Sun Microsystems, a company that has been making noise lately in the Linux desktop market with StarOffice 7 and Java Desktop System, is currently working on an experimental 3D successor to Java Desktop that they believe will change the way we interact with computers, and in the end elevate the popularity of Linux in general.
Sun Wants to Make Linux 3D
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
“I think in and of itself, it has a big wow effect. It’s cute to see these things like 3D animations of stuff moving around and think of collaborative space, but how does it make my business more productive?” Murphy wonders. He believes, they need to tie the product to business productivity to make it successful. “As it’s shown (in the demonstration), if it’s not going to make me more productive, then who cares?”
I couldn’t have said it better. Is this just feature bloat? Linux 3d drivers are going to have to mature or go open source for this to be viable. Maybe Sun is trying to create demand for high end Sun 3D workstations with this.
I think if they put something out that makes WinXP look like really old technology, it’ll lend credence to the Linux movement among the less technically inclined.
that I have submitted an news-suggestion to OSNews about the “Looking Glas” technoloy from Sun several weeks ago and it never shows up 😉
That’s because we have _already_ talked about Looking Glass MANY times here on osnews (I was in fact one of the first people reporting on it after Sun demoed it to me a few months ago) and the article you linked was offering nothing new. THIS new article does offer some news though.
OK – I think I have to check OSNews even more frequently than I do right now to stay online with the news *shame_on_me* 😉
You know what would be interesting (looking to the videogame industry for ideas). Dropping the mouse and taking a gamepad for moving around in 3D areas. I’ve been playing FFXI (ahem) and I can manipulate every aspect of that game much faster than I can move around windows. Plus the added bonus is that just about everyone <25-30 will probably feel right at home with the new technology.
Just a thought:) I’ll be quiet now;)
When I first saw looking glass at linux world in NY, and now seeing it here again a few things always come into my mind. For starters… what kind of hardware does this require? It seemed to run pretty smooth all along, but I can’t imagine that this is anything close to resource friendly. Personally I just dont’ find it that great looking… as far as usability goes, I don’t need a note for every application/window, I’d rather not have my apps stack to the side as he displays because then I’m losing horizontal screen real-estate, and as far as flipping windows around… I just couldn’t care less. It’s nice to look at, and it’s refreshing, but I don’t think I’d ever use it myself. Lastly, I’d just like to note that when I saw this at Linux World they mentioned this was the side project of a single developer, maybe more have jumped on the bandwagon now, but the way they came across was that this single developer decided to have some fun and see what he could do and now they’ve sort of… ummmm… taken control of his work so to speak. It makes me wonder if Sun has one of those clauses that says any work you develop belongs to them, if that’s the case and this is why it’s “Sun’s project looking glass” then I’m sorely disappointed. If this guy just gave it to them or pushed it to them for the sake of seeing what the company could do with it, all the power to him, but I’d be giving a lot more credit to the guy who made it than saying “Sun is looking at the future of desktops” when it was, at least up until Linux world, some Sun employee’s little side project.
But they’d have to make it possible for both Gnome and Kde application to be 3d enabled, by providing a window manager able to use it for example.
I personally would like to have something like that to juggle windows. That would be something between the icons and the full sized windows. Something like expose but with a different implementation. Microsoft is trying to patent something like expose in 3d for virtual screens by the way.
Regarding real estate, large very high resolution desktop screens should be with us within a couple of years (at least I hope) so that would be a good use for it, but there is nothing wrong being ready in time.
Is is already possible with fdo’s Xserver? Using the composing manager shouldnt it be possible for the composer to just draw the window with 3d effects?
I don’t know why no one gives me a whereabout of the demo. So here is one.
http://wwws.sun.com/software/looking_glass/demo.html
That and considering the “issues” we have with 2D interfaces. Are we really ready to tackle a 3D interface?
I have yet to see a 3D interface that could not be projected onto a 2D interface with no loss in functionality. I mean, do people really want to put notes on the “back” of a window? What happens to those notes when the window closes? Do people really annotate things like this?
Solution looking for a problem.
This statement I find most important:
What we are doing is dissecting the original concept and rebuilding it using traditional desktop graphics technologies,” Soto says.
what exactly are they rebuilding, will it be have a built in Xserver, will it be similar to KDRIVE.
what traditional desktop graphics technology? Will it be again an Xserver, built on top of a OpenGL rendering model.
