jdx-news — Get the latest news about the JD4X project with this mailing list.

[jdx-news] Help with JD4X v0.6.

[Hock K. T. <tay...@ho...>]

I have completed the initial beta release for v0.6. I need your help to test 
and help fix bugs if possible. Also help to test old bugs that you have 
reported (regression testing, if any). You may need to upgrade your Xorg 
server to 6.8.2-77 for this version to work correctly. Well, what the hack 
just test it with whatever you have got. :-)

New features for v0.6 include:

- Quick key support. (Refer to configuration files in 
jd4x/config/zesktop/keymap/ for details)
  + Mouse pointer need to be outside any input window to activate key 
mapping. (To be fixed)
  + Currently, [ctrl+t = xterm] and [alt+tab = cycle windows]
  + To add new key binding, just insert a new XML configuration file to 
.jd4x/zesktop/keymap/ before starting desktop session.

- Better support for single Swing threading model.

- Additional popup menu on window icon in taskbar to restore, minimize or 
close rogue windows.

- Bug for Motif support, eg. XMMS. (To be fixed)

- System slows to a crawl when multiple Java applications are running. (To 
be fixed)

- Many other bugs ... (To be fixed)  :-(

You can get the latest project source from sf.net CVS.

All my current time have been tied up in developing other commercial 
applications, so help chip in some work if you can affort the time and 
interest.

Regards,
Tay.

[jdx-news] [ 1334825 ] Bug closed.

[Hock K. T. <tay...@ho...>]

[ 1334825 ] Cannot manage applications without traditional frames.
https://sourceforge.net/tracker/index.php?func=detail&aid=1334825&group_id=33051&atid=407132

The bug above has been closed and resolved as of the current CVS revision. 
Window focus model has also been fixed to behave correctly like most other 
window managers.

Tay.

[jdx-news] CVS updated.

[Hock K. T. <tay...@ho...>]

Hi everyone,

Finally back from my project break. Here is what is new!

I have finally uploaded JD4X v0.6 to the sourceforge CVS. The current public 
release is v0.5. You should be able to check it out using the pserver 
anonymous access. However, most of you will not have CVS write access unless 
you already have contributed code that the project is already using. If you 
decide to start contributing stuff again to the project, just let me know 
what access you require. I will assist as much as I can to allow greater 
collaboration among us without compromising our project security. There is 
also an eclipse project file include in the CVS, so if you use eclipse for 
the development it will be a great bonus.

For those of us who already has CVS write access, please check with me or 
the original author first before altering somebody else code!

Not everything in v0.6 is working correctly, so don't compile and use it 
yet. I'm still working at improving the window focus model. However, SSL 
certificate exchange (public key), SSL file transfer and chat facilities are 
already working correctly. Every online operation is encrypted using your 
generated digital certificate in the Java key store. If you are really 
worried about security, then just don't join any online groups (make sure 
you are offline!). Also new simpler makefiles have been included. DefaultWM 
has been removed and replaced with Zesktop.

Ensure that the makefiles in linux/ and moth/ have the JDK variable pointing 
correctly to your installed Java toolkit.

To compile:
1) make clean
2) make

To create a installer.jar (entire package public release with GUI 
installer):
1) make clean
2) make release

You will need to create a build directory under jd4x/ source tree before 
compiling. All output files are generated in the build directory.

Have fun and enjoy accessing the source,
Tay.

[jdx-news] Help and Suggestions.

[Tay I. <tay...@ho...>]

Development update: The digital signature system for auto signing messages 
and request is completed and fully integrated. Future versions will allow 
encrypted network request and messages to be send from desktops to desktops 
with digital signatures to verify its authenticity. More on that in the next 
release!

I'm looking for Java tools to include for future releases. I have added a 
public forum to accept other inputs from the general public. Please refer to 
this thread to post your ideas.

https://sourceforge.net/forum/forum.php?thread_id=1275474&forum_id=103792


Thanks.

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/

[jdx-news] Upload errors updated.

[Tay I. <tay...@ho...>]

Update: Downloads before April 15 are incorrect JD4X v0.4 versions! If you 
experienced errors with jdk 1.5, try downloading again. Also make sure the 
mirror site where you downloaded from has been updated with the latest copy 
by comparing the file sizes.

My apology for the errors.

_________________________________________________________________
Get MSN Hotmail alerts on your mobile. 
http://mobile.msn.com/ac.aspx?cid=uuhp_hotmail

[jdx-news] Java Desktop for X v0.5 released

[Tay I. <tay...@ho...>]

JD4X v0.5 supports both jdk v1.4 or v1.5 (jdk 5). The binary version 
released is however, compiled for jdk 1.5 (jdk 5). Users interested to run 
it on jdk v1.4 should download the source code and compile it again with the 
specific v1.4 toolkit.

