Blender Java Port
Jason LaDere has started a Blender to Java port as a personal learning project. So far, he claims that basic functionality is already working (unconfirmed, as it doesn’t run on OSX yet). I must say that I have serious doubts about the usefulness of a Java version of Blender (maintaining it would be a hell of a job), BUT I do think it’s a tribute to the spirit of Open Source. What’s your opinion?
I have started a Java port of the Blender 3D software. Currently it is capable of loading/saving files, manipulating UI, rendering, and editing meshes. I am curious to see how much interest there is in this project, so please check it out.
So far this a proof of concept work to see how many people might be interested in this project. The Blender Java version available through webstart and depicted in most of the screenshots is based on Blender 2.48a code. I also have another version based on the current 2.5 code which can be seen in the last screenshot. The 2.5 version is not as far along yet. Blender Java uses JOGL for 3D rendering.
Links
I’m astonished. It really looks like Blender ;)
it opened in a flash, using netbeans I’ve almost forgot I have a quadcore in my pc – it’s so sluggish
I wonder if this Blender Java will open that fast even when it will be completed. (IF it will get ever completed)
In particular I fear it could have problems handling big scenes.
On portability and hackability of java source code java is far better than C, no doubt about that.
Good and interesting work!
I really had to double check it wasn’t april 1st.
That aside if this is true I wish him the best, it would be very, very useful to have Blender in Java code. Yes it may be slow but machines are always getting quicker and with JIT compiling and all the smart tricks he can probably do it would be very interesting indeed.
Could this lead to a blender that would work in a web browser, I guess that’s what the JNLP/WebStart version ? Something like the blender browser plug-in except as a java applet so it might be more cross platform?
Interesting idea, didn’t work for me on my Linux box. The JNLP started with javaws but all I got was a grey widow with my normal window manager menu controls ( maximize, minimize, close).
Edit: I tried again from the command line and it did start up (once) and I saw the 3D view and buttons but most of them didn’t do anything. I could zoom in and out with the scroll wheel but that was about it. Very nice start, looks very promising.
Edit: Reading through the about I guess my problems as known and common on the Linux version due to a JOGL threading issue.
Blendroid? (Blender for Android, I heard Java was used in their new googlephones OS)
It could also be used to make some ports for symbian?
More interestingly, I wonder if it could be used in a cloud. This could be interesting if someone could use Blender UI from distant servers (like in the case of a render farm)
Don’t shoot me. I know I am ign’ant. Especially when it comes to java…
:)
That’s pretty cool, I must say. But given that Blender runs on all major OS, I don’t see the point of porting into JAVA, especially when C is so much faster.
Browser embeded BGE games, similar to quakelive.com … that would be interesting :D
That’s impressive, especially for a project of one guy. And I see a lot of potential in this port if it succeed to be ported totally, particularly in prototyping new features.
Next time, please keep that post for April 1st ;)
I highly doubt that a single guy alone can work through the port of a codebase developped by dozens of developers for decades. But you never know, congratulations to him if he makes it ;)
I think the project is great as a learning tool, but not as something many people will get interested in.
It would be much more interesting if he was porting it to D for the core, and Lisp as a replacement for python.
Kindest regards to Jason, and sincere props for all his hard work. Remember that while I rant. :^)
I really want to wish this effort the best of luck, but I can’t find it in my engineering heart to do so. Not because I love C and hate Java, but because if this succeeds, some hobbyists and one-off projects are going to hack in specialized functionality not by using the elegant Python interfaces, but by cramming some more Java code into this version. And maybe this additional functionality is useful, but it doesn’t run under IBM’s JVM, or Sun’s, or this-or-that version number, or when the planets aren’t aligned. This sounds a lot like paranoia, but it’s the reality of diverse developers plugging chunks of Java code into one another.
(and now to try and be more constructive)
What I’d love to see instead is an effort to expose what we have as the Python API to other languages, Java included. Folks can then write their plugins and addons in whatever language they’re comfortable in, but to common and well-defined entry points and without all the wheel-reinventing and fork-maintaining and JVM-worm-can-opening. :^)
It’s not that a fully Java-implemented Blender couldn’t provide and even encourage this kind of behaviour, it’s just that there are swarms of Java “programmers” out there who will instead think, “But I don’t need to bother with the API, I can just hack my piece into this existing class. …And that one. …And bring in this other library…”
Anyway, I really do applaud all your hard work. Happy hacking, and here’s hoping I’m way off. I love being proved wrong when I’m pessimistic. :^)
Respect!!!
Please, stop. Anything I have ever seen done in Java was disgusting. Focus on something that will actually bring good things to the world.
Good work but… what is the purpose of this?
Very impressive, all this in just a few months! Kudos to you.
And this is really fast, almost unreliably fast for Java!
This could have a lot of value to attract new developers, get new features and port them to the original Blender.
And as Dread Knight said, a java version of the Blender player, to get embeded BGE games/apps would be very nice.
Congrats!
I think this is great :D -first send people there to try blender out, if they like it, they download the real thing and use it..
Some of us seem to forget that there are people without pcs, or can use someone else’s but don’t have administrative rights to install it but love 3d so much.. –like.. what if there’s no way to install it, but you’d love to use it? -Seriously this is great!!
-even though it’s on every platform.. I couldn’t install it until I got my own pc, so I admired it and prayed to finally be able to use it :D -this would’ve been a great start ^^
I LOVE IT!!
Well, I hope some day the BGE work on this port.
I hope you can make a BlenderPlayer Java for web.
I just downloaded the software,
it works (on ubuntu 9.04 and java 6)… there is the 4-way screen, i was only able to move the 3d cursor, but it is definitely blender :)
good work!!
As somebody said previously I had to double check if it was the 1st of April..my goodness, that’s a nightmare of job to port Blender to Java and honestly I don’t see an usefulness for Blender users. However, I’m not saying that this guy is doing a stupid thing, I really admire him for his “crazyness” and iron will to carry out such a big project by himself. As stated in the article, that’s the spirit of opensource! Let’s keep on this way!
Astonishing!!! This is truly a mammoth task!
Well done so far!
I don’t get the point of that port unless there’s a need to make Blender slower for like three times or so.
Nice.
Perhaps you should take a look at the JMonkey Engine:
http://www.jmonkeyengine.com/
It’s a 3D game engine written in and for Java.
i agree with auria. i wonder how this is possible? blender has several 100000 lines of code. hm… is this some automatic or at least semi-automatic c to java translation?
having a java version of blender certainly is interesting though.
One of my professors stated some weeks ago, that java wasn’t slower if you just use the right compiler (which isn’t for free…). Another professor did some peformance testing on sorting algorithms, the jdk-quicksort being faster than the c++-equivalent.
As penduin stated, there are quite some issues to discuss about concering further development (and who convices brecht to code in java ;) )
I’d also like to know whether it’s semi-automatic porting.
Nevertheless it’s a great proof of concept and might lead to some interesting development in combination with java-webstart and programmers contributing plugins that might be ported.
