I mean this entirely literally: LOL
That is all.
edit: unless of course it was a serious suggestion - in which case I’m 100% behind it.
I’d just like to say I’m eternally grateful for the work people have done maintaining GL4Java (Sven/Ken + others) years ago and in turn JOGL. It’s helped me to learn loads about OpenGL using my main programming language (Java) without having to get back up to speed in C++ and I’ve been able to realise a lot of project ideas that were kicking around my head.
This is not a pro-JOGL / anti-LWJGL post (LWJGL is awesome too), I’m just showing my appreciation to contributors.
Thanks! 
On the anagram, I’m growing to like ‘JUMP’ … it’s catchy.
[quote]I’d just like to say I’m eternally grateful for the work people have done maintaining GL4Java (Sven/Ken + others) years ago and in turn JOGL. It’s helped me to learn loads about OpenGL using my main programming language (Java) without having to get back up to speed in C++ and I’ve been able to realise a lot of project ideas that were kicking around my head.
[/quote]
Indeed I can fully relate to that. I have a similar background with GL4Java and I think these guys have done a great job o bring it this far.
I’m especially excited about the new features in JOGL2 like ES2, newt and CL support. Breaking the AWT tie has to be a good thing too when you look at what’s coming up with google os/android based netbooks/laptops/phones etc. It would be great if JOGL can keep it’s very light weight core with all the optional jars approach, and then if the best bits of other libs could live also as optional jars that sit on top, or along side, then that wold make a developers life easy. As always I’m thinking about deployment issues and I think the focus should not just be on the name of any unified package, but also on how do we compete with flash and silverlight. Whatever comes out of the process here has to be proffesional/serious enough to contend with these technologies, otherwise we may as well not try. I don’t think FX is the solution to Java on the desktop, I think what we build here could be though!! The idea of accessig the browser context for direct rendering sounds like it would help alot, but otherwise we need to find a simple way to deploy ‘JUMP’ components to work on all desktops/OS’s. Then we can start thinking about our own store…!!
Peter
I Agree…
Fixing JOGL is one thing, but when we look at the overall user-experience of accessing some JOGL/LWJGL content on the web, the situation is almost hopeless:
- Java is far from being installed everywhere. Many reasons for that:
- poor content offer
- installation is not seamless (as Flash is…)
- lack of clear “Desktop Strategy” from Sun, etc.
- The “Java Web Start Experience” sucks:
- It takes ages until the JVM is (first) loaded
- The jar downloading process is painful (we all know the “download stalled” issue, regardless of the quality of the connection)
- The need for certificate is a big annoyance both for users and developers (3d content should not require any certificate)
Well, nothing new. People have been discussing these issues for years.
I’m reading here and there that OpenJDK could solve (on the paper) many of these problems, but frankly, I don’t believe in a massive world adoption…
Any hope doc?
I wish. One thing you can say in Java’s favour is that it has proper Linux support.
I would feel better if we did the opposite, we keep JOGL underneath and we create a wrapper to expose a LWJGL-like API for those who wish to use it because JOGL has an OpenGL-ES support, it is an excellent reason to keep it as the core OpenGL/OpenGL-ES binding. However, I understand that some LWJGL users may prefer your suggestion.
Java is installed on 91% of the desktop computers whereas Flash is installed on 92% of the desktop computers as far as I know and I cannot install Flash on some computers under Windows Vista because the new installer works bad, it is not seamless, it seems to work bad with some firewalls, I have not found the exact root cause, the installation does not even start. There is a lack of strategy but the content is far from being poor on my view. When I see so many Flash games with a so limited gameplay, I don’t call this a rich content, sorry for Flash programmers and artists, I see many Flash games using very similar artworks or concepts, it looks like several games made with the same cake pan, it becomes quickly boring… I prefer MineCraft, Tesseract, Bloodridge, Futuristic Arenas, Incredibuilder! …
Therefore, don’t use Java Web Start, nobody forces you to do this. It is almost enough to deploy quickly a blue print but using an installer gives a more professional result (IzPack for example). 3D content should not require any certificate but the use of certificate for games accessing to the file system is a good thing on my view, it is safer, you don’t allow anyone to do anything.
That would have been nice, but we shouldn’t believe Sun propaganda.
