Skip to main content

Summer of Code 2007

19 replies [Last post]
Anonymous

Hi list
I'm sure that somebody already knows this, but the application period
for mentoring organizations is already opened for this year's Summer of
Code, it will be open until march 12th, so this looks like a good moment to
decide whether lg3d will apply this year, start discussing which projects
would be interesting this year (wonderland should get some good proposals
for sure) and find who's volunteering for mentoring.

Just starting the discussion.
Regards
[att1.html]

Reply viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
joesimms
Offline
Joined: 2007-03-09

Thanks, It sounds like an amazing amount of work. Do you all have any suggestions?

Pierre D.

lg3d@javadesktop.org a écrit :
> Thanks, It sounds like an amazing amount of work. Do you all have any
> suggestions?
Well... You should try to find as many documentation as possible about Mac OS
X and its API, and try to know whether it's possible or not to redirect
windows display and have a program interacting with them instead of the Mac
OS X tools. Is it possible to create a new expose tool from scratch on OS X ?
That's the first part.
Then map that feature in LG3D.
Then try to get event redirection working without any system modification...
And that part may be impossible.

In the current Linux/Unix scheme, the X server is sending "2D" events to LG3D
display server (DS), for instance "click at 123x456". Then the DS try to
guess where the event is targeted, for instance "click at 42x42 in window
42". Then the DS sends back the event to the X server, with the right
settings, and the X server can then send the event to window 42.
The X server isn't aware of the windows positions. Only the display server is
aware of that. And I fear you won't be able to do that in MacOS X.
Try to find for instance a software zooming windows and still allowing mouse
and keyboard events in the window. If you find one, then look at how it
works, and try to understand how to do that...

But really, you're going far far over the Summer of Code. Two months will
probably not be enough for that.

Perhaps I'm wrong. Perhaps OS X provide great API for that, available
immediately from Java. But that'd surprise me.

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

Krishna Gadepalli

Hi Bjorn
Thanks for the great ideas.

I strongly recommend everyone to go ahead and edit our SOC wiki page
http://wiki.java.net/bin/view/Javadesktop/SummerOfCode2007
and list your ideas there after posting it to the forum. The more
information and project ideas we have on that page, the better chance we
will have of getting our SOC request accepted and more students assigned
to us.

Thanks
-krishna

lg3d@javadesktop.org said the following on 03/09/2007 11:22 AM:
> Hello everybody,
> at first I would propose to change the "SOC-Link" on the main project page to the current (2007) wiki page. It currently leads to the 2006 wiki page.
>
> I had some Ideas for Projects that I sent to Paul some time ago. Now, since this thread is there for brainstorming and discussing Ideas I want to post them here.
>
> Idea 1: Widget Set
> -------
> The idea is designing and implementing a set of reusable "3D-Widgets" that come in handy and do take common tasks, so that not every developer has to reimplement them. Besides this, a result could also be some design patterns derived from the implemented components.
>
> Examples for components to develop could be:
> - some three-dimensional layout-managers (e. g. a ferris-wheel layout)
> - a set of cursors that the developer can use out of the box
> - some standard "window layouts" that can be used as templates or some sort of standard application layouts
>
> This could make the further development easyer.
>
> Idea 2: Swing3D
> -------
> Maybe a bit connected to idea 1. I read that a 3D view layer for Swing UI widgets in planned. If I got that right, there is no implementation of this up till now, right? So that could be a nice project for SoC.
>
> Idea 3: Messenger3D
> -------
> One thing I would be missing if I used LG3D for my daily work is a (3D) messenger. So the target of this project would be to design and develop a three-dimensional instant messenger for multiple protocols (regarding the short time maybe at least a structure for multiple protocols and only one protocol implemented). I saw the 3D AIM client in the incubator project, but this project seems to be not started yet (at least there are no usable source files in CVS).
>
> It could also be interesting to make this messenger Wonderland-ready meaning it uses the Wonderland infrastructure to communicate with other persons instead of the usual messenger protocols. This might be a good starting point for more complex Wonderland-apps.
>
> Idea 4: Notebook application
> -------
> Looking at my desk at work I realize that I always have a bunch of papersheets that I use for writing down spontaneous ideas or the ToDo-list for the next days or weeks. I could imagine that a notebook-like application or a briefcase with sheets in it would be a nice 3D app. On the sheets the user could take notes or draw simple sketches on them. Of course this is quite near to the possibility of taking notes on the back of X11 windows, but I would like a window-independent app of this kind.
>
> This is also an application that could be made Wonderland-ready. I imagine that dragging a sheet to a Wonderland cell would make up a collaborative notepad for all the users attached to the channel.
>
> Idea 5: Object visualizer
> -------
> To me as a developer, a nice 3D application would be some kind of obect or class visualizer for Java- or UML-classes. The presentation could point out the relations between the classes and like this increase the understanding of a class structure. This could also serve as a basis for a visual programming environment where you create programs by plugging classes/objects into another. I think there has been a lot of academic research on this field that can be built on. IMHO such a project could be further developed to a full 3-dimensional IDE in the future.
>
> Again this could also lead to a collaborative Wonderland IDE that a project team could use to distributedly create applications and/or create distributed applications. Looks quite fascinating to me, but is probably far too much of work for the limited time while SoC.
>
> Idea 6: Implementing HUD support
> -------
> To me this is the most exciting idea, but also the hardest one to realize. The idea is to add support for HUDs (or at least one model of HUD) to LG3D so that the user has a REAL 3D picture in front of his eyes. The biggest problem with this idea is probably the cost of testing hardware. By quickly googling I found out that this model [http://www.3dvisor.com/] costs more than 1400 €. One task of this project would therefore be to find cheaper alternatives and maybe find out if there are standard protocols that most HUDs support, so that not every model needs an own driver or other adoption. Together with the GoMonkey system this would be a highly interesting and exciting feature.
>
>
> These are my project ideas. I would be happy if you comment on them and say if you think these ideas are feasible and/or good ideas.
>
> I will definitely apply to the SoC, I just am not yet sure about the idea(s).
>
> Regards
>
> Björn
> [Message sent by forum member 'thebear' (thebear)]
>
> http://forums.java.net/jive/thread.jspa?messageID=207166
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
> For additional commands, e-mail: interest-help@lg3d.dev.java.net
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

