jMonkeyEngine 3.0

Originally I was also a bit startled by that blog-post renanse made. But many things said in this thread, especially Ilama's points in the end have completely put my mind at ease… It is true. I did not come to jME for what it could be and would be. I have always embraced it for what it is.



Also, I think that this thread has been a good thing for us. If people have doubts and fears, it is important to tell about them and talk about them. This holds true about real life relationships and online communities also I believe. We come out of this with a more clear vision of ourselves and a more realistic plan for the future.

darkfrog said:


That's because we're already doing it...


Have the bugs of the last few months been checked in ??
theprism said:

Have the bugs of the last few months been checked in ??


jME has had 11 commits total in the last 30 days.  So yeah, many have been addressed. :)

I love developing games and 3d apps in Java (the years of doing games in C for pc and consoles leaving me with a bad taste in my mouth), so I really want the engine tech to move forward and be embraced by a bigger part of the industry.

For that to happen I think it's crucial that it exists several collaborating and/or competing open-source engines, to borrow ideas and code from each other, and trying out different paths and solutions.



Also, this is probably what jME needs in order to get some activity going again, not counting on daddy Renanse to take care of everything. A little nail up ones scrotum can do wonders sometimes :slight_smile:



In the end I believe this will be good for Java and gaming as a whole…

MrCoder said:

Also, this is probably what jME needs in order to get some activity going again, not counting on daddy Renanse to take care of everything.


I like that...hey Renanse, you should change your nickname here to "Daddy Renanse"....maybe even "Big Daddy Renanse", it has a bit more of a ring to it.

MrCoder said:

A little nail up ones scrotum can do wonders sometimes :)


Great imagery, thanks for that. 
Vardamir said:

You are absolutely right. Looking at the current developments, i'm involved in, it's all going in the right direction. We have already a stable foundation with the engine. When we all work together, we'll have a reference implementation someday.
I agree with darkfrog, if the project fails, because one or two developers leave, then it failed from the beginning, but that's not the case here.
I can only speak for myself, and i'll stick with the jME for now ;)


Same here!

So a bunch of the grizzled old ancients are leaving to work on another java engine, to be designed AAA from the start (which is always a good thing too), doesn't mean we, as a community, can't get organized! If jme is to live on after Renanse and Mr.Coder, despairing would only fulfill in killing it!
MrCoder said:

I love developing games and 3d apps in Java (the years of doing games in C for pc and consoles leaving me with a bad taste in my mouth), so I really want the engine tech to move forward and be embraced by a bigger part of the industry.
For that to happen I think it's crucial that it exists several collaborating and/or competing open-source engines, to borrow ideas and code from each other, and trying out different paths and solutions.

Also, this is probably what jME needs in order to get some activity going again, not counting on daddy Renanse to take care of everything. A little nail up ones scrotum can do wonders sometimes :)

In the end I believe this will be good for Java and gaming as a whole...


So this was kind of a wakeup call. At least i see it this way. And as we can see here, we are all awake! That's good.
The argument was, that jME is not capable of beeing used as an engine for those AAA game projects, is that correct?
To be honest, i can't tell, because i have never been part of such a game project, not even a game project at all. But i'm sure there are better engines out there ;) I always mention the CryEngine here, because this was created by some fellow bavarians in my region here.

Maybe we have already all parts that would make up a state-of-the-art engine, but those parts are only loosly coupled.
When we talk about the jMonkyeEngine here, we are only talking about the core. maybe i'm wrong here, but the following are the parts that form the engine:
- jME Core
- Physics
- JGN
- Terra (AFAIK we have two competing engines)
- MW3D
- Monklypse (for those who don't know: http://code.google.com/p/monklypse/)

Why not follow the example of eclipse and form some sort of foundation? Lets call it the jME-foundation for now, that would have some organization like this:

- jME foundation
  - Core engine
  - Physics extensions
  - Networking extensions
  - Terrain extensions
  - Tools
  - IDE integration

Just like the eclipse project, i am also a friend of a fixed schedule development cylce. With all those subprojects synchronizing their releases.

We can also think about beeing open for some kind of sponsors. Companies, that have a strong interest, that the engine they are using is kept on par with all the other AAA engines out there.

Competion is good, but cooperation is even better!
Starnick said:

So a bunch of the grizzled old ancients are leaving to work on another java engine, to be designed AAA from the start (which is always a good thing too), doesn't mean we, as a community, can't get organized! If jme is to live on after Renanse and Mr.Coder, despairing would only fulfill in killing it!


Ewww, I change my mind, Renanse, please change your nickname to "Grizzled Old Ancient"...that's so much better.

Vardamir said:

So this was kind of a wakeup call. At least i see it this way. And as we can see here, we are all awake! That's good.
The argument was, that jME is not capable of beeing used as an engine for those AAA game projects, is that correct?
To be honest, i can't tell, because i have never been part of such a game project, not even a game project at all. But i'm sure there are better engines out there ;) I always mention the CryEngine here, because this was created by some fellow bavarians in my region here.