Thumbs up, great job so far.
It could have potential if you could run in a browser; think of an online version of blender where you could access your projects from anywhere in the world. Time will tell
Wouldn’t this mean all those nice python script Blender comes with wouldn’t work?
“””Wouldn’t this mean all those nice python script Blender comes with wouldn’t work?”””
the java blender could use jython.
As anything other than for a personal learning project, I feel that this endeavor would be a disastrous idea. I, personally, find python + c quite adequate and shudder at the idea of potentially waking up one day to ports of Blender for so many different programming languages advances in one port would get lost in translation of other ports…
Edit: Tried it…didn’t work.
Finally! This is fantastic! C development takes twice as long as Java development (Microsoft itself admitted this more than a decade ago). The world of computer graphics evolves at such a high rate that a Java version will eventually become mandatory to manage the enormous increase in complexity and code bloat of future releases within a shorter development cycle. Computers are getting faster, however the latest version of Java and its highly optimized JVM is gradually approaching the CPU efficiency of C. According to TIOBE.com there are many more Java programmers than C programmers out there. Count me in!
http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html
@John Talbot: Do you have any clue what you’re talking about? Java coding isn’t twice as fast a C coding, it simply lends itself better to dealing with stupid programmers, which in large projects with many developers, adds up to a significant practical benefit.
Java isn’t sufficiently more advanced then C to speed up development much. I’ve seen C code that was much smaller and more maintainable then most Java code. Java doesn’t lend itself well to small, compact codebases as well as C does (and even C isn’t that great in that regard). If you really want the most maintainable, simple code you’d have to switch to a language like Slag, or python, or come up with your own.
Certainly Java isn’t good enough to make switching to it “mandatory.” Especially for performance-critical applications; while Java may be fast these days, it’s memory management is still crap, which simply isn’t acceptable for many applications.
I can sum this up in one word, STUPID! The only possible worse thing that could be done is to rewrite Blender using .NET or any other M$ only language.
But there’s hope here. Jason LaDere can reclaim his dignity if he took on the task of coding a java version of the runtime so that games can be run under java.
And as for Mr. John Talbot who wrote “According to TIOBE.com there are many more Java programmers than C programmers out there. Count me in!”, there is always a reason for everything. There is also more c developers than asm developers. There are more asm developers than computer engineers. There are more computer engineers than physicists. As you can see, there is a pattern here. The lesser intelligent will always outnumber the more intelligent.
If you read his site, he goes over the reasons he’s doing this. It’s not *that* bad, it seems more like a learning project then an attempt to entice all of us devs over to java.
Why do I use jEdit? Because it will be there on any platform with a JRE because it’s Java.
Watch out for that memory management thing mentioned above. Try to make certain that you can develop an application/applet with no memory leaks before proceeding full steam ahead.
The Java Ranch Big Moose Saloon forum may be good to get responses from Java programmers.
Edit: Read your ‘About’ – sounds like you’re very experienced in this area. Good luck.
I will concede that this attempt is a noble one for gaining a better understanding of java and/or Blender. I too find myseld doing wierd things in Java so I can see if it is even possible. As far as usability is concerned, JAVA just isn’t all that great. In fact, all OO languages suffer from the same defective gene; class extensions. JAVA just makes this worse by never reaching the machine code level at compile time. Comparing JAVA to C is like having two runners in a race where the first one is on solid ground and the other is on wet clay. The runner on solid ground will stay clean all the way through while the other will continously gain unnecessary baggage (the clay).
Great work. Please make the Linux version work. Thanks.
it’s sad to see the reaction of some people here, who just try to start a troll over C-vs-java.
Why don’t they just ignore it and focus on their own C-only app, instead of yelling on every roof that java is the Evil incarnated?
Java is simply a hot topic; it’s been over-hyped to hell by Sun, so people tend to argue back and forth over it’s supposedly advantages. Personally, I think it’s an antiquated, out of date language in bad need of revision. All the concepts in it are decades old, and it’s not been significantly improved since they added Generics.
C is hardly a paragon of virtue itself. It’s even more antiquated then java, and is a bit more complex to work with. It’s main advantage is simplicity; C is so simple that it lacks all the overhead associated with most OO languages. It also provides greater control over memory management, which can be very important in many cases. Still, I’m sure it’d be hard to argue which language is “better.”
In my opinion, if you want to choose a language for it’s maintainability factor, choose python. Or Slag, though it’s a fairly new language. Something that has modern OO constructs, like aspect-oriented programming, continuations, etc.
Hmm. It seems crazy. Still, you can compile Java code to your machine’s native language if you really like to (it would suffer in portability though).
I tried it and most functions doesn’t seem to work (maybe because I’m in linux?). Still I like the icons(they are cute and colorful >.<). Speed seems a tad slower (because I'm In linux again?).
I like most the icons. Reall, really like them. C version icons is dull in comparison.
So would this work on PPC linux?
—————————————————————————————————————————-
.
*** C O N G R A T U L A T I O N S ___J A S O N___FOR___M A K I N G___A___J A V A___B L E N D E R ***
.
*** AND ***
*** TO___THE___B L E N D E R___ C O M M U N I T Y___FOR___E N A B L I N G___THIS! A W E S O M E! ***
—————————————————————————————————————————-
.
This has been discussed over the years by a few people as a potential project.
.
It was inevitable that someone would ‘just do it’ sometime.
.
.
::::::::::::::::::::::::::::::::::: YOU, Jason, have made personal and Bender HISTORY! :::::::::::::::::::::::::::
.
.
But your success stands on the shoulders of NaN, Ton, The Blender Foundation, the user community, etc.
———————————————————————————————————————
.
.
So, please let me please write a note of appeal to you, the Blender Foundation, and the community.
.
——————-
.
Java may not appear to many who are enamoured by the efficiency of C to be the best choice of language for a production speed 3D app.
.
But the comments in this forum and in Jason’s blog make the same points that illustrate why his project should be considered as an OFFICIAL Blender Foundation Project – NOW – and not before Jason’s project gets runaway momentum:
.
1. There’s a larger base of active Java developers than C developers – from uni students, to senior coders, and academics;
.
2. It will be easier for existing Java coders and those from other languages to learn Blender’s architecture than learn C;
.
3. A Java version will have unique benefits of use eg run in a browser or mobile platform.
.
.
Yes, there are obvious issues to be addressed as have been mentioned in this forum, perhaps the most important is the forking of functionality.
.
But, guided right, this doesn’t have to be a problem. For example, a web based Java version of Blender perhaps should have have different functionaltiy to a fuller client based app written in C. Or as bandwidth and CPU power, etc, increase, they may be/become identical. Let’s thinks about this and make some goals!
.
It would be very easy to knock Jason for his timing – pre 2.5, but he has obviously committed his own time to this project and as he states on his blog, has good intentions from wanting to learn Blenders architecture, to becoming a committer to the C Blender project, to helping others who know Java better do the same.
.