The new JD4X v0.5 + Zesktop v0.3 includes drag and drop support for adding 
all types of desktop components. Eg. drawers, launchers, applets, workspaces 
and etc... Adding multiple taskbars are also now supported. In addition, 
utility programs like clock, resource applets and etc are also included in 
the desktop applet section of the control panel. GUI device mounting support 
has also been added to mount and unmount devices on the desktop. Eg. CDROM, 
Floppy and etc... This version should be simpler to use but with a much 
richer set of features than the previous versions. For the latest set of 
cool screen shots please refer to 
https://sourceforge.net/project/screenshots.php?group_id=33051.

Enjoy and feedback.

_________________________________________________________________
Get an advanced look at the new version of MSN Messenger. 
http://messenger.msn.com.sg/Beta/Default.aspx

[jdx-news] help

[<Hdi...@ao...>]

Please unsubscribe me. 
 
Heidi A. Dietrich
 
 

RE: [jdx-news] About jxlib and some errors in JD4X

[Tay I. <tay...@ho...>]

Great to have you back in the development.

>1) were should i send sources of lib? (you ask me join to development list 
>so - were?)

We only have one mailing list for now. Join at 
http://lists.sourceforge.net/lists/listinfo/jdx-news.
You can send the source to me or copy (refer to SF.net for details) them to 
http://jdx.sourceforge.net/tmp so that everyone can help test, suggest and 
improve them. More heads are better than one.

>2) what sources of moth should i use for test integration with jxlib?

Use current v0.4. The Moth Java sources have not changed much. Try replacing 
the window manipulation codes like raising window, resizing, moving and etc. 
Leave the workspace support last. I'll test them from my end. Then we can 
merge our efforts to start replacing libjxwin.so with your Java jxlib. Keep 
your code changes to only Moth and don't spread them to the other modules, 
it is cleaner and easier to maintain that way.

>3) some errors in last JD4X distributive - after install (or compiling) it 
>does not start up normaly (with
>blackdown-1.4.2 JDK) problem in new desktop (project core work greate) - it 
>can't do something and all my system
>freez in that case only one method could help - reset (console switch 
>ctl_alt_Fn doese not work). I cant give you
>any logs becouse system freez (and shell >> doese not put any errors in 
>file). I'm using gentoo (with last emerge --sync) :((
>But with SUN jdk 1.5.0 all works fine. ^)

I have noticed that problem with blackdown-1.4.2 JDK too. However, Sun's 
1.4.2 JDK works fine. I think I know where the problem is and have fixed 
them for v0.5. The new v0.5 fully supports Sun's JDK 5 (v1.5) and I may 
consider to drop support for JDK 1.4 to use some new JDK 5 features. The new 
JDK 5 XToolkit is very different from the old motif toolkit and works 
better, however, it is not well implemented yet and suffers from some minor 
bugs. Sadly, it looks like JDK 5 on Linux is only targeted at Gnome/Metacity 
environments.

The new JD4X v0.5 + Zesktop v0.3 includes drag and drop for adding all 
desktop components. Adding multiple taskbar support. Additional utility 
programs like clock, resource applets and etc. Finally, GUI device mounting 
support. It should be simpler to use with more richer features. Unless there 
is a better JD4X WM, we should work towards supporting Zesktop for now.

>PS: I have some ideas about jxlib. Why we could't do java x
>server? Becouse we can't write native java code for
>hardware => we can't paint on screen in native java way.
>BUT we can (for example) make x server as applet and
>embedded JD4X in web. User could get access for his machine
>from any device that have brouser and JVM. Is it
>interesting or no? :))

It seems that we now have more downloads (close to 2000 for the first 
release of our live CD) for the live CD than the core JD4X desktop itself. 
Perhaps the Windows people are more interested in what we're doing than the 
Linux people. I think we can have JD4X run on 2 modes, one on the browser 
for Windows people and the other a full fledge desktop on Linux. I 
personally prefer running JD4X as a Java application on the native platform 
than in a browser. But either way we need the X server or else move to the 
Linux frame buffer/Windows API. We also need to ask the question on why 
would people need 2 desktops running in a single desktop session?

There are some problems with the browser idea. Browser support and 
permission can cause major problems because that is out of our control. The 
other issue is that the resource requirement and dependencies will become 
huge and hard to manage. Performance may also greatly suffer.

We need to think of a solution that will allow the access of all the data on 
the web and bring them right to the desktop for the user as part of their 
desktop environment and interaction. In my opinion, the web was not 
implemented very well. It should be a web describing how and what 
data/information can be accessed and retrieved. Instead, it currently 
determines how it is displayed leaving the how and what to be determined by 
the human user. The machines should have handled this, all the human user 
needs to do is decide on their preference and want.

Tay.

_________________________________________________________________
Don’t just search. Find. Check out the new MSN Search! 
http://search.msn.click-url.com/go/onm00200636ave/direct/01/

Re: [jdx-news] About jxlib and some errors in JD4X

[Jeff W. <jef...@gm...>]

There is a pure java X server called wierdX. I've had some interesting
problems with it (it doesn't intercept keystrokes, not all input
drivers work), but maybe it's been fixed since then.

