Yes, as said before I’d rather fork an immature lib to handle my own project, since we can’t start a serious project (an experimental worldwide GPS software in 3D using sat maps to handle logistics) using immature tools (JOGL, LWJGL or whatever). Better to develop it in-house so you can support what you need and ditch the other stuff you’ll not need and have a stable release ASAP.
It’s the centerpoint of software engineering, the vision must be consolidated before starting the development.
But the more you try to reach 1 the more your
specifications should be consolidated.
Yes.
But if there are some things which doesn’t make much sense, you have to change them. Remember the days where we had a display and window class? I always mixed em up and I’m really glad that they were merged. It just makes much more sense this way.
Many moons ago there where even pointers in lwjgl. Do you miss em? 
Maturity is a good thing, but it takes it’s time. It’s evolutionary. If you don’t want to do any changes every once in a while… you certainly don’t need to. You can continue using an older version and - eventually - do all the changes at once at some point in the future (eg if you’re entering the beta phase).
[quote]>But the more you try to reach 1 the more your
specifications should be consolidated.
Yes.
But if there are some things which doesn’t make much sense, you have to change them. Remember the days where we had a display and window class? I always mixed em up and I’m really glad that they were merged. It just makes much more sense this way.
Many moons ago there where even pointers in lwjgl. Do you miss em? 
Maturity is a good thing, but it takes it’s time. It’s evolutionary. If you don’t want to do any changes every once in a while… you certainly don’t need to. You can continue using an older version and - eventually - do all the changes at once at some point in the future (eg if you’re entering the beta phase).
[/quote]
Agreed.
I only feel odd that supporters continue to stress the in-development of the library but, on the other hand, continue to support the idea that it is a “pro” tool.
Sticking on an alpha or beta (and outdated) build to avoid misalignment is not my way to develop things.
[quote]Matzon saying stupid to others and keeping things close & personal will not make you be smarter, just childish.
[/quote]
I never said you were stupid, but rather what you said was stupid. Your whole post was basically invalid, and based on no logic or truth - and IMO it sounded very uninformed (as if you used LWJGL 0.5, turned away and never looked back).
[quote] If JME is not able to run for a simple 0.01 update is your fault no matter how right you think you are.
[/quote]
You’re kidding, right? LWJGL has been in alpha release since day one. It’s not something we slapped on because we like the name Alpha, but rather because it is alpha. We reserve the right to change the library any way we see fit, and in any way we like. Thats not to say we enjoy breaking stuff. But we have a goal, and if we need to break stuff to get there - so be it. As for jMe, for each version they release they also break stuff. It’s the nature of any evolutionary project. But for each release they typically specify which LWJGL to use with that version. Like any other program. Be it a 0.1 increment or not.
[quote]Anyhow, let me remember you that most of the features removed where because you cannot stand the effort to support them cross platformly (like the input devices or EAX sound) at the moment. Will we must assume “blinking” features? Better to leave them incomplete to than removing them to be added later.
[/quote]
Once again (and not being personal) you come off rather uninformed.
1: EAX cannot be supported crossplatform since it’s only available on Windows.
2: The controller stuff was removed, since it only worked on windows, as stated in the release thread on JGO. It was decided that it is better to support controllers completely than to have half an implementation blow up on users on non-win32 platforms.
We think it’s a good decision to do so, those that disagree are more than welcome to build from a date prior to the cut.
LWJGL does not cater for all users, but rather for the common case. As such, some will be burned by some of our choices - but most will benefit. Should someone feel something is bad or wrong, they’re more than welcome to raise their concern on the boards or on IRC. But the way I see it, those that do the actual coding gets to decide the direction… it’s not like we’re flooded with controller patches or software EAX implementations for non-win32 platforms each day. It’s actually been some time since I’ve seen a patch from anyone ('cept some mispelling or very minor patch (line or two)).
I would like to point out that all previous versions of LWJGL are still available. We break nothing. If you’re happy with a particular release, stick with it.
If it’s any consolation every time the API is broken I now have 3 entire games to keep up to date - and it’s really a trivial matter to do so.
Cas 
I’ll pipe in on the jME part…
Current release 0.8 uses 0.94, as it was released prior to 0.95. We periodically have people on the boards mention problems, we tell them to use LWJGL 0.94 and the issue is resolved. We missed that release by about a week, which is too bad, but oh well.
CVS will have 0.95 support soon (getting some other core changes in first). And when we release 0.9 it will probably use LWJGL 0.95 when LWJGL 0.96 comes out. 
It’s the nature of the beast and we (jME team) grunt and groan now and then, questions some of LWJGL teams decisions every once in awhile. But overall, moving from one version to another has not been difficult.
Our getting started guide specifically mentions which version of LWJGL to get, so that keeps many of the issues from cropping up. We also put the LWJGL version into cvs lib directory to allow the CVS users to always be current and that seems to work well.
We never intended LWJGL to behave like an “extension” or a “dependency”… why not just bundle the binaries along with jME and solve the problem completely?
Cas 
[quote]You’ve just broken support for one of the best 3D library out there.
And what about the countless professional games developed around the library? It’s stupid “whining” (or it was suggesting, point of views) for an unfinished library but is intelligent (and pro) using an immature artifact to develop professional things? You’ve a quite confused opinion about how “pro” software is developed (I must assure than starting on the base of an alpha lib is not part of the equation, JOGL is not excluded from these considerations).