nicoley
Offline
Joined: 2007-02-12

Hi. I'm one of the people involved with the Sun Labs MPK20 Project. We will soon be moving this content into Wonderland. I have quite a few project ideas for the Summer of Code that would help to enhance this type of work-related virtual environment. Here are the ideas that I have added to the wiki page so far (http://wiki.java.net/bin/view/Javadesktop/SummerOfCode2007).

Interactive 3D Whiteboard for Wonderland
I am imagining functionality similar to the Thinkature whiteboard. In Wonderland, avatars should be able to stand in front of a whiteboard and collaborate. Extra credit for integrating physical electronic whiteboard technology such as eBeam.

Content Creation Tool for Wonderland
There needs to be a way for non-programmers to get content into Wonderland. This project would involve figuring out how to import models from one of the free 3D editors, such as Google Sketchup. It would also be extremely valuable to have a simple scripting language to specify how the 3D content should be animated in the world.

3D Buddy List for Wonderland
In addition to including standard instant messaging functionality that works with Jabber, features could include viewing a window into the virtual location where your buddy currently is, teleporting to your buddy's location, inviting your buddy to teleport to your location, and initiating a private voice chat with your buddy (the Sun Labs Voice Bridge already supports this functionality, so this would just be creating a user interface for it).

Social Network for Wonderland
Interactive 3D visualization of a person's social network. The network could be created based on a variety of relationships. Types of relationships might include: buddy list contacts, email contacts, who you sit near in the physical world, who you sit near in the virtual world, shared community involvement, joint authorship of papers, articles or on-line material, blogs in common, music tastes in common, hobbies in common, etc. The social network might offer a daily suggestion of a person you would like to meet and why (E.g, "Tom works for your company and is also interested in classical music"). The display could have a "contact" button which could initiate a voice chat or could teleport you to the person's virtual office door.

Connecting Sensors to Avatars and Other Objects in Wonderland
In virtual worlds it's often difficult to tell how remote people are reacting. If we could capture real data from the humans driving the avatars, it would be possible to make avatars much more responsive. Here are just a few examples. With a simple eye tracker, you could properly position an avatar's head to show what the human is looking at in the world. A seat sensor could trigger presence information, figuring out when a person is around or away. A motion sensor and accelerometer on a wristwatch could potentially detect certain arm gestures and translate those into avatar animations. A speech detector could automatically set an avatar's state to busy. A simple thermometer could change the in-world scenery to either match the real world climate or be the opposite (on cold days, a person might want to see a warm beach outside their virtual window).