Will Keiths be finished before theirs, or am I just making wrong assumptions.
I’ve heard quite a bit of criticism of Looking Glass and the concept of a 3D graphical environment in general. These complaints seem to be stemming from the idea that Looking Glass makes window management more cumbersome rather than less, or that existing technologies do a better job of addressing the same desktop usability problems addressed by Looking Glass.
Much like Expose, Looking Glass provides a solution to a set of window management problems which novice or intermediate users (and much more, but I won’t go into that here). With Sun pitching JDS for enterprise deployment for a set of users who will fall primarily into the intermediate skill level category, having an intuitive technology to address window management problems the users were previously experiencing with Windows, thereby increasing employee productivity, is a major selling point.
I have great appreciation for an easy-to-use, intuitive technology which addresses a commonplace problem. Many of the intermediate users I support here have trouble managing virtual desktops, and I think could greatly benefit from the window management provided by either OS X with Expose or JDS with Looking Glass.
So, I ask that before you ridicule Looking Glass because of how you think it behaves from watching the demo, give it a try yourself. This may be somewhat difficult if you are not friends with a Sun employee and don’t attend trade shows where it is being actively demoed, in which case I’d simply suggest you refrain from such general comments as “Looking Glass/3D desktops are useless” until you can actually see for yourself what real world problems Looking Glass is addressing.
While all the hype has been about Suns 3D desktop, there is another 3D desktop available: http://www.3dwm.org. One wonders if the Sun developer just backpacked this development and presented it to his Sun management team. If Sun does want to release LookingGlass as a Linux desktop, which license will it be under?
will it be have a built in Xserver
Looking Glass acts as an X server, but uses Java3D as a backend. In theory it should work on top of any system which supports Java3D.
will it be similar to KDRIVE.
No, kdrive is currently tied to the Linux framebuffer device as a backend and is consequently non-portable.
what traditional desktop graphics technology?
The X11 wire protocol, and potentially Sun’s Sunray server.
Will it be again an Xserver, built on top of a OpenGL rendering model.
Basically, but all the underlying hardware management and everything that entails is left to the host specific window server/GL implementation being utilized by Java3D.
Will Keiths be finished before theirs, or am I just making wrong assumptions.
The question isn’t so much “finished” as “released”, and we don’t really know Sun’s plans for releasing Looking Glass.
I’ve always assumed that Looking Glass would be utilized in conjunction with the new 3D Sunrays that Sun is currently developing (no public information available, sorry). The Sunray server is essentially an X11 server (and Sunray 3D will introduce GLX support), so the Sunray protocol, which has enough performance to play DVDs over the network, will be used to send textures while all GL commands being issued will simply be tunneled over the Sunray protocol, and the actual 3D rendering will be done on the Sunrays themselves. The Looking Glass Java application will run on the system running the Sunray server, and other X11 applications will simply connect to that rather than the Sunray server itself.
Judging from the screenshots, this is just a 2D desktop mapped to a cube. The latest screenshot looks to show a 3D model, but that doesn’t make a desktop.
Sun’s Looking Glass is so much more than this. 3DWM looks to be a simple gimmick type thing that actually makes it harder to use the desktop. It should also be noted taht 3DwM has been around a long time, and has barely progressed at all in the years I’ve known about it. It’s still boring looking. As usfull as VRML web browsing was.
— “While all the hype has been about Suns 3D desktop, there is another 3D desktop available: http://www.3dwm.org. One wonders if the Sun developer just backpacked this development and presented it to his Sun management team.” —
I don’t see anything even remotely similar here. 3dwm looks like a completely useless gimic, where as Looking Glass has some actual utilty too it that some find useful.
Frankly, I can’t wait to see what they come up with! It will either be something new and usefull, which is great, or it will not, which is no skin off my back.
There is simply no bad side to this.
We won’t know what can be done with it until
it is out there in the wild. You all sound so old.
