Oolite and OE make them almost two different projects already. Oolite is focused more on the independent trader, while OE pushes the gameplay in different directions, with uber ships, player controlled stations, etc.

Another open source example that springs to mind is the mySQL & Postgres projects. While their codebase is completely different, they are both SQL engines, and they're both doing very well.
Why? Most people seem to agree that's because they are focusing on different SQL engine needs.
While mySQL is relatively lightweight, and used for smaller projects, it doesn't scale as well as Postgres, making Postgres the better choice for big databases.

A lot of the not-so-Riedquat-friendly comments might well be a result of the contrasting gaming styles: if you want something as streamlined as possible, you don't want to add stuff you're only going to use once in a very long while. By the same token, if you want to have all possibilities open to you, missing out on 1 out of - say - 250 oxps is already less than desirable, never mind missing out 20 or even 50 oxps.

With that in mind, I do think that OE as a stand alone project - including its own exe - would make a lot of sense.
As well as having an executable with all the 'experimental' features that Oolite definitely won't include this side of the MNSR, a future stand alone OE project can have all the properly licenced OE.oxp stuff included in Resources, making a lot of ships part of the standard OE distribution.
This leaves out the oxps for which the author never wrote an explicit licence, or that contain a no-derivatives clause. This second bunch of oxps could be all placed in a second installer that would pre-populate the AddOns directory with all the separate no-derivatives oxps!
That way OE would have two downloads (OE base + OE NDs), wich - while not ideal - compares pretty well with the original OSE.oxp idea (1 download to get Oolite, + 1 for OSE.oxp).

Lestradae's 2 years work won't be in vain, and people who either can't run or don't need all the extra ships & features as standard can still quite happily download the original Oolite instead of OE, and just add the oxps they're happy with.

In hindsight, the whole Sung saga might well have been a misunderstanding of monumental proportions, considering that in the end he's got no problems with his Oolite textures being available to the Oolite community: if I'm not too mistaken, Lestradae has been hosting them on his own box.net space on behalf of Sung for the last 6 months+, as well as including them inside OE.

If we simply agree to disagree, and a fork is a pretty official way of saying so, then we shouldn't really have any problems in the future if OE develops the way the OE team feels is best. If a fork happens, I personally don't see any problem with the idea of the OE project & Oolite.org remaining in friendly terms, & having links to each others' forums, etc...

Anyway, this is my personal opinion. I really don't quite know what your opinions are (by you I mean other devs, oxp makers, forum admins & members) but I really would like to know if what I said above does make some sense.

Lestradae, your opinion on what I just said is particularly welcome!

Joy. Yet another topic about this.

Sorry if that sounds harsh - is that three separate threads about this now, including another two or three about the licensing issues that spin off from it?



When will it end?

Selezen wrote:

Sorry if that sounds harsh ... When will it end?

Selezen ... I don't want to insult you, seriously not, but are you being naive or are you somehow trying to make some black humoured fun of me?

It won't end. The question is how exactly it will continue, without making the non-OE-interested people unhappy. With the possible exemption of hypothetical people who will only be happy if the OE ship has been sunk. Which won't happen.

So either you don't care at all if my two years of work are going to be binned and that's it, in which case perhaps don't mention the words "respect", "moral" and "feelings" ever again in your favour, or you do stop asking "for the end" of OE.

The licensing discussion is over, OE not.

Kaks wrote:

If we simply agree to disagree, and a fork is a pretty official way of saying so, then we shouldn't really have any problems in the future if OE develops the way the OE team feels is best. If a fork happens, I personally don't see any problem with the idea of the OE project & Oolite.org remaining in friendly terms, & having links to each others' forums, etc. ... Lestradae, your opinion on what I just said is particularly welcome!

I think that, if enough knowledgeable and skilled people are ready to help me form a new OE team, if aegidian is OK with the whole idea, and if the licensing mess can be resolved in a meaningful way (I will actually attempt to ask all of the ~200 oxp authors that did not use a share-alike derivatives allowed license if they are OK with inclusion and in some cases, adaptation of their work for OE), this is a great idea that could well become the way forward.

The only other idea that sounded doable would be ClymAngus' oxp-library: With checkboxes and automatical upgrade downloads, but that would be something to which I could only contribute the OE unique stuff, and the old inconsistencies amongst oxps would reappear then. I ironed out quite a few of those over the course of the last year.

So yes, one of these two ideas could become the way forwards for OE. And Kaks' idea would mean bigger options to broaden the scope of a standalone Oolite (Extended) branch, as that one would not have to be as careful to the sensibilities of the more purist Oolite fans as Oolite itself.

That is my opinion.

L

Kaks, a well reasoned argument. regarding the Cobra points tho, as a particular reason for your decisions : Firstly, i think it's established that the Oolite universe is some years on from the Elite one. Advancements have been made. Also, in my mind, making the Cobra now NOT the best ship is more realistic for a game start - if you're a rookie pilot, you should have to WORK to get that best ship, not be given it by default. Therefore to me (assuming youhave a bunch of ship OXPs added in) starting in a Cobra MkIII is now a slightly more realistic scenario.

Well, I brought it up in the first place, therefore I am clearly supporting a forking-approach, especially because the other alternative, an OXP-manager, while desirable, is not actually in sight. All debate over it so far is purely hypothetically.