On Apr 1, 2005 5:46 PM, wi...@ma... <wi...@ma...> wrote:
> Tay i'm realy sorry for a long silince after may promise
> about jxlib. Unfortunaly was some minor changes in my life.
> Hoever i'm back now to my normal network and developing
> life :))
> 
> So:
> 1) were should i send sources of lib? (you ask me join to
> development list so - were?)
> 2) what sources of moth should i use for test integration
> with jxlib?
> 3) some errors in last JD4X distributive - after install
> (or compiling) it does not start up normaly (with
> blackdown-1.4.2 JDK) problem in new desktop (project core
> work greate) - it can't do something and all my system
> freez in that case only one method could help - reset
> (console switch ctl_alt_Fn doese not work). I cant give you
> any logs becouse system freez (and shell >> doese not put
> any errors in file). I'm using gentoo (with last emerge
> --sync) :((
> But with SUN jdk 1.5.0 all works fine. ^)
> 
> PS:
> I have some ideas about jxlib. Why we could't do java x
> server? Becouse we can't write native java code for
> hardware => we can't paint on screen in native java way.
> BUT we can (for example) make x server as applet and
> embedded JD4X in web. User could get access for his machine
> from any device that have brouser and JVM. Is it
> interesting or no? :))
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> _______________________________________________
> jdx-news mailing list
> jdx...@li...
> https://lists.sourceforge.net/lists/listinfo/jdx-news
> 


-- 

Existence matters only to itself.

[jdx-news] About jxlib and some errors in JD4X

[<wi...@ma...>]

Tay i'm realy sorry for a long silince after may promise
about jxlib. Unfortunaly was some minor changes in my life.
Hoever i'm back now to my normal network and developing
life :))

So:
1) were should i send sources of lib? (you ask me join to
development list so - were?)
2) what sources of moth should i use for test integration
with jxlib?
3) some errors in last JD4X distributive - after install
(or compiling) it does not start up normaly (with
blackdown-1.4.2 JDK) problem in new desktop (project core
work greate) - it can't do something and all my system
freez in that case only one method could help - reset
(console switch ctl_alt_Fn doese not work). I cant give you
any logs becouse system freez (and shell >> doese not put
any errors in file). I'm using gentoo (with last emerge
--sync) :((
But with SUN jdk 1.5.0 all works fine. ^)

PS:
I have some ideas about jxlib. Why we could't do java x
server? Becouse we can't write native java code for
hardware => we can't paint on screen in native java way.
BUT we can (for example) make x server as applet and
embedded JD4X in web. User could get access for his machine
from any device that have brouser and JVM. Is it
interesting or no? :))

[jdx-news] Research options.

[Tay I. <tay...@ho...>]

This is an interesting research paper that I found. Zesktop was mentioned 
inside and that is how I  got a hold of it through google. It is the closest 
concept to what I had in mind the last time we discussed about the future of 
JD4X.

Link: http://www.louisville.edu/~prsaha02/thesis-pankaj.pdf

So those of you out there who is interested should consider the concepts 
proposed and see if it is possible to work together and implement something 
similar in JD4X.   ;-)

Let me know if you have any ideas,
Tay.

_________________________________________________________________
Find just what you are after with the more precise, more powerful new MSN 
Search. http://search.msn.com.sg/ Try it now.

[jdx-news] new Live CD out.

[Tay I. <tay...@ho...>]

A new live CD has been released.

Live CD includes JD4X v0.4 and Zesktop v0.2. Some bug fixes to the beta live 
CD (console mouse, proper device state update and etc ...) and additional 
features such as ejecting the live CD when restarting or shutting off the 
computer. However, you might need more RAM for this version, an estimate 
would be 256MB.

You can custom configure your graphic card with "xf86cfg" or "xf86config". 
If you logout, you can use "startx" to boot to the desktop again. However, 
note that changing to a smaller resolution screen might cause some problems.

The live CD will move to Xorg in the next release as the current auto 
probing on Xorg is not as good as XFree86. We will also include Firefox and 
Thunderbird (both are huge to fit into a single 700MB CD).

Enjoy and feedback.

_________________________________________________________________
Get an advanced look at the new version of MSN Messenger. 
http://messenger.msn.com.sg/Beta/Default.aspx

Re: [jdx-news] compiling JD4X to a native binary

[David K. <az...@ro...>]

The key here is to work with swingWT, a swing api that maps to SWT.
Using SwingWT, you can compile swing code to binary. The issues revolve
around heavy manipulation of swing classes, extentions...etc. check it out:

http://swingwt.sf.net

      --David Keymel

Alex Kotchnev wrote:

>Enrico,
>  It doesn't sound to me that you will be able to compile with gcj.
>Basically, as far as I know, gcj doesn't do swing - I remember seeing
>that they had compiled Eclipse to a native binary with gcj, but that is
>only because Eclipse is built on top of SWT (and not Swing). JD4X is
>Swing based.
>
>   That is however based on my general knowledge about gcj - I might be
>totally wrong.
>
>Thanks,
>
>Alex K
>
>Enrico Weigelt said:
>  
>

Re: [jdx-news] Re: JD4X future development + jd4x 0.3 impressions

[Bernhard F. <Ber...@ar...>]

=20

Pythagoras:

Alles, was der Mensch den Tieren antut, kommt auf den Menschen wieder=20
zur=FCck.