Many many open source projects have gone through the same historic forking, and it is the ones that don’t shy from this, but embrace it maturely ie keep the communication open, that survive by growing from the strength of each other rather than wittling away each others user bases with unnecessary language wars.
.
This is a challenge to the Blender Foundation and its community to be as open as possible, and a challenge to you Jason, to be mature in aligning your project with the Foundation too.
.
.Net has been mocked above (and perhaps rightly so), but there may be some users who could use a Mono compatible version of Blender someday?
.
So, don’t think shouting down Jason is going to stop him or anyone else porting Blender to other languages, runtimes, etc.
.
Would someone be ‘crazy’ enough to try to make a pure Python version? Would some of the detractors of Java here be as harsh on such an ‘experiment’?
.
.
.
Jason has opened historic floodgates that can’t be closed now.
.
.
::::::::::::::::::::::::It is arguable that this is the test of Blender truly being ‘open source’.:::::::::::::::::::::::
.
.
Let’s not collectively blow it, but make it a success.
.
.
.
I suggest Jason and the Blender Foundation quickly meet/discuss how to turn this into an official project that:
.
.
1. Works to simlar UI etc guidelines as the C project;
.
2. Develops some unique goals for the Java project and own timeline eg web browser use;
.
3. Synchronises the benefits between each project eg teaching developers C/Java, porting code bi-directionally, identifying better algorithms, test UI conventions, etc, etc.
.
.
Then, this will not deplete from the C project, and let 2.5 get to ‘market’/community without interuption.
.
But perhaps the world will see a 2.5 version of the Java project out around the same time for mobile or whatever?
.
.
Why?
.
.
Because although Blender may become the poster-child of open source media apps, a Java version is more likely to become the poster-child of Blender’s open source nature, being that it will be more easy to read the code.
.
Don’t believe me?
.
Best example(?) :More universities will play with the Java than the C version, integrating it into more course assignments.
.
.
If the Blender community, including the Blender Foundation fails to do 1 – 3 above, you can expect some of the bad things mentioned in the posts above to become worse in reality, and another open source project (Blender) being f@@ked (I mean forked ;-) .
.
.
So I APPEAL to everyone concerned, please learn from the history of Unix to many open source projects today – embrace this new project as a sibling – don’t outcast it, but invite it into the home of the community to strengthen it so that it may strengthen Blender in return.
.
.
You’ll be surprised by the benefits a Java version will bring, besides many many more code committers.
.
(For the ‘record, I’m not a big fan of Java as a language, but it’s wider benefits/popularity over C are obvious.)
.
Let’s not kill Blender’s future by trying to kill history.
.
.
And, Jason, please meet the responsibility you have created by opening your project by trying to integrate it as an OFFICIAL Blender Foundation Project, so that the benefits or your work, which stands on the shoulders of the BF, may continue to grow with Blender into the future.
.
.
And, once again… CONGRATULATIONS to Jason, the BF, and Community – a truly AWE INSPIRING DEVELOPMENT!!!
Oops, I meant…
Blender HISTORY not ‘Bender HISTORY’.
Nothing implied!
;-)
This is effort that could have been better spent on improving Blender’s current code.
A port someday to PyPy if it can offer speed surpassing hand optimized C makes sense since part of Blender is already Python and we can slowly transition to that as the PyPy project and it’s JIT matures.
I give it a try and I think it’s a Good work! :O
Keep working :)
@Goathead:
No, but no! Stop saying that’s a waste of effort, even if it’s because you subjectively dislike a language. Jason said himself that he’s learning Blender through this port. If he didn’t how the heck could he improve the current code? By adding randomly tons of lines without understanding what’s he’s doing? Be serious.
Sorry goathead,
but as easy as it may be to write that Jason’s effort ‘could have been better spent on improving Blender’s current code’, if you read his blog, he stated that he did the Java port so that he could better understand Blender’s use of C.
So, in light of what Jason has written, such points as yours (no offense), are not valid.
He was better to spend his time learning C by porting Blender to Java, and then if he chooses to, he can contribute to the C version – but as a better C coder.
And the benefits of a bi-product Java version of Blender – significantly outweigh what he may have been able to contribute to the Blender C code base in respect to his C coding skills at the time of porting.
So… let’s stop the trivial arguments against this, wishful thinking – ‘I wish he had ported to bla language’, and embrace this HISTORIC development.
If you or others wish to port to PyPy, then do – no one will be able to stop you – but forum posters negative comments (like yours) may directly or indirectly attempt to.
In the meantime, please don’t try to stop the energy, benefits, and pure awesomeness that will be Blender in another language, one that most Computer Science students to industry Java coders will be able to immediately jump into learning.
If embraced properly, Jasons ‘spent time’ will only help others learn how to build ports to other languages and how to work with the BF, including people like you in the community.
Embrace, don’t disgrace.
( sorry gokudomatic, I think I just wrote the verbose Java version of the C version of the same ideas as in your post – you beat me to post reply to goathead while I was writing mine. )
Lol! I hope Jason isn’t actually reading all this nonsense. If you get a chance what would awesome is an alternative way to create standalone blender games, perhaps java? If you could do that I would have your babies. Anyways, good luck man! Have fun.
Any full-fledged programmer should be able to pick up C (or any other procedural language) on the fly; that’s part of a software developer’s job. I’ve never met any pure “java developers”, who apparently can’t think outside the java box and whose disability to learn any other language we should pander too. Do they even exist?
And please, no uber hyping. There will be no production-usable java port of blender; it’s simply too much work. I know this from personal experience from the bmesh project, and that’s just a small piece of blender. This seems like more of an exploratory project, to see what’s possible. Personally, I think a java port of the game engine would make sense (BGE web applet, anyone?), and I imagine there are other parts of blender that might be useful in e.g. phone apps, web applets, whatever. But a full-blown port just wouldn’t make sense. It’d take a team of programmers months to do, anyway.
Lol blndrusr, you always manage to entertain with such vibrant ideas. :p
“Any full-fledged programmer should be able to pick up C on the fly; that’s part of a software developer’s job.”
yeah, yeah, sure. programming == C, eh?
@gokudomatic: thanks for putting words in my mouth. that’s not what I meant at all; anyone should be able to pick up java too, or php, or C#, or whatever language you need for what it is your doing. programming is more about the problems your trying to solve, not the language.
Wow. I’m impressed.
I tried the webstart and it loaded pretty fast, then again I’m on a broadband connection. However, it’s nice to see you managed to get this far.
I’m quite surprised by many of the comments. I mean who cares if he makes a port of this to Java or not? You may think the developer is wasting his time but the thing to remember is that it’s his time, not yours. To me it’s better that he be coding something and contributing to the freedom, open source spirit, and experimentation of the Blender code base than not do anything at all. This seems like another thing Blender can brag about in terms of the projects it’s inspired and the community supporting it. Porting Blender to Java is not an easy task so I think this guy deserves props for at least getting this far.
Besides Blender is still being developed in C and will still continue to be developed in C, read the Blender forums and the devs have stated this many times.