Nicole.

diabeteo
Offline
Joined: 2005-11-08

Björn: I'm also interested in creating 3D IM. I would like to use Smack API to implement Jabber protocol support. AIM3D is available to download from CVS. There is a couple of classes that can be re-used. Also, there should be a discussion about user interface (it needs changes). So, if you think that I can be anyhow helpful, just drop me a line.

Regards,

Stanley

thebear
Offline
Joined: 2007-02-21

Hey Stanley,
thanks for your offer for help. Since the SoC is not an option anymore I decided to try to get into LG3D anyway. So I will put the IM3D on my personal hobby project list :-)

I'm afraid I can not promise anything, since I have studies at college and am also working at a company. But I would like to give it a try anyway, because LG3D (and Wonderland / MPK20) is highly fascinating to me.

I will drop you a line (or two or three ;-) ) soon.

Greetings

Björn

[edit] I finally found the AIM3D sources :-). It was the package org.jdesktop.lg3d.apps.aim3d.com.wilko.jaim that irritated me. This seems so be just an empty package. [/edit]

Juan González

Hi Pierre,

>
> This year, I think it's better for me to be a mentor. Soc 2005 and 2006 were
> enough for me on Looking Glass, this time I'll try to work on KDE...
>

I'm also planning to follow the mentoring path, but I'm not sure
whether it will be compatible with working as an student for another
project. Last year, I wanted to do that exact thing and it was
rejected to avoid problems, maybe this year they'll accept it, but
it's worth asking...

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

Krishna Gadepalli

I have started a new wiki page for SOC 2007 @
http://wiki.java.net/bin/view/Javadesktop/SummerOfCode2007 using the
last years page as a template. Please take a look at it and make
changes/additions as required.

Here is what needs to be done:

- please do what you can by publicizing Project Looking Glass'
involvement in SOC 2007 and encouraging students to apply

- start filling the above wiki page with new project ideas to
help/encourage students to apply

- we need mentor volunteers - if you are interested please send me your
gmail account address and I will add you to our application. The more
mentors we have the stronger our application

- we need interested students - if you are interested start using this
forum to brainstorm project ideas and then update the above wiki page

- take a look at the SOC timeline
(http://code.google.com/support/bin/answer.py?answer=60325&topic=10729).
The most important date is March 24 which is the student application
deadline. So we have just more than 2 weeks to solicit student applications.

Thanks
-krishna

Juan González said the following on 03/07/2007 10:24 AM:
> Hi Pierre,
>
>>
>> This year, I think it's better for me to be a mentor. Soc 2005 and
>> 2006 were
>> enough for me on Looking Glass, this time I'll try to work on KDE...
>>
>
> I'm also planning to follow the mentoring path, but I'm not sure
> whether it will be compatible with working as an student for another
> project. Last year, I wanted to do that exact thing and it was
> rejected to avoid problems, maybe this year they'll accept it, but
> it's worth asking...
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
> For additional commands, e-mail: interest-help@lg3d.dev.java.net
>

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

William Tracy

> - we need interested students - if you are interested start using this
> forum to brainstorm project ideas and then update the above wiki page

Maybe we could get some students (probably not me...) to help with the
port to Darwin? How about a FreeBSD port? (I know one person expressed
interest in that in the past.)

As time permits, I'm working on a video mock-up of a possible
alternate "start" menu (not that there has been any shortage of menu
UI ideas in the past...). I'm also tempted to fix up and resubmit the
chess game idea I pitched last year. :-)

> - take a look at the SOC timeline
> (http://code.google.com/support/bin/answer.py?answer=60325&topic=10729).
> The most important date is March 24 which is the student application
> deadline. So we have just more than 2 weeks to solicit student applications.

Wow. That's coming up *fast*. Eeek!

William Tracy
--
Some cathedrals took a century to complete. Can you imagine the
grandeur and scope of a program that would take as long?
-- Epigrams in Programming, ACM SIGPLAN Sept. 1982

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

William Tracy

Another idea: How about Jython bindings for LG3D? (Or JRuby, or
JavaScript/ECMAScript, I don't care.)

There always seems to be slightly more people with ideas than with
real Java programming skills, and some script bindings could make it
possible for more people to build working prototypes (or even complete
applications!).

William Tracy

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

paulby
Offline
Joined: 2003-06-13

How about F3 bindings

thebear
Offline
Joined: 2007-02-21

Hello everybody,
at first I would propose to change the "SOC-Link" on the main project page to the current (2007) wiki page. It currently leads to the 2006 wiki page.