=09

Mahatma Gandhi:

Die Gr=F6=DFe und den moralischen Fortschritt einer Nation kann man daran=
=20
messen, wie sie die Tiere behandelt.

=09

Leo Tolstoi:

Solange es Schlachth=E4user gibt wird es auch Schlachtfelder geben.

Leonardo da Vinci:

Die Zeit wird kommen, in der die Menschen den Mord an Tieren ebenso als=20
Verbrechen betrachten werden wie den Mord an Menschen.

=09

Paul McCartney:

Ich esse nichts, was Augen hat. Wenn die Mauern der Schlachth=F6fe aus=20
Glas w=E4ren, w=FCrde jeder Vegetarier werden.

=09

George Orwell:

Wenn Freiheit =FCberhaupt etwas bedeutet, dann vor allem das Recht,=20
anderen Leuten zu sagen, was sie nicht h=F6ren wollen.


An Interview with God:
http://www.ikfk.de/interview.html (2004/12/29 04:00:52)
http://en.wikipedia.org/wiki/User:Fasten (2004/12/29 04:00:52)
http://de.wikipedia.org/wiki/Benutzer:Fasten (2004/12/29 04:00:52)
http://homepage.mac.com/bfastenrath/interview.html (2004/12/29 04:00:52)

RE: [jdx-news] Re: JD4X future development + jd4x 0.3 impressions

[Tay I. <tay...@ho...>]

>Did you mean "cues" (hints/suggestions) instead of
>"queues" (waiting lines) ).

It is times like these that I wonder how I manage to graduate from 
university. ;-P  Yes, I meant cues. I hope at lease it was a good laugh ...

>    where do you keep a copy of jd4x 0.4beta ? Can I get my hands on a copy
>if it is at least minimally functional ? Also, any chances of setting
>up cvs  or svn for jd4x ?

I'm still hacking on v0.4 beta. But it is functional, we use it to launch 
apps on other colleagues machines (running JD4X) and of cause instant 
message each other at our office/lab. However, we have a pretty tight 
intranet so JD4X runs unreserved (i would not dare to do this on a public 
network). The certificate/signature verification process is not done yet, 
only the SSL so I still hope to hold on to it for a little longer until that 
too is somewhat complete. As for CVS, I'm just too lazy to do it at the 
moment. My apology. :-(

>    These are some great ideas that you mention below. Do  you have any
>recommended reading/books that relate to the desktop design you mention?

Unfortunately, no.

>Regarding the 2nd point in your email: can you give more examples of what
>you mean by "easy, self initiated live interaction beyond mere
>communication across the network" ? If you had to pick 3 examples that jd4x 
>would need
>to illustrate this concept, what would they be (e.g. chat, web service
>consumption & serving, video conferencing) ?

I guess a combination of them. Here are some possible use cases.

An example for users:
Say for example, I'm surfing the web and found this cool piece of free 
music. I then realize that my friend Jane and John is also on the same 
network group as I am (on JD4X). I want to share with them this piece of 
music, so I send a XML script that contains the command and arguments (eg. 
URLs, file types and ...) needed to play this music to their desktops. Once 
the XML message reaches their JD4X desktop, JUNK checks and verifies the 
senders certificate/signatures. Assumming Jane and John already has placed 
my public certificate as a trusted source. JUNK reads and executes the XML 
script. XMMS appears on Jane and John's desktop and starts playing the 
music. Following that, my instant message appears on their desktop saying, 
"Did you like that?" - from Tay.

This is just a specific example. It can be generalize to include services, 
voice, streaming video and etc ...

An example for services:
A new Linux user wants to try installing ZerahStar's Linux on his/her 
Windows PC. But have no idea how to begin. The user pops the ZerahStar live 
CD into the CDROM and runs JD4X. The user finds (of all people) Tay on the 
default JD4X network group. The user sends me an instant message to help. 
Assumming that my public certificate is on the live CD and trusted (I work 
for ZerahStar). I first request from JUNK (on the live CD) for the system 
information. Then I help the user, mount, partition and start the 
installation process through sending XML scripts. Every step of the way, I 
guide the user through with instant messaging. The user could clear his/her 
doubts by simply messaging those questions. At the final step, the user 
thanks me and reboots the machine.

I think the key to truely free network communication is removing the need 
for coordination. Coordination information can be automated by the machine 
leaving the user to just worry about the task they want to get done.

>Also, regarding the p2p component, what do you plan to use as the backbone
>(jxta?? - sounds like it..).