Maybe we have already all parts that would make up a state-of-the-art engine, but those parts are only loosly coupled.
When we talk about the jMonkyeEngine here, we are only talking about the core. maybe i'm wrong here, but the following are the parts that form the engine:
- jME Core
- Physics
- JGN
- Terra (AFAIK we have two competing engines)
- MW3D
- Monklypse (for those who don't know: http://code.google.com/p/monklypse/)

Why not follow the example of eclipse and form some sort of foundation? Lets call it the jME-foundation for now, that would have some organization like this:

- jME foundation
  - Core engine
  - Physics extensions
  - Networking extensions
  - Terrain extensions
  - Tools
  - IDE integration

Just like the eclipse project, i am also a friend of a fixed schedule development cylce. With all those subprojects synchronizing their releases.

We can also think about beeing open for some kind of sponsors. Companies, that have a strong interest, that the engine they are using is kept on par with all the other AAA engines out there.

Competion is good, but cooperation is even better!


I happen to agree that a start from scratch is a much better idea in many ways.  jME carries a lot of baggage and whether it's a new project or jME 2.0 as I was anticipating when that development first started, I think taking a step away from what we've clung to all this time and asking ourselves what the best way is to do each task is the smartest approach.

No worries though, even if Uber3D comes out so much better than jME as long as it's open-source we can always use their work as a basis to further jME as well....giving credit of course and I'm sure all other engines do for us. ;)

Sure, a fresh start is not the worst thing to do. I'm this kind of radical as well sometimes :wink:

But we need to consider, that there are many projects out there already, that rely on the engine in its current state. When we decide to start from scratch, with all the expierience in mind, we should really think about backwards compatibility, even if it would only be some compatibility layer.

Before thinking about that though, we have to put jME 2.0 in its final state.



And for the renaming of renanse, i would go for "gray eminence"  8)

In all honesty, how many projects are there, which are genuinely likely to come to fruition and which are truely pushing the limits of what JME can already do?

Lead developers will automatically lead developers if they do a lot of work (only lead dev right now would be Mojo himself I'd say). They won't be before they do a lot of work.



Making a roadmap is pointless if there isn't anyone commited to actually making the features on the roadmap.



Coding will help jME more than discussing on IRC every week.



If enough people start doing some coding, meetings, roadmaps, etc will follow naturally.



Well that's my opinion anyway.

Vardamir said:

Sure, a fresh start is not the worst thing to do. I'm this kind of radical as well sometimes ;)
But we need to consider, that there are many projects out there already, that rely on the engine in its current state. When we decide to start from scratch, with all the expierience in mind, we should really think about backwards compatibility, even if it would only be some compatibility layer.
Before thinking about that though, we have to put jME 2.0 in its final state.

And for the renaming of renanse, i would go for "gray eminence"  8)


That also sums up nicely, why folks leave and create something from scratch. Once something starts to mature and get used, every version simply cannot be a complete redesign (as you mention), else no one would use it.

And, heh "Grizzled Old Ancient" as a title does sound pretty good come to think of it...

I completely respect Renanse, but I also share some fears about JME's future.



I too think that some management is needed, 1) a group of lead developers including fresh ones, 2) a roadmap, 3) a better documentation project.



In any case, maybe not so many things need to be added urgently, but a team taking care of the engine is a must. I don't think open write access to the repository is a good idea.



We could also discuss in the IRC.



Maybe a weekly or fortnightly meeting between lead developers in order to discuss a roadmap could help too.



I can put some time into this, although I am myself a quite basic 3d coder I could help with other tasks. There are few persons with on going developments that maybe could take some responsability. Just to keep things going.



All the best to Renanse and to JME, although I also think Renanse could have directly forked JME and maybe many would have followed.


llama said:

Lead developers will automatically lead developers if they do a lot of work (only lead dev right now would be Mojo himself I'd say). They won't be *before* they do a lot of work.

Making a roadmap is pointless if there isn't anyone commited to actually making the features on the roadmap.
No doubt about it.
llama wrote:
Coding will help jME more than discussing on IRC every week.

If enough people start doing some coding, meetings, roadmaps, etc will follow naturally.

Well that's my opinion anyway.
Are you sure about this? I would be very careful to downplay the importance of active, perpetual communication when dealing with large-scale teams like this one. How many "online meetings" have you really tried to schedule and hold in the past? When did you last talk with two or more other developers at the same time about jME?

I love how all boys between fifteen and twenty are in a constant state of hubris. "I have zero experience to back it up, but I know everything"  8)

What large scale team?

llama said:

What large scale team?

probably ">1" = "large scale" :P

I think he's calling us fat. 

Starnick said:

That also sums up nicely, why folks leave and create something from scratch. Once something starts to mature and get used, every version simply cannot be a complete redesign (as you mention), else no one would use it.


That is the reason, why design is so important in the first place.
llama said:

What large scale team?
I was referring to the team of administrators, developers and board moderators.

I mentioned the live meetings not as the one miracle remedy, as it might not work out at all for the perticular type of collaboration you've gotten used to over these past year, but I put it out there as one of many possible means to sharing opinons, personal agendas and other types of information, which is what I think is necessary in order for a proper roadmap to eventually appear.

Maybe coding alone will get all the necessary wheels rolling eventually, just figured I might be of help by presenting you with other options.