I had some Ideas for Projects that I sent to Paul some time ago. Now, since this thread is there for brainstorming and discussing Ideas I want to post them here.

Idea 1: Widget Set
-------
The idea is designing and implementing a set of reusable "3D-Widgets" that come in handy and do take common tasks, so that not every developer has to reimplement them. Besides this, a result could also be some design patterns derived from the implemented components.

Examples for components to develop could be:
- some three-dimensional layout-managers (e. g. a ferris-wheel layout)
- a set of cursors that the developer can use out of the box
- some standard "window layouts" that can be used as templates or some sort of standard application layouts

This could make the further development easyer.

Idea 2: Swing3D
-------
Maybe a bit connected to idea 1. I read that a 3D view layer for Swing UI widgets in planned. If I got that right, there is no implementation of this up till now, right? So that could be a nice project for SoC.

Idea 3: Messenger3D
-------
One thing I would be missing if I used LG3D for my daily work is a (3D) messenger. So the target of this project would be to design and develop a three-dimensional instant messenger for multiple protocols (regarding the short time maybe at least a structure for multiple protocols and only one protocol implemented). I saw the 3D AIM client in the incubator project, but this project seems to be not started yet (at least there are no usable source files in CVS).

It could also be interesting to make this messenger Wonderland-ready meaning it uses the Wonderland infrastructure to communicate with other persons instead of the usual messenger protocols. This might be a good starting point for more complex Wonderland-apps.

Idea 4: Notebook application
-------
Looking at my desk at work I realize that I always have a bunch of papersheets that I use for writing down spontaneous ideas or the ToDo-list for the next days or weeks. I could imagine that a notebook-like application or a briefcase with sheets in it would be a nice 3D app. On the sheets the user could take notes or draw simple sketches on them. Of course this is quite near to the possibility of taking notes on the back of X11 windows, but I would like a window-independent app of this kind.

This is also an application that could be made Wonderland-ready. I imagine that dragging a sheet to a Wonderland cell would make up a collaborative notepad for all the users attached to the channel.

Idea 5: Object visualizer
-------
To me as a developer, a nice 3D application would be some kind of obect or class visualizer for Java- or UML-classes. The presentation could point out the relations between the classes and like this increase the understanding of a class structure. This could also serve as a basis for a visual programming environment where you create programs by plugging classes/objects into another. I think there has been a lot of academic research on this field that can be built on. IMHO such a project could be further developed to a full 3-dimensional IDE in the future.

Again this could also lead to a collaborative Wonderland IDE that a project team could use to distributedly create applications and/or create distributed applications. Looks quite fascinating to me, but is probably far too much of work for the limited time while SoC.

Idea 6: Implementing HUD support
-------
To me this is the most exciting idea, but also the hardest one to realize. The idea is to add support for HUDs (or at least one model of HUD) to LG3D so that the user has a REAL 3D picture in front of his eyes. The biggest problem with this idea is probably the cost of testing hardware. By quickly googling I found out that this model [http://www.3dvisor.com/] costs more than 1400 €. One task of this project would therefore be to find cheaper alternatives and maybe find out if there are standard protocols that most HUDs support, so that not every model needs an own driver or other adoption. Together with the GoMonkey system this would be a highly interesting and exciting feature.

These are my project ideas. I would be happy if you comment on them and say if you think these ideas are feasible and/or good ideas.

I will definitely apply to the SoC, I just am not yet sure about the idea(s).

Regards

Björn

William Tracy

Hello,

> Examples for components to develop could be:
> - some three-dimensional layout-managers (e. g. a ferris-wheel layout)
> - a set of cursors that the developer can use out of the box
> - some standard "window layouts" that can be used as templates or some sort
> of standard application layouts

My understanding is that all of these, to some extent, are already
there. LG already has several 3D layout managers and cursors, but it
could probably use more of both.

> It could also be interesting to make this messenger Wonderland-ready meaning
> it uses the Wonderland infrastructure to communicate with other persons instead
> of the usual messenger protocols. This might be a good starting point for
> more complex Wonderland-apps.

To the Wonderland developers: Will this sort of functionality be built
into Wonderland? (i.e., would this program be redundant?)