I have developed a very simple framework for it using raw multicast and SSL 
facilities. No JXTA. I've gotten a few mail suggestions on keeping JD4X in 
the free domain using free toolkits and Jvms, if you know what I mean. 
Anyway, I use p2p in a slightly different way then it is commonly done. JD4X 
has always been highly experimental, so I'm still experimenting/researching 
on the possible techniques to use and I remain undecided at the moment. 
Hence, the need for more brains and hands to work on it ...

>Finally, could you describe in a few sentences the design/architecture of
>the desktop that you have in mind that would enable this "easy,self
>iniiated live interaction" that would be sufficiently secure at the same
>time ?

The use cases above should have answered some parts of this question, 
hopefully. Basically, we need a component to make coordination information 
available (to answer the where and how question). As in, where are they and 
how to contact them. Then we need another component to answer the who and 
what. As in, what needs to get done and who can get it done. The who and 
what is already completed in JUNK's basic architecture. The where and how is 
what I'm building now.

>3. What are the issues with JDK 1.5 ? The reason I ask is that from
>personal experience it looks much better and feels faster...

It is not possible to know what java types are associated with each window. 
JDK 1.4 allows this to be done through X's protocols but JDK 1.5 hides them 
and reveals only a generic type. Example in 1.4 you can determine that a 
window has type jd4x.window.JXWindow but in 1.5 every window is of type 
sun.awt.peer.Window. (the types used here are just examples and not the real 
vaules) So it becomes almost impossible to define which window as the 
background and which one to behavior like a taskbar. There is a work around 
but I have not implemented it yet.

>4. Where did the "process viewer" go - where one could see what java
>processes were running in the jvm ?

It is not available in v0.3 but I've already added it in v0.4. I'm slow in 
trying to improve the GUI but we're getting there.

>5. I ran the setup for jd4x 0.3 and it did not create the .xsession file
>for me (after I had cleaned up the existing one from my previous
>installation). Is that by design or is it an omission ?

You only need .Xclients, .xsession is just a soft link to .Xclients. Older 
systems uses .xsession or .xinitrc so unless your Linux system uses them, if 
not it should be OK to leave them out.

Thanks for the feedback,
Tay.

_________________________________________________________________
Get MSN Hotmail alerts on your mobile. 
http://mobile.msn.com/ac.aspx?cid=uuhp_hotmail

RE: [jdx-news] Re: JD4X future development + jd4x 0.3 impressions

[Alex K. <po...@li...>]

Tay,

   where do you keep a copy of jd4x 0.4beta ? Can I get my hands on a cop=
y
if it is at least minimally functional ? Also, any chances of setting
up cvs  or svn for jd4x ?

   These are some great ideas that you mention below. Do  you have any
recommended reading/books that relate to the desktop design you mention
? I feel totally unknowledgeable in the area of your first suggestion
(as I don't know much about the theory of UI design and user
interaction. Did you mean "cues" (hints/suggestions) instead of
"queues" (waiting lines) ).

Regarding the 2nd point in your email: can you give more examples of what
you mean by "easy, self initiated live interaction beyond mere
communication
across the network" ? If you had to pick 3 examples that jd4x would need
to illustrate this concept, what would they be (e.g. chat, web service
consumption & serving, video conferencing) ? I totally agree that
currently, users kinda have very limited "windows" to the network/externa=
l
world - e.g. they fire up a browser to look at pages, start up a IM clien=
t
to chat, etc - these really should be features in the desktop itself, and
not necessarily "patches"/external apps to the desktop.

Also, regarding the p2p component, what do you plan to use as the backbon=
e
(jxta?? - sounds like it..).

Finally, could you describe in a few sentences the design/architecture of
the desktop that you have in mind that would enable this "easy,self
iniiated live interaction" that would be sufficiently secure at the same
time ?

  Some first impressions from 0.3 :
1. I did like the progress that the zdesktop brought to jd4x, especially
the ability to minimize/maximize/shadow the windows. I also liked the
extensive setup utilities that come with it. I also liked the "skinning"
features - that was quite slick !!
2. However, the general feel if the icons and the look of the widgets is =
a
little bit clunkier - I did like the "start" menus that were with 0.2 at
the task bar. Also, the "drawers" and the "start buttons" are HUGE now -
my resolution (on a laptop 1024x768) is not too high, and they take a
disproportinate amount of space. Finally, the icons and buttons in 0.2
just looked MUCH slicker - not they just have the "Metal" L&F which feels
very outdated.
3. What are the issues with JDK 1.5 ? The reason I ask is that from
personal experience it looks much better and feels faster...
4. Where did the "process viewer" go - where one could see what java
processes were running in the jvm ?
5. I ran the setup for jd4x 0.3 and it did not create the .xsession file
for me (after I had cleaned up the existing one from my previous
installation). Is that by design or is it an omission ?

I am currently running this from jd4x and I feel quite comfortable in the
environment (which was not quite the case with 0.2). Great Job !