Bare least you should say is nice job good luck on your project.
Anyway, let me get off my soapbox.
Nice job, Jason LaDere and good luck to you and your project.
Some of you guys crack me up with your degree of anti-everything-not-invented-here bullshit, ignorance and insecurity.
I’m a 20-year C, C++, ObjectiveC, ADA… (long list) programmer and I still much prefer Java for it’s hygiene over any of the closer C derivatives. The guy isn’t reimplementing because he is suggesting Java is better – just more useful for his purpose.
There are definitely more Java noobs out there than in non-GC languages but that is more of a reflection on Java being an excellent language for teaching OO principles and it consequent popularity with University course developers than on the language’s quality or suitability for a given purpose.
Many of the posters have alluded to poor performance – pure bullshit. Well optimised Java compilers produce native code that generally outperforms C++ and gets very close to C performance. The only place you’ll notice any real difference in speed is in areas of code where you would normally use asm inline – such as matrix-quaternion conversions and the like. Java is MUCH faster than the Python interpreter anyway.
To joeedh – compact code != clean, self documenting and therefore easily maintained code – read some books on software engineering. Verbosity in Java is syntactic only – it does not translate into verbose bytecode. On the contrary the HotSpot JVM has received so much tuning attention that you’d have to be very good to do better by hand.
Obviously OpenGL is a very procedurally oriented framework, and the JOGL library does not attempt to create an OO layer of abstraction – this is why JOGL is surprisingly fast, calling directly through to native code. But there are some areas, particularly in user interface design, where an OO paradigm is far more convenient, intuitive, modular and maintainable.
Java lets you have both with a modest penalty of garbage collection – something many experienced programmer f%&k up incidentally.
I use C because I have to – I use Java because I prefer to. I don’t use Python because it does not offer me any compelling reason through features I need which I don’t already have implemented with more maturity and thorough testing.
Kudos Jason. Nice work.
I don’t see for the moment the use of this version but we have to wait, to see the quality of the code at the end of the project and compare the performances with the official blender.
I am a little scared when we don’t know what will happen with the repurchase of Oracle about Sun.
Hmm… Modelling on my cellphone on the way to and from work with a Bluetooth mouse, then BT it to the Pc for rendering… :)
@ AD-Edge, well, at least they entertain someone! =)
But seriously…
My main ideas is this:
Blender is an IDEA, ultimately NOT tied to ***1*** language, but can only improve as a consistent artist-driven philosophy IF it is allowed to be tested in different guises, languages, etc.
—————————————————————————
.
Please make BlenderJava an Official Blender Foundation Project
.
—————————————————————————
… including having its own pages on the official Blender web site.
The only genuine criticism floating in this forum is the amount of work needed to complete it.
But from reading the more positive comments on Jason’s blog, and knowing how popular Java is in uni’s and workplaces, it’s obvious that there is enough Java love out there for a contributor community to form quickly and sustainably around it.
The main point of my verbose ‘vibrant ideas’ (hehe) is that BlenderJava is happening with or *without* this community, and to avoid obvious forking issues and prepare for other inevitable language ports (and attempts), the BF and this community should embrace BlenderJava *officially*.
This doesn’t mean giving up developers time or priorities on the C version or anything silly, but simply officially acknowledging it, scheduling some meeting time with Ton and crew to ink copy/paste/modify some guidelines/goals eg the unique features it would have ie for web, mobile or university course study/dev, give them some admin access on a wiki, and let Jason and his Javanauts get on with helping Blender become the dominant 3D app by propagating it to another language – a very popular one.
And the comments defending the speed of Java are valid – it’s come a long way – and is good enough for ‘the job’.
There’s no harm in having ‘sub-Tons’, holding the fort for heading the development of other Blender projects. We have this already. And it seems we have a new one, Jason for BlenderJava. :D For the long-term growth of Blender, the development of nascent coder leadership is as good as the development of new code. It strengthens the community – if we embrace the diversity – not disgrace it.
Whether Jason spearheads BlenderJava into the future or not remains to be seen, but it will be cool when the main Blender page shows Ton is impressed enough by this nubile project that he grants it some web space and finds a bit of spare time in his day (damn he must be crazy busy right now) to help it fuse as a real BF Project around some official goals, guidelines etc. I hope he gets a chance to even read the news item this forum thread is attached to.
As is said, ‘the most sincere form of flattery is mimicry’, and BlenderJava is at least this at this time and potentially a lot more.
It could develop into something that brings unique benefits to this community and helps Blender move into new user spaces eg mobile, while interoperating with blender_C.
There is cool news happening daily now in the world of Blender, which rocks!
.
.
::::::::::::::::: :::::But how often does it get semi-functionally ported to another language??? ::::::::::::::::::::::::
.
.
Let’s make it stick, grow, and feedback to help the C version. There are SO MANY MANY ways that it could.
We’re all great at using our imaginations to think of animation stories etc, but it seems we ignore them when thinking of the most basic marketing/operational/community-development ideas.
.
.
Jason has handed this community a GIFT not a curse.
.
.
My last last point is this – BlenderJava is a sign on the maturity of Blender’s code and community, is inevitable, and is going to be WAY BIGGER than even I can imagine.
I’d better stop here> all I wanted to say is in previous posts. This was meant to be a summary. Ooops. =D
(Yes, I am aiming for shorter posts, sorry ;-)
As for programming practice this is might be good project, but for real usage its just not needed when we have original version that is faster, up to date and runs on many platforms.
Yeah, i’ve seen benchmarks where java is even faster than c with runtime optimizations, but all programs that I have tested are slower.
@Dread Knight
That is one area I would really be interested in seeing working.
I don’t see the rationale behind this port, except for personal improvement for the author. Blender is small and fast as it is in C++. Portability is also no problem at the moment. Moreover, Java is slow and memory hungry which makes it unsuitable for large projects.
A wxPython or wxWidgets C++ port of Blender would have been somewhat more logical, as many users are put off by its non-standard user interface. I initially was put off by it as well, but over time I’ve came to appreciate it. For example, Blender looks exactly the same on Linux as it does on Windows. Also, the OpenGL UI is scalable which native UI’s aren’t.
Conclusion: Blender is IMHO perfect as it is. No need for a Java port.
@Dave W:
You’ve loosed 20 of your years… I CAN produce any C code with equal or more speed than ANY Java/C++… without writing any lines of code!
Because it’s just how Java/C++ are designed, that’s all!
@all:
Seriously, please stop to argue about which programming languages is better, faster, stronger, w+er…
Just respect the design, the purpose and the hierarchy of them.
Java/C++/Python/… are HIGH LEVEL PROGRAMMING LANGUAGE
C is MEDIAN LEVEL
ASM = LOW LEVEL
– Lesser is the level, higher is the coding time for a task, higher is the risk of bug, stronger should be the skills for the maintenance.
– Higher is the level, lesser is the coding time for a task, lesser is the risk of bug, blablabla…
And why these rules are? Because 99% of CPUs work using a machine language, not a higher one.