> Idea 4: Notebook application

Cool idea. :-)

> Idea 5: Object visualizer

Another cool idea. :-)

> By quickly googling I found out that this model [http://www.3dvisor.com/] costs
> more than 1400 €. One task of this project would therefore be to find
> cheaper alternatives and maybe find out if there are standard protocols that
> most HUDs support, so that not every model needs an own driver or other
> adoption. Together with the GoMonkey system this would be a highly interesting
> and exciting feature.

The FAQ on that site says that the 3D features are not supported under
Linux, so it could be an uphill battle to make these things work with
LG3D. Other than that, it wounds awesome. :-)

William

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

joesimms
Offline
Joined: 2007-03-09

Hello,
My Name is Joe, and I'm a student very much interested in working on a project affiliated with google's summer of code. I must admit that I do not know a large amount about your project, but java is the language I'm most comfortable with. I notice that you do not have an OS X implementation on your site. Would this be an acceptable summer project? Please let me know as I'm a devout Mac user and this sounds like an interesting challenge.

Joe Simms
4th year, Classics Major
Mount Allison University
Apple Certified Portable Technician
Apple Certified Desktop Technician
Jmsmms[(at)]mta[(dot)]ca

William Tracy

> I notice that you do not have an OS X implementation on your site.

See:
http://forums.java.net/jive/thread.jspa?threadID=19795&tstart=0

> Would this be an acceptable summer project? Please let me know as I'm a
> devout Mac user and this sounds like an interesting challenge.

A lot of X-server code needs to be hacked to make lg3d-app run on
Darwin. This is C code (unless there's C++ code I don't know about),
not Java code.

Just giving you fair warning. :-) An OS X port would be very welcome.

William

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

Pierre D.

William Tracy a écrit :
> > I notice that you do not have an OS X implementation on your site.
>
> See:
> http://forums.java.net/jive/thread.jspa?threadID=19795&tstart=0
>
> > Would this be an acceptable summer project? Please let me know as I'm a
> > devout Mac user and this sounds like an interesting challenge.
>
> A lot of X-server code needs to be hacked to make lg3d-app run on
> Darwin. This is C code (unless there's C++ code I don't know about),
> not Java code.
And you would be able to launch only X11 apps, no Cocoa apps...
Such a feature would require really a lot of work, and not using Java. I don't
know whether it's feasible using ObjC or C or C++ in OS X to write a sort of
composite manager, redrawing windows... If it's feasible, then you'd have to
do a lot of things to "map" this feature in Java through JNI. And rewrite
every part of LG3D relying on X of course.
I'm not sure it's possible to handle all that in one summer of code...

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

Krishna Gadepalli

Hi Juan
Thanks for starting this thread... I have been meaning to do this.

Yes, we will indeed be applying to be a mentoring organization for the
SOC this year. This year I guess we will be doing it jointly under
Project Looking Glass and Project Wonderland. I will start a wiki page
for the SOC today and send out a link so that we can start gathering
project ideas.

So if there is anyone out there who will be interested in being a SOC
student on Looking Glass this is a good time to start thinking of
project ideas.
We are also looking for volunteers who can be mentors and guide the
students.

Thanks
-krishna

Juan González said the following on 03/07/2007 06:46 AM:
> Hi list
> I'm sure that somebody already knows this, but the application
> period for mentoring organizations is already opened for this year's
> Summer of Code, it will be open until march 12th, so this looks like a
> good moment to decide whether lg3d will apply this year, start
> discussing which projects would be interesting this year (wonderland
> should get some good proposals for sure) and find who's volunteering
> for mentoring.
>
> Just starting the discussion.
> Regards

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

Pierre D.

Krishna Gadepalli a écrit :
> Hi Juan
> Thanks for starting this thread... I have been meaning to do this.
>
> Yes, we will indeed be applying to be a mentoring organization for the
> SOC this year. This year I guess we will be doing it jointly under
> Project Looking Glass and Project Wonderland. I will start a wiki page
> for the SOC today and send out a link so that we can start gathering
> project ideas.
>
> So if there is anyone out there who will be interested in being a SOC
> student on Looking Glass this is a good time to start thinking of
> project ideas.
> We are also looking for volunteers who can be mentors and guide the
> students.
Hello

This year, I think it's better for me to be a mentor. Soc 2005 and 2006 were
enough for me on Looking Glass, this time I'll try to work on KDE...

Pierre

---------------------------------------------------------------------
To unsubscribe, e-mail: interest-unsubscribe@lg3d.dev.java.net
For additional commands, e-mail: interest-help@lg3d.dev.java.net

Krishna Gadepalli

[att1.html]