Thanks,

Alex K

Tay Ivan said:
> Hi Alex, great to hear from you again.
>
>>I'm trying to come up with a good topic for my master's thesis in
>>Software Engineering (starting sometime in Sept 2005).  My idea is to
>>take an open source project and to either re-work and improve a part of
>>it (if needed), add needed or missing functionality, etc. Now, since I
>>had learned a little about JD4X , I thought it might be a good match
>>since I do find it quite exciting, and there seems to be a lot of room
>>to develop the project.
>
> Your idea is truely music to my hears. I consider JD4X to be still very
> primitive because of the many missing features that still remains
> undeveloped. Any form of help is greatly appriciated.
>
>>I would be interested to hear ideas of what might be a decent (not too
>>small, not too wide) scope project within JD4X that I could take on and
>>contribute back to the project.
>
> There are 2 major areas lacking in JD4X that I can propose (for your
> consideration):
>
> 1) The question: How to enrich the user's experience on the desktop.
> (You'll
> need to narrow this question down with an implementable technique.
> Example,
> by mixing effectively audio and visual queues.) Sound for example is
> abused
> on most desktops. Most implementations does nothing more than notify ab=
out
> silly events that contribute very little to the user's desktop experien=
ce.
> In my opinion, audio queues should be used to aid the user in using the
> desktop (like an audio tool tip). When combined appropriately, the desk=
top
> experience can be very impressive. Current UI theory has been largely o=
n
> visual queues (which is very well done). However, what a user is able t=
o
> see
> is only a small portion of what contributes to his/her desktop experien=
ce.
> We'll need to learn from the visually and hearing impared users, as wel=
l
> as
> games, to gain better insight on this issue.
>
> My opinion: I think that current desktop implementations have turn out
> this
> way because of the actual (code) architectural design. The MVC model ha=
s a
> big part to play in influencing this sort of event driven desktop
> experiences. There is a lot that can be researched into on this area.
> Example, what if data (both mobile and localized/saved to file) itself
> could
> have an associated behavior that defines how it wants to be experienced=
 on
> the desktop? and so on ... The key idea is not only viewing data but
> experiencing them on the desktop.
>
> 2) The question: How can desktops encourage users to build e-communitie=
s
> on
> the network. (Again, you'll need to narrow this down to implementable
> techniques. Example by using peer to peer computing.) The trend that se=
ems
> to stand out most the last few years in popular software has been, no n=
ot
> technical superiority but supporting e-communities (eg. Linux, firefox,
> openoffice and so on ...). In my opinion, desktop users are still isola=
ted
> even with web browsers and chat programs. The reason is because they la=
ck
> self initiated real time interactive elements in them. The desktop must
> provide easy, self initiated live interaction beyond mere communication
> across the network. Imagine if your friend could popup on your desktop
> through his/her video camera (securely) without the need for any
> coordination between both parties. Trusted service providers could help
> users fix bugs and do upgrades, installations interactively all in real
> time
> through the network. I'm currently working on such a framework within J=
UNK
> but I'm sure more brains working on it is better than one.
>
> My opinion: The key problem to solve here is a secure, automated and
> transparent communication architecture. I've completed the first phrase=
 of
> development using a combination of peer to peer and web service concept=
s.
> Security is done using SSL and certificates (public and private keys).
> Current JD4X v0.4beta allows users to join network groups that will mak=
e
> available those within the group. Users can then instant message (more
> features are planned later) each other at real time without the need to
> know
> about where and how. No central servers (decentralized design), purely =
a
> peer to peer implementation. I'll introduce the beta to the mailing lis=
t
> when I'm done building the UI around the framework (as usual with any
> beta).
>
> I'm always happy to see more code contributions. So whatever you may
> decide,
> I'm happy if it helps JD4X move forward.
>
> My best wishes to your master's thesis,
> Tay.
>
> _________________________________________________________________
> Get cheap fares online with MSN Travel http://www.msn.com.sg/travel/
>
>
>
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users=
.
> Discover which products truly live up to the hype. Start reading now.
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> jdx-news mailing list
> jdx...@li...
> https://lists.sourceforge.net/lists/listinfo/jdx-news
>

RE: [jdx-news] Re: JD4X future development

[Tay I. <tay...@ho...>]

Hi Alex, great to hear from you again.

>I'm trying to come up with a good topic for my master's thesis in
>Software Engineering (starting sometime in Sept 2005).  My idea is to
>take an open source project and to either re-work and improve a part of
>it (if needed), add needed or missing functionality, etc. Now, since I
>had learned a little about JD4X , I thought it might be a good match
>since I do find it quite exciting, and there seems to be a lot of room
>to develop the project.

Your idea is truely music to my hears. I consider JD4X to be still very 
primitive because of the many missing features that still remains 
undeveloped. Any form of help is greatly appriciated.

>I would be interested to hear ideas of what might be a decent (not too
>small, not too wide) scope project within JD4X that I could take on and
>contribute back to the project.