It’s more like 78-82% (depending on the country) and about 40% still using the old plugin (pre 6u10), which is unreliable. In my own gathered statistics, about 67% of non-technical users on a non-technical website, were able to launch an applet that was able to connect to its own host.
I didn’t mean to prefer LWJGL, but purely at technical level it seems that LWJGL static methods are exposing OpenGL directly as-is, whereas JOGL approach seems little more highlevel, so this approach make most sense technical-wise. I’ll look to both LWJGL and JOGL sources to check if this is really true.
About OpenGL ES, I don’t see any problem, in any way there would be just one generator, so the ES would be supported by both APIs.
You’re quite of an optimist monsieur.
But anyway, user-experience wise, you can’t compare the overall “Java on the Desktop” experience with what Flash has to offer…
Java Web Start reflects the state-of-the-art of what Sun is capable / willing to propose on the Desktop.
It’s been here for years, it sucks (user-experience wise) and it’s broken.
Logical conclusion: it’s never going to be better (confirmed by the recent drop of Jogl by Sun…)
I don’t believe in that direction either.
We live in Browser Land, and the future will be even more browser-based (think Chrome OS…)
In my humble opinion, the hope _ in term of Java + OpenGL _ resides in Android. Not exactly as it is today but in the not-so-far future (when Android enabled devices become bigger, with decent GPUs and JIT compiling…)
Let’s try not to miss the train this time!
And the only guy who knew anything about fixing it (Ken, of plugin2 fame) got pushed out and went to work for Google. Leaving the rest to come up with the brilliant Java Store application.
Cas 
I disagree. 82% is pretty good, and since java6u10 plugin2, applets now work at least as well as webstart. As more and more computers are upgraded to 6u10, applets will only get better. The webstart ‘download stalled’ problem is annoying, but it doesn’t happen often in my experience with small 5mb apps.
To be honest, I’m always surprised at how many computers have java installed. Especially computers in universities and work places where the user account does not have admin privileges. Invariably, the restricted account will not allow program installations or exe’s of any kind, yet I always find that I can fire up an unrestricted java web start app from the net, unrestricted!
That was such a pity. I wonder what he’s working on now?
It happens for me all the time, on tiny apps (a few hundred Ks)
I have to admit that I never gave Applets a chance with Jogl…
I will make a test to see how it works on our “problematic platforms” (namely Macs with Java 1.6 and PCs with Intel GPU’s…)
I hope Ken is working on making OpenGL charged Android apps available on Java enabled desktops
Webstart stalled can be an issue with Comodo firewall (maybe other firewalls as well?). There is a well-known bug there.
I’m getting it often enough, but don’t really care since I loathe webstart’s habit of cluttering up add/remove programs and being a bit of a pain with uninstalling webstart apps.
Hooray for reliable applets
He’s working on WebGL. More reason to sort this situation out and start pushing for Spasi’s idea above
You idealize Flash, it is buggy too, Flash installer is sometimes worse than Java Web Start lol, programmers that use ActionScript have to work around some known bugs that have not been fixed for years, it is not any better.
With Flash, it’s the developer that must workaround the bugs.
With Java, it’s the end user that must workaround the bugs.
Indeed.
Check this out for a user experience:
http://www.silvergames.com/game/quake-flash/
Now the gloves are off, I want to have the same experience (or better) with Java. Go!
Cas
Gosh, the world is moving too fast with Apple and Google in the house!
Javascript (as did Flash’s ActionScript over the years) is becoming faster and faster (see it in Chrome…)
Now add OpenGL acceleration (exactly what Flash is still missing): that’s it, we’re there (i.e. a point where the need for Java can be by-passed…)
Yes, if only every browser out there was the latest version of Chrome… it’ll take years and years to catch up to the penetration that Flash 10+ will enjoy.
Cas
As of today: I guess it’s possible to state that standard web browser Javascript execution on the Desktop is actually blowing away Java running on any Android device.
And still, many people out there enjoy OpenGL on the Android platform (and it’s just a beginning…)
All this to say that Javascript execution speed is not necessarily the main bottleneck.
I’d say it will depend on how smart and open is the WebGL standard, and of course on how much time it will take to be adopted by developers and users…