I think having a fork is the natural way of going forward. Why? Because, if we're honest, OE has been a project of its own for quite some time now. I see it like a body-kit for a car (Oolite being the car in this analogy), or actually more like a full customization à la "Pimp my Ride" or "Overhauling". In the end you might even wonder what the original car was. Anyway, the original and the fully customized car will end up not only looking, but also behaving very differently. It's only an analogy, and like any analogy it only works so far. But I hope it helps to visualize my point.

The point is: there is nothing bad about the proud owner of the fully customized car joining a different car club from the one for the "vanilla" cars. The customized car will get improved in its own way, and next year's version of the "vanilla" car may go into a different direction.

What I want to say: We don't need to be afraid of forking. Yes, in the end we may have two very different games. But there is nothing bad about that. After all, they will have been created in order to suit different approaches and emphases of gameplay. Kaks' explanation of the two different emphases makes a lot of sense to me, and I am not surprised to find myself on the "focused more on the independent trader" side of things. Thanks, Kaks, for clearing this up!

So I am with Kaks in calling Oolite a re-interpretation of Elite. OE could then be called a game based upon Elite. Both have their merits. Both will have their fans. None of them is superior over the other. Personally I am with one of them, but that doesn't mean that other people shouldn't be with the other one.

Bottom line: I think that forking is as good a win-win situation as we are going to get, considering the technical and practical matters. For me it is where the road leads us.

PS: My OXPs will continue to be available for everybody to use and build upon. All of them are clearly designed with the "focused more on the independent trader" emphasis in mind. Therefore they will fit naturally into the Oolite environment. But this doesn't speak against transferring them into the OE environment and seeing what happens. You just have to know that their natural habitat is Oolite, not OE.

Killer Wolf wrote:

Kaks, a well reasoned argument. regarding the Cobra points tho, as a particular reason for your decisions : Firstly, i think it's established that the Oolite universe is some years on from the Elite one. Advancements have been made. Also, in my mind, making the Cobra now NOT the best ship is more realistic for a game start - if you're a rookie pilot, you should have to WORK to get that best ship, not be given it by default. Therefore to me (assuming youhave a bunch of ship OXPs added in) starting in a Cobra MkIII is now a slightly more realistic scenario.

I suggest sticking to Selezen's time frame. I've used this as a background to my stuff, so it looks like this:

The Dark Wheel - 3115 AD
Elite - 3125 AD
Status Quo - 3138 AD
Oolite - 3140 AD
FE2/FFE - 3199 AD

Cheers,

Drew.

To me its pretty clear that a lot of stuff that OE is trying to do would be a tonne easier if it was dont in the core game (in Objective-C) rather than in javascript in OXP. This means a fork.

One major advantage of forking is that with OE needing a better hardware spec than oolite, you could probably drop a lot of the paranoia about maintaining backward compatability and maybe support for all the OSes, this would allow easier progress.

Another thing is about direction, oolite is fundamentally elite with a few nice bits added. OE intends to be much more, I believe it wants to be a really up to date space fighting and trading game, which doesnt need to worry about the elite legacy. For that reason, I think there are different aims, and a fork would make sense.

An OXP manager would of course be lovely, but I am deeply skeptical of how easy it would be to implement (although I have myself previously posted some possibilities). Of course with OE as a fork, you could mandate certain requirements for plugins, that would make this easier. One of the chalelenges for oolite is that there are existing OXPs which somehow need to be brough into a new system -- a forked OE wouldnt need to feel bound by that.

To be clear about this, a fork isnt a negative thing. A fork is a complete copy is the code, which L can maintain himself, will have all the features of oolite, and everything he needs to get going, unless you change it, all the OXPs will still work etc. All he will need is source control (can probably use berlios, just like oolite), and people willing to work on it. What this would do is take control of OE into your hands. No more debate about who can do what, or arguing over features. It becomes yours do wo with as you wish.

I agree with the forking Oolite idea then.

My timeplan will look like this:

1) Leave a breather to everyone and come back to the topic in 2-3 weeks so that this does stay an Oolite and not an OE board, and so everyone can calm down.

2) Open a new thread (then! don't panic anyone) in which I will ask people about what can be included, what can be altered, what can not be included into a future OE branch concerning oxp materials.

3) Get into talks with who would want to join an OE branch effort teamwise - for I could never ever do this alone, neither concerning skill nor workload.

4) I'll found a new Oolite Extended board, talk with Oolite's devs about branching procedures, and hopefully establish and maintain a good relationship between the two boards.

For me there is no need for further big public debates - the outcome was not what I expected and not what I originally wished for, but if a solution has been found that everyone can live with that is good.

L

my online time is limited as of late, so i have been quiet for a long time. in fact i missed the entire debate until now. i agree with a fork.

attempting to find, contact and obtain information about the wishes of the creators is a good idea... eventually though, for some oxps, i expect them to fall into the abandonware category. ^^

when a fork occurs, i'll be around both groups, and likely still quiet most of the time. i'll still want to be in with beta testing, though i haven't been able to do much since my first attempt at the dyson sphere (for which i have some possible ideas about resolving the invisibility/failure to draw issue....) then again, with a fork, it might be able to be coded directly in. ^^

anyhows, not that my opinion is needed in any way, but i'm willing to give what (limited) support i am able to give. go fork. go blue rajah. ^^

Legally there is no such thing as abandonware. This is on reason its a good idea to give a clear license so that people know how they can use your work