There are 2 major areas lacking in JD4X that I can propose (for your 
consideration):

1) The question: How to enrich the user's experience on the desktop. (You'll 
need to narrow this question down with an implementable technique. Example, 
by mixing effectively audio and visual queues.) Sound for example is abused 
on most desktops. Most implementations does nothing more than notify about 
silly events that contribute very little to the user's desktop experience. 
In my opinion, audio queues should be used to aid the user in using the 
desktop (like an audio tool tip). When combined appropriately, the desktop 
experience can be very impressive. Current UI theory has been largely on 
visual queues (which is very well done). However, what a user is able to see 
is only a small portion of what contributes to his/her desktop experience. 
We'll need to learn from the visually and hearing impared users, as well as 
games, to gain better insight on this issue.

My opinion: I think that current desktop implementations have turn out this 
way because of the actual (code) architectural design. The MVC model has a 
big part to play in influencing this sort of event driven desktop 
experiences. There is a lot that can be researched into on this area. 
Example, what if data (both mobile and localized/saved to file) itself could 
have an associated behavior that defines how it wants to be experienced on 
the desktop? and so on ... The key idea is not only viewing data but 
experiencing them on the desktop.

2) The question: How can desktops encourage users to build e-communities on 
the network. (Again, you'll need to narrow this down to implementable 
techniques. Example by using peer to peer computing.) The trend that seems 
to stand out most the last few years in popular software has been, no not 
technical superiority but supporting e-communities (eg. Linux, firefox, 
openoffice and so on ...). In my opinion, desktop users are still isolated 
even with web browsers and chat programs. The reason is because they lack 
self initiated real time interactive elements in them. The desktop must 
provide easy, self initiated live interaction beyond mere communication 
across the network. Imagine if your friend could popup on your desktop 
through his/her video camera (securely) without the need for any 
coordination between both parties. Trusted service providers could help 
users fix bugs and do upgrades, installations interactively all in real time 
through the network. I'm currently working on such a framework within JUNK 
but I'm sure more brains working on it is better than one.

My opinion: The key problem to solve here is a secure, automated and 
transparent communication architecture. I've completed the first phrase of 
development using a combination of peer to peer and web service concepts. 
Security is done using SSL and certificates (public and private keys). 
Current JD4X v0.4beta allows users to join network groups that will make 
available those within the group. Users can then instant message (more 
features are planned later) each other at real time without the need to know 
about where and how. No central servers (decentralized design), purely a 
peer to peer implementation. I'll introduce the beta to the mailing list 
when I'm done building the UI around the framework (as usual with any beta).

I'm always happy to see more code contributions. So whatever you may decide, 
I'm happy if it helps JD4X move forward.

My best wishes to your master's thesis,
Tay.

_________________________________________________________________
Get cheap fares online with MSN Travel http://www.msn.com.sg/travel/

[jdx-news] JD4X future development

[Alex K. <po...@li...>]

   I'm trying to come up with a good topic for my master's thesis in
Software Engineering (starting sometime in Sept 2005).  My idea is to
take an open source project and to either re-work and improve a part of
it (if needed), add needed or missing functionality, etc. Now, since I
had learned a little about JD4X , I thought it might be a good match
since I do find it quite exciting, and there seems to be a lot of room
to develop the project.

 I would be interested to hear ideas of what might be a decent (not too
small, not too wide) scope project within JD4X that I could take on and
contribute back to the project.




Thanks,

Alex K

[jdx-news] compiling JD4X to a native binary

[Alex K. <po...@li...>]

Enrico,
  It doesn't sound to me that you will be able to compile with gcj.
Basically, as far as I know, gcj doesn't do swing - I remember seeing
that they had compiled Eclipse to a native binary with gcj, but that is
only because Eclipse is built on top of SWT (and not Swing). JD4X is
Swing based.

   That is however based on my general knowledge about gcj - I might be
totally wrong.

Thanks,

Alex K

Enrico Weigelt said:
> * Tay Ivan <tay...@ho...> wrote:
>
> Hi,
>
>> >How can I build it to a native binary w/ gcj ?
>>
>> I've never tried it before, so I don't really know how. It has been
>> suggested in the mailing list before (I've forgotten who it was). You'=
re
>> on
>> your own on this one! If someone in the list has done it before, pleas=
e
>> help enlighten us. Better yet, if you do succeed in doing this yoursel=
f,
>> contribute it to the project!
>
> hmm, I need to know the code structure. Which classes belong together
> in which ways (i.e. which one is containing the main(), etc) ?
>
>
> cu
> --
> ---------------------------------------------------------------------
>  Enrico Weigelt    =3D=3D   metux IT service
>
>   phone:     +49 36207 519931         www:       http://www.metux.de/
>   fax:       +49 36207 519932         email:     co...@me...
>   cellphone: +49 174 7066481
> ---------------------------------------------------------------------
>  -- DSL ab 0 Euro. -- statische IP -- UUCP -- Hosting -- Webshops --
> ---------------------------------------------------------------------
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: InterSystems CACHE
> FREE OODBMS DOWNLOAD - A multidimensional database that combines
> robust object and relational technologies, making it a perfect match
> for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8
> _______________________________________________
> jdx-news mailing list
> jdx...@li...
> https://lists.sourceforge.net/lists/listinfo/jdx-news
>

Re: [jdx-news] Has anyone tried...