Sure it exists now some chipset with JVM integrated in HW, but not used in public computer.
But now, we’re falling in an argue of using specialized chipsets or not.
So now, about the port itself:
– it’s not a 100% stupidity thing: this port can be useful for private system w/o C SDK for exemple, but these systems have not a goal to run a 3D modeler/renderer software.
– Clouding Computer? (current popular concept…) Yes, maybe! but limited to the popularity of the concept.
You see: the audience of this port is not the one of the current Blender. It’s normal that many of current users don’t see the benefit.
So this port has a benefit but please don’t try to increase it at the same level of the current C implementation… this last works, very well, performs well also for the generation and type of CPUs used, and it’s strongly portable, even in Java :-D ! (I’ve made the port of Blender/Python on MorphOS, a non-Unix/PowerPC platform… on which there is no JVM port!)
Just a note about one of the first post: a Java program is not more portable than C program, just because a C compiler can be found for 99.9999% of any CPU manufactured in this world.
It’s like the concept of ‘maturity’ of a code just because its written in such or such language… LOL :-)
Depends on which church you pray, that’s all!
/me a 24 years programming XP guy :-P
Polling: How many of you want a BlitzBasic2 port?
Go Jason go! If you succeed then maybe you will get interested in recoding Carrara 3D to Java too?
I have no Idea why you guys are even taking the time to formulate such long posts regarding this.
You aren’t going to get a fully featured Java port of Blender, so get that idea out of your heads. It’s too impractical. Joeedh made the most valid point in this whole discussion, and he’s being jumped on for some reason.
Sure, it’s interesting as someones learning project, but nothing more. In terms of sheer practicality, you might as well ask a cheese sandwich to make your dinner for you.
Dave.
I am a web developer, and currently working on a blender model repository. (istvan.us/bmr – unstyled, and semi-functional). I can tell you that having a java applet that works enough to show a scene, not edit it, but just enough to see the models in 3d would be awesome. This would be great on forums too, a feature to upload a .blend file and show other people your WIP without them having to download the file. One small step that could make it even more useful would be grease pencil. Instead of the huge posts trying to explain which part of the model they are talking about, each user creates a grease pencil layer, and can mark it up all they want. I see that as being really useful
However, I see little to no use in having a version for actually making/editing blender files in java, as it is faster and already extraordinarly cross platform as a binary. If some people write features for the java blender, then they won’t make it to the ‘real’ blender. Also maintaining the java version with all of the features would be a pain, as there are so many SVN commits that patches would need to be ported by the tens or hundreds.
In short, I think this could be useful to fill the gap that blender has – the internet. As for working on files, I see no advantage, and a few dis-advantages.
~Best of luck, and I can’t wait to see how it turns out, Istvan.
Blndrusr said:
“My main ideas is this:
Blender is an IDEA, ultimately NOT tied to ***1*** language, but can only improve as a consistent artist-driven philosophy IF it is allowed to be tested in different guises, languages, etc.”
I think that’s a really dumb idea. maintaining multiple versions of blender, all coded in different languages would be a colossal task. It would also significantly impact development speed as well. Think of the logistics of communicating between multiple teams, and how much extra time and effort that would require.
And the net benefit for the end user? Very little.
@ Istvan
That would be sweet!
This is an awesome idea since a web-started java app would load the program instantly from any computer (with the java runtime environment installed), if it lagged in performance you could always download the executable, but having it like this would probably open Blender up to a wider audience.
@Dave saying “You aren’t going to get a fully featured Java port of Blender” is like saying you are going to get a full feature 3D animation package for free….. It is absolutely possible, and imagine the potential of high level programming classes in universities all over the world working on the code in java…. adding new features and then the development team ports those new features back into the main branch in C. Jason congrats on having the iron to take on such a huge project I wish you the best of luck. Long live open source.
@Dave W: Of course compact code by itself isn’t self-documenting or maintainable by itself. Making it so doesn’t take *that* much work, however. And compact code is much easier to read and figure out what it does then verbose code (by far), *if* it’s commented correctly.
Of course I’m different then most people; I read code quite a bit faster (it’s more like scanning code) so if the code is too verbose it takes forever to figure out what it does, and drives me crazy. The “ideal” code for me is compact, has enough comments to be clear, etc. It’s all about being as clear as possible without being too verbose within the code structure itself, which would slow me down quite a bit.
Wow, awesome.
How about an eclipse RCP application build with the plug-in paradigma (OSGi components)?
Cheers
Boessu
@Dave Parsons – Blender isn’t currently *fully* constructed from 1 language now.
…………………..You totally missed my point – Blender is about a useability philosophy not 1 language.
@Jeromew – that’s been my point too! :D
……………Universities will pick up BlenderJava whether the BF and community here accept it or not.
Hundreds if not thousands of students/academics will become partially aquainted with BlenderJava’s code, and some will decide to become contributors to its ongoing development.
So the BF should make this an OFFICIAL BF Project.
Just give it graphical link on main Blender site, some wiki space etc, BUT let Jason find the coders and do the heavy lifting.
And if done right, blender_C will see the benefits flow back to it while probably still being developmentally in the lead for a long time.
It’s not a race and one project will not deplete the other – but if set up right now at the beginning – they will compliment and strenghten each other.
@Yomgui – Lol re ‘Polling: How many of you want a BlitzBasic2 port?’.
…………..But seriously, a Java port was inevitable, it’s been raised so long ago, back in the NaN forum(?) – you can’t just ask ‘please don’t try to increase it at the same level of the current C implementation’ – Jason and/or others will attempt to do so anyway…
the cat’s already out of the bag!
So let’s just embrace BlenderJava, give it some growing space (wiki, etc), and see what awesome benefits emerge!
right. don’t try to resist us, you will be assimilated. :D
@blndrusr
I didn’t miss that point, I just ignored it. The reason for that is because I just don’t see how having a Java port is any way related to Blender’s usability/design philosophy.
I am aware that blender is not written entirely in one language; But creating and maintaining a totally separate port is completely different to what the devs are currently doing.
blndrusr: “Universities will pick up BlenderJava whether the BF and community here accept it or not.
Hundreds if not thousands of students/academics will become partially aquainted with BlenderJava’s code, and some will decide to become contributors to its ongoing development.”
I agree, they sure will pick it up and turn blenderjava into perfect gem, just like they did with java port of umm, eh, is there any popular software coded with java?
Guess im postin in troll thread
@juri:
you want an example of java gem? netbeans. too bad it wasn’t picked up by students, eh?
@gokudomatic = ‘assimilated’ as you write is OTT and immature. My point is obvious – BlenderJava will grow *without you, the BF, etc – but it’s best for all if it everyone is included.
.
Lol, re your ‘netbeans’ and uni students comment. I ran the BlenderJava idea past comp sci grads in wkend who mostly studied Java, and all were interested in at least looking at the code, if not contributing. Some have an interest in 3D and Blender, but the commonaility was their interest in graphics code and a decent Java media app – very different from ‘netbeans’ that they have only ever cared for in a paid working environment. Chalk and cheese.