So far the Sun demo looks like it’s creating more eye candy and more overhead without any particular purpose. This could be just me; I don’t know when this Looking Glass project began in relation to Longhorn development… but it looks like Sun’s attempt to create a competitor to Longhorn. Seems to do much of the same sort of glitzy tricks that are novel at first and eventually get in the way as you spend more time spinning and tilting things than you used to by tabbing between windows. Hopefully in their next demo they’ll be able to show more meaningful features that make my computer do more or do it better instead of just making it more fun. Fun is nice, but in the end more often than not I’m looking to get something done, not play games with my desktop.
…except replace “3D” with GUI and “2D” with command line. C’mon we’ve been wrong before I say we give it at least 5 years befroe we come to a judgement, it’ll take time. Just look at how useless plain X is and then compare KDE/GNOME/XFCE/etc.
The alt-tab shortcut works great for quickly cycling through open programs.
I’ve never seen a design nor implimentation of a design that was 3D based that was efficient or easy to use as a UI. More importantly, none that was BETTER than the existing 2D designs. My opinion is that it’s just not going to happen. I don’t want people to stop trying to do new things, but I do wish that bad ideas wouldn’t constantly get rehashed every few months due to the cool factor or because some people refuse to do the research and/or believe the results. A 3D UI is not the way to go and there are tons of usability and psychological research results to prove exactly why it’s the wrong way to go.
A 3D UI is not the way to go
Why?
and there are tons of usability and psychological research results to prove exactly why it’s the wrong way to go.
Where?
Here’s some research to the contrary of your conjecture:
http://research.microsoft.com/adapt/TaskGallery/
“Our cognitive psychologists ran user studies to see whether the TaskGallery successfully engaged the knowledge and abilities people use to navigate physical space. Study participants told them that it seemed more natural to them to place tasks on the walls of the Gallery rather than on the floor or ceiling. Users were also very adept at remembering the depth ordering of tasks. This suggests that people feel compelled to treat the Gallery like a real-world environment, rather than just as an interesting desktop theme. In general, users preferred the TaskGallery to the existing Windows interface. Further studies are ongoing and will attempt to tease out the various advantages and qualities of using a 3D environment for normal productivity tasks.”
re: 3d. is welltris better or more popular than tetris?
Well I do more mousing than typing but wouldn’t it make sense to use one mouse on the left and one on the right. The 2nd mouse would be used for navigation in the 3d desktop and the first for highlighting objects in the 3d world.
Doesn’t sound any more complicated than pushing a clutch, gassing, shifting and steering at the same time. Plus I could fire my chiropractor after my arm muscles reach equilibrium.
I remember seeing “Unix” in Jurassic Park the movie. The girl was flying around in a visual representation of a filesystem and I thought, being the naïve youth that I was, “Wow! So that is Unix”
on the other hand, i doubt sun will publish their 3d desktop to the ‘community’.
between looking glass and fd.org Xserver i rather choose the latter.
wat linux users want are more ‘just work’ reality, quality software, out of the box eXPerience. Not some ‘3d desktop’. i bet i need to ‘exit’ their 3d desktop before playing Doom3.
Except the GUI doesn’t replace the CLI. GUIs and CLIs are two different things, and are good at different tasks.
But I’ve been waiting for a 3D desktop for quite some time.
I was hoping to see that some here were innovative but something seems to be missing.
When this stuff get’s launched you can’t restrict your mind to the current input models we have today (keyboard+mouse). We probably will see some heavy changes to HOW we work, not only to what we see.
I’m not exactly sure how this new input model might look, but some sort of glow might be it? Maybe something that your feet work with? How about some Eyecam watching how your eyes focus on some part of the screen and while your pupils grow, it zooms for you and when you want a wider perspective the eyecam analyzes that your pupils shrink (BTW is pupils right word here? English not my native) and you suddenly have some really cool input models.
Think about possibilities before reject things as “not working” etc… I still think the 2D desktop stink as there never seem to be room for all the stuff you want in there without having 6 screens…
Even if it is useless, as long as it looks great (which it does) and runs fast (which it does, although I’m really wondering what are the demo video laptop’s specs) and is free software, I’ll give it a try.
With the right controls though, I think it could really be a very useful and nice addition…
A 3D interfaces is somewhat redundant and useless for desktop computing. In games, 3D perspectives exist for a single purpose. That is to create a feeling of immersion or an atmosphere that mimicks the flexibility of real space. I certainly don’t need to be immersed to use my email, nor do I need to be immersed to search my computer for files or play music files.