[Jeff W. <jwa...@co...>]

I've decided to bail on the X window as JComponent idea for the moment. 
I don't really have that much time. ;)

I'm creating SimpleWM, a JXMinorModule that's driven entirely from one 
persistent menu. I've been wanting a window manager like that for a 
while but never found one.

RE: [jdx-news] Has anyone tried...

[Tay I. <tay...@ho...>]

Hi Jeff,

It has been a while since we heard from you!

>I hope to get going again soon. I"ll be developing another WM, something I 
>haven"t actually seen implemented anywhere else.

There are 3 examples for you to reference from. The defaultWM, Zesktop and 
Jodfee (code is on our web space, however, it supports only v0.1). Have fun, 
it should be an interesting project to do!

>Has anyone here tried creating an extension of JComponent that represents 
>an X window?

For your consideration (may not be accurate because of my limited knowledge 
of X):
This would mean mapping an X window to a Java component. A scary thought, 
but not impossible as Moth does something like that too! The Java toolkit 
would have little to no idea of this window, so it is hard to integrate 
nicely. Most toolkit implementation only deals with the windows they create 
and leave windows created by others alone. The most that they do is report 
the desired changes by sending a notification to the owners (if they do make 
any changes). Handling windows not own by you is dangerous because it may 
result in a race condition like effect (a sure crash formula), hence the 
need for ICCCM. The Jvm will not handle these windows because they are not 
created by the Java toolkit so the application will not be able to 
manipulate them in Java code. The only way (I think) to achieve this is to 
define your own Java toolkit, eg. Gnu Classpath and extend it by hacking the 
lib's code.

There may also be a problem with the J2SDK license to do that but I'm not 
sure on this. You'll need to check the license agreement in detail!

Hope this helped,
Tay.

_________________________________________________________________
Find love on MSN Personals http://personals.msn.com.sg/

Re: [jdx-news] Version 0.3: first impressions

[Jeff W. <jwa...@co...>]

I hope to get going again soon. I'll be developing another WM, something 
I haven't actually seen implemented anywhere else.

I didn't disappear, just got caught up in developing another 
client/server application for work.

Speaking of which, I hadn't noticed before that swing is now WAY faster 
than awt, both in load time and execution. Interesting...

[jdx-news] Has anyone tried...

[Jeff W. <jwa...@co...>]

Has anyone here tried creating an extension of JComponent that 
represents an X window?

In other words, each window is a JComponent that could be used as the 
view component inside an InternalFrame, allowing the jvm to handle 
windowing considerations while to the application the connection to X is 
transparent.

Have I asked this question previously? :)

[jdx-news] Re: Version 0.3: first impressions

[mp <ma...@yu...>]

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
  <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
  <title></title>
</head>
<body>
<br>
<blockquote
 cite="mid...@ma...%3E%20(added%20by%20p...@ma...)"
 type="cite">
  <pre wrap="">One question. Need to know this to fix it.
  </pre>
  <blockquote type="cite">
    <pre wrap="">- the icons are nice but all have a white caret which doesn't look
    </pre>
  </blockquote>
  <pre wrap=""><!---->
What is a "white caret"?
  </pre>
</blockquote>
Sorry, that's a mistake, it had to be a 'square' instead of 'caret' (I
hope this is the right word ...).<br>
<br>
Marc<br>
<br>
</body>
</html>

RE: [jdx-news] Version 0.3: first impressions

[Tay I. <tay...@ho...>]

Thanks for these good feedbacks, I'll keep them and work at correcting them 
for the next release.

One question. Need to know this to fix it.
>- the icons are nice but all have a white caret which doesn't look

What is a "white caret"?

>- no menus, this is a really essential, together with the fact that it 
>seems not possible to add new icons and applications to the desktop or am I 
>missing something ?

Currently you can add new stuff by creating new XML files in the 
.jd4x/zesktop/... directory. Models can be added by adding new XML files to 
jd4x/config/zesktop/.../model/... (eg. jd4x/config/zesktop/launcher/model) 
Hopefully, I'll have enough time to work on a more extendsive GUI 
configurator, I also think this is important.

We'll get the desktop to be a daily usable desktop in time to come. :-) I 
intend to support this desktop long term, it is a more well thought out 
design then the tutorial WM (more like a quick hack). Help in anyway to 
improve it will be deeply appriciated. Feel free to build better tools to 
make the desktop easier to use. Lets collaborate on extending the 
functionality of this desktop.

Thanks again,
Tay.

_________________________________________________________________
Get MSN Hotmail alerts on your mobile. 
http://mobile.msn.com/ac.aspx?cid=uuhp_hotmail