.
Keeping pushing the negativity. If it fails, return to my first point – your negativity is irrelvant – there’ll be HUGE interest across many new and old developer communites in a BlenderJava, and quite obviously for different ‘reasons’ than you are imagining. Try to work on that.
.
.
.
@Dave Parsons = you may selectively read and ignore any points I make, it doesn’t validate your replies.
.
So to REPEAT the point I made several times in earlier posts, in reply to a point of mine you IGNORED:
.
‘Just give it graphical link on main Blender site, some wiki space etc, BUT let Jason find the coders and do the heavy lifting.’
.
Yes, ‘creating and maintaining a totally separate port is completely different to what the devs are currently doing’ – no kidding – and *not* my point.
.
MY POINT: the Blender_C dev’s are not expected by Jason (***read his blog and not selectively***) to work on his port.
.
BlenderJava will attract its *own* dev base and have its own unique uses, some of which are likely to overlap with Blender_C, and if you know anything about coding, this overlap can only be a good thing as it will highlight better algorithms etc. And the differences will be what makes each port worth pursuing eg BlenderJava for web/mobile/educational benefits.
.
.
.
If my replies here still don’t make sense, perhaps the following will:
.
@juri = a few individual uni students have extended parts of Blender over the years, but only a few, and that’s no doubt because the code base, tbh, was rather difficult to work with and less students know C as well as they do Java.
Even with a ‘clearner’ 2.5 code base, Blender_C will still be beyond the interest and possibiy the skills of many students.
.
And from experience, those I’ve asked, and the overwhelming positive response on Jasons blog – the problem is there haven”t been any good multimedia, in particular 3D Java apps with sufficient popularity, for students and academia to care for.
But it’s apparent that BlenderJava will change that.
.
If you don’t get this, then perhaps you were the type of person who wouldn’t have wanted the game engine in Blender, which had a LOT of opposition but now has Apricot and healthy future.
.
For many I know who are interested in BlenderJava, the difference is as obvious as the historical translation of the bible from Latin to a common language making the bible overwhelmingly more accessible and popular.
.
It’s all very well to have a popular and ‘accessible’ open source Blender for artists – but it will be a LOT more accessible to developeres when its code is more easily readable and extendable, and a Java version is this compared to a C version.
.
.
So to summarise for the historical ignorant, imaginatively challenged, and those who selectively read posts:
.
1. BlenderJava (like Blender’s game engine) is already growing without you. But I hope it doesn’t.
.
2. I hope it grows with the support of you, the community, and the BF so it can compliment the Blender_C version;
.
3. It *only* needs a front page link to a few pages on the main site, wiki access, and some guideline/goal notes;
.
Let me repeat this for selective readers – Jason will easily find his own developers.
.
4. Whether or not students pick it up or not, many grad’s I have talked with are keen on BlenderJava. The ‘market’/community exists that has been waiting for an accessible 3D app ie in Java that they can learn/extend/use. It is more likely to be picked up by students than the existing Blender_C version. But its development will not be dependent on students. This is just one of the communites that wil pick it up.
.
5. In no way will BlenderJava negatively impact the Blender_C version, but on the contrary, it can only help it, the BF, and this community, just as the game engine has.
.
6. Don’t take my word for it – read Jason’s blog and follow its growth!
blndrusr: “BlenderJava” isn’t going to be at all what you think it is, so please stop preaching. Whatever uses it’s going to find, it’s not going to be a simple clone of blender. There’s just no point in that. There’s other ideas that make more sense, like cloning specific parts that would be useful in web apps or whatever.
+1 @joeedh
If the goal is simply to make blender in java, then there are no upsides, and many downsides. For any use that java blender has, the official blender can do too, but faster. Sure it can run in a browser….. but who cares…. just download a binary. On the other hand, like I was saying before it can be used to fill the gaps between. My example was letting people create grease pencil layers on a file in forums. These would be great to illustrate people’s points in forums, and is a task that is specific to the web. It is for these interfaces, from web pages to animation studio that a java version of blender is quite useful. In replacing blender I see no advantage. The only place that the official blender is at all shaky is for the web.
@joeedh and Istvan = again, if you actually read what I wrote, you are agreeing with me, lol:
.
.
‘And the differences will be what makes each port worth pursuing eg BlenderJava for web/mobile/educational benefits.’
.
.
I have been writing a variant on this in most of my posts and Jason himself states similar possible goals in his blog.
.
.
.
My main points, if you didn’t read them… have simply been…
1. The BlenderJava project cannot just be dismissed – Jason is working on it anyway;
.
2. It’ll keep progressing SO it should become OFFICIALLY supported by the BF eg get recognition/graphical-link on http://www.blender.org/ to some pages for Jason and BF to outline guidelines and shared + unique goals, and some wiki space. BUT let Jason find the coders and ‘do the heavy lifting’ so no negative impact on Blender_C.
.
3. We as a community should be positive about this project, whether Jason continues working to a full port first or not – unique benefits will come. But they will happen a LOT quicker if the community throws its negativity aside.
I don’t want to bring up the past… but don’t forget there was a time when Ton didn’t want to open source Blender and there was paid version, a LOT of people were against a game engine in Blender, a modern rewrite of Blender was delayed and delayed and delayed sometimes for no good reason or just because of poor historical decisions… yet now Blender is open sourced, a rewrite has been happening nicely towards version 2.5, and the game engine is used in a version of Apricot.
Many ideas ‘the community’ initially thought bad (or *didn’t think through*) have made it into Blender, against a tide of negativity by a few people, and utimately have helped improve Blender and the community.
So rather than be specious about Jason’s goals… I’m simply saying we should show him encouragement to continue and if we can get to point 2) above – make it an OFFICIALLY recognised BF Project, then we can all be contributors to its goals, and make them compliment Blender_C, whether its web/mobile based editor/player or whatever.
The alternative is that Jason will continue doing what he’s doing now, open source it, and BlenderJava may fork from Blender_C in a non-complimentary manner.
No amount of negativity in this forum will help prevent this – Jason never asked for our opinions when he started the project and is likely to keep going without us.
Anyone who says he is wasting his time isn’t thinking – it’s HIS time – NOT YOURS – and he probably is contributing more to Blender’s popularity and future growth pursuing what you think is a silly project than most here have (or will) to this community other than download a free app or video or use free bandwidth on this site.
The BEST contribution most here can make is to review their negativity and try to get Jason’s project to point 2) above – get it recognised as an OFFICIAL BF Project so that its goals can be synchronised with the Blender_C version, compliment it, and also be partially unique so that it makes Blender more accessible, for example on different platforms.
Please, before you post in reply to this or any of my other posts on this topic – READ them. You may find that I agree with you, at least in part.
But until we all agree that BlenderJava should become an OFFICIALLY recognised BF Project, then we are all ‘pissing in the wind’ as Jason is free to do what he wants with BlenderJava and our words, whether rants or raves about it, are wasted.
