JSR 231

How it is related to jogl ? Is it based on jogl, with a bit of cleanup in few places, or totally new project ?

Any chances of having entire java-gaming.org community promoted to ‘expert’ status ? :wink: Or at least having one person from Sun who would sneak some info out of expert group, posting interesting problems for discussion here ?

Link for JSR: http://jcp.org/en/jsr/detail?id=231

2.11 Please describe the anticipated schedule for the development of this specification.

First Draft: January 2004
Public Review: March 2004
Final Draft: May 2004

Damn, gonna be awhile…

[quote]2.4 Why isn’t this need met by existing specifications?

There is an existing 3D API for Java, called Java 3D… [snip]
[/quote]
Hmm… are they even aware of the existance of Jogl? Surely the organisation within Sun can’t be that bad…

On the other hand, I don’t see any familiar names on there, and you’d have thought that they’d mention Jogl somewhere in there. Difficult to imagine them wanting to start over completely, yet their implementation description starts from scratch. Nothing mentioned in the section about depreciating other APIs either.

They also metion OpenGL 1.5 spec to be confirmed Dec 2003. Isn’t this already out? Could this just have been written ages ago and only now surfacing?

And it says it’s going to be a separate download. Damn. =/

Uhm, they confused me now. Isn’t that third OpenGL API from Sun this year? (OSS JOGL, Java2D in Tiger with internal one, and tih 231 official one). Can someone from Sun clarify? Some of those will be merged somewhere in future, I hope.

Tiger isn’t supposed to be out until the summer anyway. Compared to other jsr’s this seems to be on a fast track.

There’s also the OpenGL that Sun and SGI announced earlier this year - or is this it? SGI aren’t in a lead role, so I guess not.

So that’s four OpenGL APIs from Sun this year? ???

The organization within Sun IS that bad. Read my notes on the Sun Network Conference. As I was talking to people at the conference from Sun, some even demoing graphics technology - NONE of them were aware of the Java Gaming Initiative APIs.

I have a close relationship with people from SGI and am working on getting some of these pieces of information together. Right now it is very hush hush all around. I hope that the Java community can come together and figure this out. I am trying to make contributions on a personal basis to JOGL, but will eventually move to whatever “standard” there is for Java/OpenGL bindings.

-moorej

[quote]The organization within Sun IS that bad. Read my notes on the Sun Network Conference. As I was talking to people at the conference from Sun, some even demoing graphics technology - NONE of them were aware of the Java Gaming Initiative APIs.
[/quote]
What’s sad is that it is so easy for us to see it from the outside. Maybe we could be on Sun’s payroll to point out areas where they can avoid doing duplicate work :).

[quote] Right now it is very hush hush all around. I hope that the Java community can come together and figure this out.
[/quote]
Hmm… the big question is, why so secretive? Its not like this is some new revolutionary technology here, its ‘just’ allowing programmers to access and existing API. Jogl etc. were produced in complete secrecy, and what did it gain them? A whole heap of buggy work to start with, missing platform support and a various other glitches (thankfully getting slowly ironed out now ;D ).

I fail to see the need to keep all this stuff hidden, especially when it has the ability to swing from successful to neglected depending on the communities use & reaction to it…

Because it’s more fun that way :slight_smile:

Will.

Jogl/Joal/JInput are all open source, which means that there is no need to switch to the “newest” OpenGL implementation.

I would really like a future, where Java is completely open source (OpenOffice shows that it’s possible). I am not an insider, but it looks like Sun isn’t able to manage it properly anymore. If Sun isn’t even aware of the Javagaming Initiative how can they advertise it to others? How can they make Java a gaming language if they don’t know their own APIs?

Okay, relax everyone. The games group is the group responsible for all of these actvities. The intent is that this JSR starts with JOGL, and modifies as nessesary. Becausr of the way the JCP works, you can’t force anything on a JSR, so it could not be submitted as such. There is no mass disorganization at Sun. This is the result of the agreement between SGI and Sun. SGI decided not to join the JCP, but will be involved - at a minimum through their involvement in the OpenGL ARB.

There is a focused strategy to all this - even though it sometimes is confusing. Sun does not have lots of money to go around doing competing initiatives. We are all focued on the same things. Having said that, Sun is still a big company, so some people in the company may not know what others are doing. That is not the case for any of the graphics technologies today.

Doug.

[quote]Okay, relax everyone. The games group is the group responsible for all of these actvities. The intent is that this JSR starts with JOGL, and modifies as nessesary. Becausr of the way the JCP works, you can’t force anything on a JSR, so it could not be submitted as such.
[/quote]
Then why not just focus the scope of the existing JSRs. Why are there pretty much 3 expert groups working on the separate JSRs? If SGI wants to go and work on their own, fine - no biggie; but I still don’t see any reason for JSR231. Where is this JSR going to diverge from the already existing JSR…s.

While there may not be any specific disorganization within Sun (though as an outsider talking to folks working at sun and in sun research I just can’t understand why certain groups are unaware of these JSRs), the impression that is definitely coming from these proceedings is one of confusion and duplication.

There are no other JSR’s for this activity. There is only one JSR for Java bindings for OpenGL - 231. SGI is not doing their own thing. We are all working together. I don’t know where the impression came from that everyone is doing their own thing.

As I said, Sun is a big company. JSR’s do not become compleetly public until they are filed. I don’t know who you have been taling with, but it would not suprise me if some project off in a corner did not know what the games group is doing - we are only 6 months old.

Okay. So if I understand you correctly the J2ME OpenGL JSR, the JGI JSR and this new JSR 231 are the same thing or at least the same set of people?

I was talking to the hardware group - they are actively trying to sell graphics workstations and showing cool demos of those hardware platforms doing nice stuff in OpenGL. Since I know for sure that they sell into the government which uses these things for government simulation and training - it would be nice if they knew there was a Java solution already in place.

There is no J2ME OpenGL JSR. JSR-184 is the only 3D JSR for J2ME, and it is a high level scene graph API. I am involved in that. The JGI JSR is now cancelled, as it states on its page. Those technologies are making there way to java.net. I am involved in that. And, 231 - which is the only JSR related to OpenGL has just been filed. And yes, I am involved in that as well.

I work in the software CTO office. I am the chief architect of the games group, and I am also focused on graphics strategies at Sun. So, it is my job to know and help drive all of these activities. I am very aware that lack of information is the biggest reason for all of this confusion. I hope that we can make things a little more transparent, so some of the confusion can go away.

I used to work in that group, and they definitely know what is going on. The problem is that most markets that they are working in do not currently use Java for their graphics applications. Some are moving in that direction, but they are not there yet. So, Java graphcis rendering is not at the top of their list of priorities.