[/quote]
You are quite confused about how pro software is developed. When you start developing, you pick a library version and stick with it to the end. The only reason to move to a newer version mid-stream is for a required patch or if you need/want a new feature. Everyone who makes professional production software knows the risks of changing a library version mid schedule. It is always a risk.
And jME is not broken. Each verison of jME specifically says which version of LWJGL it is made with. If you move to a newer version, you broke it.
If that is what you do, then what are you complaining about. Pull version 0.94 out of CVS and have at it.
Erh, no, that is simply flat out wrong. I’m very active in the development of LWJGL, since we’re using it for our Tribal Trouble game. And several others from the LWJGL development team is using it for their projects too, and they too provide fixes and features. So naming LWJGL a one-man-show by Cas might have been true in the beginning, but certainly not now.
And I’m not sure what you have against our Mac donation drive. We had been asking for help for ages, and then decided to ask for money instead. It worked out great (modulo problems with Paypal that we could do nothing about) and we have a stable and working port today.
- elias
Not handy, and I’ve not the time nor the inclination to mess around making one. The last official word I was told was that it was fundamentally incompatible with how the linux port did things, and wouldn’t be fixed.
One other thing to throw into the mix, and it’s not a fault of JOGL: LWJGL fullscreen works correctly on 1.4.0 and above. JOGL has unfortunately falley prey to a ton of bugs in AWT that to this day remain unfixed.
Regarding work on LWJGL, I do almost bugger all these days, as Elias is driving the implementation to get Tribal Trouble released. This is actually the best possible situation, as instead of writing an API for n00bs and wannabees and theoreticians and hobbyists, LWJGL has only ever been designed around the fundamental problem of getting a commercial game out there.
Cas 
… and not only games…
Cheers 
Mik
Hehe, yes of course, LWJGL was first actually used in production for live TV graphics and still is to this day.
Cas 
I can run Castalia at 400fps (onscreen) vs 25-50 fps of Java2D. I can have NPOT images, video textures (AVI/Quicktime), etc. Learning a bit of OpenGL and playing with it through LWJGL has been really straightforward. I needed some JNI in order to overcome some LWJGL-NIO limitations but nothing transcendental.
Now I can easily mix Java2D and OpenGL’ed stuff with very little effort.
Cheers2 
I don’t know where you heard that (if it was from me, I was wrong), but it is incorrect. We fixed JOGL’s visual selection mechanism on X11 a long time ago.
[quote]One other thing to throw into the mix, and it’s not a fault of JOGL: LWJGL fullscreen works correctly on 1.4.0 and above. JOGL has unfortunately falley prey to a ton of bugs in AWT that to this day remain unfixed.
Regarding work on LWJGL, I do almost bugger all these days, as Elias is driving the implementation to get Tribal Trouble released. This is actually the best possible situation, as instead of writing an API for n00bs and wannabees and theoreticians and hobbyists, LWJGL has only ever been designed around the fundamental problem of getting a commercial game out there.
Cas 
[/quote]
Testcase not propaganda.
I used JOGL (on a ATI, I must add) and LWJGL equally well in fullscreen without hassle. Just a coincidence? People have problems even with LWJGL, you’re about to add Swing AWT support to LWJGL too, you’re using a bug free AWT?
- AWT is buggy (that’s true, but JOGL uses only a small subset of it).
- Swing is slow.
- JOGL is slow (in fact speed is not that different between LWJGL actually).
- Java is crap.
Common statements that most of the time are completely wrong or only superficial or driven by pride.
I can only shiver if people are trying to go in production with alpha versions (I’ve doubt how pro they can be to do such decisions) of their core tech.
I simply admire Elias with its tribal trouble project (the kind of game I strived to develop but wasn’t able to find able people for, here in Italy), expecially since he is maintaining the lib at the same time… Being there done that, it is the only way to do the stuff, as said before.
You should get more easy on criticism, btw. I read almost all the board (Indiegamer too) and blatant FUD against everything is not LWJGL is rampant.
I must reckon you are not so fond to negative feedback about your work (after all people ship games from it, it must be good, wow!), but you must accept it.
After all other people are enduring your propaganda for months (if not years) without getting it close & personal like you did…
BTW I suggest to steer the 3d out of this sterile (if polite) flame war.
To whom am I getting up close and personal, and where’s this FUD of which you speak?
Cas 
As of today, this is no longer an issue. See
http://www.java-gaming.org/cgi-bin/JGNetForums/YaBB.cgi?board=LWJGL;action=display;num=1108645920
Sometimes, criticism leads to new features and fixes 
- elias
[quote]Comparison of JOGL and LWJGL seems pretty pointless to me, as their design goals are different.
[/quote]
By definition, this means it is extremely important to compare them.
I think what you actually meant was “arguing which one is “always best”” is pointless…".
The considerable differences make it essential for us users to have a convenient comparison so we can make up our own minds based on up-to-date info.
I’ll add my 2 cents…
JOGL apps crash my computer with a blue screen of death.
LWJGL apps are beautiful and not a one has ever given me problems.
EDIT: OK maybe I’m just a little peeved at how I can’t even start Wurm Online…