I agree to most of your comments. In fact a little poking around will turn up an old blender project for running blender online. This project is no longer maintained, but was official. I agree that it could become official, and that it would possibly be productive. My only concern is that it might require a separate bug tracker, and a lot of time updating the Java blender to keep up with the C blender. The C blender gets many updates every day, and all of them would need to be ported to the Java version. Weighing the cost-benefit here, I would say that it makes sense to focus the java project to target what it will be useful for. If this project is too big, and someone bites off more they can chew, many updates to port to java every day, the project might choke. I would hate to see that happen, because as I said, there are many holes that this would be great to use as a patch.
blnduser: making this project “official” would be stupid; it’s not a major project yet, it’s simply one guy’s fun project. taking official cognizance is something that can happen after the project starts to find a direction. what your suggesting is more like artificially generating hype, by making an “official, BF-‘blessed’ project” when in actuality it’s far too early to be talking about that.
Besides, I don’t see the point in doing that, except maybe to counteract other people’s negativity (which is a stupid reason to do anything). Open source is much less formal then this; by creating early associations you artificially drum up expectations, putting extra pressure on people for no reason. It’d be like if Ton made a press release announcing the bmesh project six months ago or so; it would have been immensely stressful to have such official cognizance (or the time an author mentioned another project of mine in his book, that wasn’t any fun).
Honestly, I don’t know why everything Java-related has to be hyped beyond all reason, but I think Jason and everyone else would just prefer to take things as they come, go with the flow, and not make a huge PR effort out of it.
@istvan = you make good points, and I’m partially in agreement as I’ve noted in previous posts that Jason’s goal of BlenderJava not being a full port but being adapted for a different purpose may be the best way to maintain the BlenderJava project.
If he were to try to maintain a full port on par with Blender_C, yes – you are right – there would be a LOT of daily updates, bug tracking, etc issues for Jason and any team that he assemles to get on top of.
.
But the way I imagined he would do this, is delay the lauch of BlenderJava by, for example, 6 months behind Blender_C, and with each release playing catch up a bit more to the Blender_C version. This way daily updates and bug tracking can be ignored right up to the next release of Blender_C, which the BlenderJava project can then port using automation and the team Jason would put together.
.
I know the time cycles are different, but there’s a some similarity here between Debian and Ubuntu. Each is unique but Ubuntu borrrows from Debian, and both promote Linux. Perhaps a better analogue would be how FreeBSD maintains old releases while still pushing out new ones?
.
………………………………………………………………………………………………………………………………………………..
If Jason continues to aim for full ports and gets these released, then there are many ways he could creatively synchronise releases of BlenderJava with Blender_C, and eventually if he chose to, differentiate its functionality, perhaps even porting new features upstream to the Blender_C version once development cycles are stable.
But I’m cool even if he just turns it into a player or whatever as the benefits of some of Blender_C being in a more accessible language eg Java, will outweigh the time costs.
…………………………………………………………………………………………………………………………………………………
.
.
@joeedh = my point regarding BlenderJava becoming official is really to help Jason find goals for BlenderJava that are compatible with the BF and Blender_C. I agree that it seems premature to make it an official project now – yet his announcement on his blog is ‘official’ enough for anyone who reads it – he’s started a project with or without the BF behind it. It’s offically a Java version of Blender just not an official BF project.
.
So rather than let Jason chew off too much as istvan makes a good point concerning, I’m suggesting the BF simply helps Jason build some workable goals, guidelines, and gives him some wikispace etc.
Sometimes a bit of official support actually makes the difference in an individual or teams motivation to continue through with a project. He has got quite far with this project – more than most would initially believe – BlenderJava actually works in part – if that official enough then what is? As I wrote above, it’s just not an official BF project yet, and for obvious ‘PR’ reasons and all the more substantial ones (sharing knowledge etc) it should become one.
.
In reply to your ‘hype’ related comments – my points have nothing to do with trying to hype BlenderJava beyond what it is – but just give it the right space to grow to be useful to the community without disappearing from a lack of love. And in reply to your comment about ‘everything Java-related…[being]…hyped beyond all reason’ – please don’t let your own bias cloud the real benefits of BlenderJava – its obvious that to many developers of this and newer generations, that its code will be more readable to them and thus more accessible and easier to work with.
.
As I’ve written in an earlier post (perhaps my first above), I’m not a big fan of Java – I find it verbose, but it’s less cryptic to me and many others than C. Perhaps many things Java-related are over-hyped, but often there is some truth behind hype eg the hyped ‘object’ is better than the one people don’t hype tho perhaps not greatly – but just ‘better’, and hype itself can push technologies into the mainstream. I remember hype helping people jump on the ‘net in the early days. If hype helps people learn the architecture of Blender_C or pick up 3D by first using BlenderJava, and if this was the only way they were going to arrive at that end point, then awesome – new doors are opened for them and new talent is introduced to this community! :D
.
………………………………………………………………………………………………………………………………………………….
Whether I’m right or wrong regarding BlenderJava becoming an officially ‘recognised’ project now – my baseline point is that the BF should reach out to Jason and help him work in the background to make it one in the future – encourage him, give him some basic resource access eg wiki/bug tracker, slowly work out some goals/guidelines that are good for both, and let him get on with what he appears to enjoy. But apps like Redmine are easy to deploy and he may install it on a host himself and just get on with remote managing a growing base of Java developers?
………………………………………………………………………………………………………………………………………………….
.
I’m sure Jason would just like to go with the flow, but who knows, we have all seen developers covered in news here with projects that they have worked on that have never made it into new Blender builds and they have been slowly left to drop off the edges of the community. If my pushy posts seem a bit OTT, it’s simply to stop this happening. As I’ve written from the first post – it’s Jason’s project not ours – but history tells us that he has a better chance of finishing and maintaining what he has set out to do if given some official support and this increases the probability of this being something that is good for the whole Blender community too.
.
And it would appear that his project/he is at a crossroads. No doubt it has been ‘fun’ up to now and he would like to keep it that way as all projects should be fun, but it seems that he is ‘serious’ enough about taking his project to the ‘next step’ that he has blogged about, it has become news here, and he would like some feedback from the community, as any next move on his part could cost him a lot more time and possibly lose some of the ‘fun factor’. Losing the fun factor is often the first sign that a project will die.
.
And unlike your ‘go with the flow’ – leave BlenderJava to chance approach, I believe it merits a sustained development environment, and of course, one that is still fun. But one that is less likely to fail.
.
Once you take off your ‘all Java-related things are overhyped’ blinkers – a Java version of Blender and Jason’s efforts deserve some recognition – enough to not be lost in the archives of Blender Nation – but helped to continue the fun he is having and the benefits this will bring many developers out there interested in media apps and working in media but that are more accustomed to Java than C.
.
So forgive me for seeming OTT, but someday you will see that BlenderJava is an HISTORIC achievement. I hope Jason continues to ‘go with the flow’ and develop it, but as I keep repeating, he seems at a crossroads where he’s asking if there’s enough community love to take it to the next step and dedicate some real work to do so. I hope the fun continues, but history tells us that it probably only will if his project is given the proper attention by this community and the BF, whether ‘front-page hype’ or not.
.
I’m not suggesting a ‘huge PR effort’, just that the BF recognise Jason’s efforts and start building some simple support bridges for his project to continue in the same fun spirit to prevent it failing as many other projects, including that which istvan mentioned.
.
And if something great happens from this, then yes, hype it! If it just happens to use Java, so what?!
@blnduser: have you ever considered is Jason *wants* any such thing? besides, you realize that in general, the BF doesn’t “officially” recognize *any* projects, except for a few cases involving collaboration with other projects (and had financial components)?
There’s nothing stopping Jason from having wiki space, or requesting project hosting on projects.blender.org (lots of people have done that). I still don’t see the point of (again, I use this word) generating tons of hype and jumping on the java bandwagon, just to, what? Keep the project from disappearing? Don’t you mean pressuring Jason to not give up on it? I have plenty of experience with that sort of attitude myself, and it’s not always fun or easy to deal with.
@joeedh =
.
1. Of course I have considered whether Jason wants his project official or not… and the added pressure… but it appears he is cool with publishing details about BlenderJava on his web site and inviting replies to his question of whether there would be interest in BlenderJava development continuing.
.
So I think it is fair to ‘push’ a little for it to become ‘official’, this way protecting his project and helping give it compatible goals with Blender_C and the BF. That’s only reason I am pushy here.
.
Jason has invited opinion… and this is mine. If he decides to give up after reading negative opinions or release BlenderJava as his own project, then I believe this won’t be the optimal outcome. But again, it’s just my opinion, and Jason has invited all to contribute their feedback. He’s free to do whatever he likes. I doubt he would bend to my opinion. =)
.
2. Sure I ‘realize that in general, the BF doesn’t “officially” recognize *any* projects, except for a few cases involving collaboration with other projects’. But what’s so different about BlenderJava from other projects that it can’t be considered a project worth collabourating with? And even if it is very different – why can’t history change???
.
At one time the BF didn’t exist and Ton’s goal was for Blender to never be open sourced. What’s changed? LOTS. So why can’t the BF change how it recognises projects? The nature of open source is that more Blender derived projects will pop up. They shouldn’t all be official as soon as they pop up, but after meeting some basic criteria.
.
Community projects officially recognised by the BF may have the following criteria: a good stage of development, good impact on Blender_C/BF, growing/potential community support, etc… and it appears Jason’s project meets these criteria (I have made up ;-) . But they are not too far from what ‘real criteria’ for BF projects could be… and if such criteria don’t exist, then BlenderJava is a good test case for them to exist, because as I’ve written – more derived works will pop up.
.
Whatever you think of Java, this is an historical moment in Blender’s open source evolution – a partially useable version in another popular production langauge. If my ‘hype’ inversely mirrors your pooh poohing of Jason’s project, then ,if that’s what I have to do to change history, I’ll do it. But I don’t believe I’m hyping his project – just pointing out its merits to this community, which would become a lot more obvious if the project ‘gets legs’.
As I wrote in reply to your first point – Jason seems cool doing what he’s doing but *wants some feedback*. Due to the ‘historical’ nature of what he is doing, and it meeting the rather agreeable criteria I have thrown in for discussion above, it seems appropriate that someone from the BF contact him and discuss whether/how/when his project could become an officially recognised project. And if this changes the way the BF recognises other future projects, then good, as more will come!
And then I won’t have to write posts that could be misunderstood as hyping projects that otherwise will be officially recognised…or given whatever due community support.
Again, if Jason decides to give up now or release BlenderJava without goals compatible with the BF, then this should be viewed by all as a sad/bad thing. It’s his project, but he wan’ts feedback. If he just wanted to give up, then he wouldn’t have asked for feedback.
.
So why not show some support by helping him take it to the ‘next level’ – whatever that is, a full port or adapted for web/mobile.
.
I just happen to be of the opinion that the best way to do this would be for the BF to approach him to discuss developing compatible goals for BlenderJava, again – whether as a full port or adapted for web/mobile.
.
I doubt anyone would consider the BF replying to his invitation for feedback ‘as pressure’.
.
It sounds like you have had some jaded experiences, which no one wants from a project, and is another reason I’m suggesting some ‘official’ support rather than let Jason’s project potentially fall apart down the line without it. I wrote ‘potentially’, as he seems to be doing well so far, and perhaps he can continue his project on his own or with the help of those who like his goals. But as I’ve written before, if he is supported more by the BF, then it’s not only more likely that his project will attract more help… but it will also help Blender_C and BF more. Simple agreeable logic.
Sure, ‘there’s nothing stopping Jason from having wiki space, or requesting project hosting on projects.blender.org’ – but did he even know he could do that before possibly reading your post here, and for the sake of *simple* community relations, has anyone from the BF actually invited or checked with Jason to see if he knows he can access these resources???
.
Such obvious simple communication from the Bf may take some ‘pressure’ off him just as the posts of support on his web site no doubt have.
.
But perhaps the BF ‘feels’ that a Java version is not needed – is a waste of time/talent/energy?
I remember many people saying the same about the game engine and many Blender_C features at first!
blndusr: the problem with your argument, is your piling heaps of special considerations simply because of the nature of the project. You also totally ignored my point about past “official” collaborations having a financial component (e.g. apricot, uni-verse). The truth is that ton simply doesn’t have time to be keeping track of all the many blender-related projects, nor does he have the time or resources to give *that* level of support to *all* of them.
Your reasoning is that this is a “historic” development, and that we should “rise to it.” I think that’s a stupid reason to do anything. And again, the BF doesn’t just go around releasing press releases of “we officially approve this project, and that project, and that one, and that one over there, and that one too”, nor does ton spend his time writing articles designed to drum up public support for every single individual blender project, especially since that would royally piss off most of us developers.
I’m not sure why you think this is such a huge development, certainly having hoards of java programmers descending on Jason’s project isn’t going to make it uber good. Remember the old saying that you can’t make a baby in one month with nine women; this applies to programming, too.
The financial component could be from universities trying to open up their students up to Open GL, I was in a graphics lecture last week and blender was mentioned but Iv done 2 years of java. The graphics module itself is in java, If only blender was in java…….
I will try to use and double check it in association with openGL , really useful one.
Thanks
Ok, I just had a cool idea: Google has a tool that lets you take Java and it creates a webpage made of css, js, and html out of it. I’m curious to see what happens if you take this java port of blender and create a webpage out of blender? I’d probably be slow, clunky, and not be practical for anything real, however might be a fun project just for fun.
I would love to see this com about. I’d like to see something like BlenderPocket on Android and this is a huge headstart!
If you are going to port it to Java you should use Maven for your build system and consider using OSGi so that we can develop plugins really easily.
For OSGi I’d look at the Apache Karaf project. http://karaf.apache.org/