We need to ask ourselves what the purpose of a window manager is. I think its purpose is to manage windows in manner suggested by the user. How many windows can a user possible handle on one desktop? What sort of manipulation patterns exist for users to manage them? How easy is it for users to understand and use effectively this manipulation patterns? Will introducing another dimension simplify or amplify already existing interfaces?
These are some of the questions I ask myself. And my response to all the questions, I listed above, suggests that 3D window managers do not serve to improve user experience both to make it more complex, less intuitive. and redundant. Like the article states, I think this project exists more for a “wow” factor, rather than I genuine interest in resolving the flaws of current 2D interfaces or window managers.
If Looking glass will double my desktop productivity, I take back all my statement. Until then, the projects on http://www.freedesktop.org have a lot more of my attention. Also, is this going to be a proprietary technology, or does SUN have any intention to promote it via open source on freedesktop.org along with the other interesting desktop projects?
I think the biggest hurdle is going to be a controller that can fully interact with the environment. I saw a post earlier about using a gamepad to navigate the environment. That’s not a bad idea, but it may need to be a little more specialized.
I can see a few practical uses for it too. (All possible in 2d, but not quite as effectively) Suppose you were keeping track of information on a certain subject. You could organize them into sort of a tablet or binder. You have a webpage on the front, you flip the page over and there’s a video that’s important to your subject on the next page, maybe a spreadsheet of statistics on the next, and so on. They’re all grouped together into one “collection” so it would be easy to keep track of all the elements, no worrying about which program to open them with, where they’re stored, etc.
Well, I’m hopeful anyway
Even funnier is this line:
” Given their recent financial issues such as having their bond rating lowered to junk status by Standard and Poors earlier this month, it’s important for them to have a high-profile desktop project like Looking Glass to show that they are on the cutting edge of Linux development. “
Or it could be used to strengthen an illusion. Such as going forwards, and backwards in time.
As far as controller. How about The Orb(TM)?
Or one of those 3D joysticks with a built-in gyro.
“3D successor to Java Desktop” I also saw it demonstrated. I also read several interviews in which it would be included as an optionl addon.
Furthermore, sun also claims to ship JDS for Solaris soon so it will run on multiple unix systems..
This article confuses me
I visited the Looking Glass demo point at CeBIT 2004 and wasn’t impressed. The only real 3D application which exists is the CD-Player rotating the discs in space. Too early and too few examples to say it’s useful.
The Gnome project has to decide which software
platform that gives them features like garbage collection,
Java or Mono.
http://ometer.com/desktop-language.html
Maybe Sun should lower its focus and try to make Java win the
battle for creating ordinary gnome desktop applications
before rabbling on about the future.
There is an important discussion going on here, don’t miss it.
I was at FOSDEM this year and I saw the freedesktop and directfb projects doing the same thing.
Actually the directfb project’s demonstration was really impressive.. running quake with real translucent window with a video running in the back and having some opengl app running at the same time with full speed.. it was really impressive..
I believe Looking Glass isa Java app. So even if it is GPL, if Java still isn’t, it’s irrelevent. These are dangerous times for the GNU/Linux and Free Software initatives. I for one won’t use Looking Glass when (if?) it gets released unless it is GPL and so is Java, or it is GPL and it doesn’t depend on Java.
For a company supposedly embracing open source, Sun is showing a lot of resistance. Perhaps they should look to Novell as an example.
You don’t have any right to other people’s work, so you shouldn’t have this “sun should release X,Y,Z or they are evil” attitude. Sun have released a lot more software into the OSS domain than probably any other company. Novell, likewise, have a business model that depends on value added proprietary technologies on top of an OSS base. Why aren’t you attacking them?
If you want an OSS Java, write it yourself, or shut up already, or harass somebody else (like IBM) to open source their JVM just for a change.
Java needs the open source world a hell of a lot less thnn open source people need Java, so perhaps its about time they brought somethign to the table other than this bad attitude that reeks of a sense of entitlement.
3ddesk
Java needs the open source world a hell of a lot less thnn open source people need Java, so perhaps its about time they brought somethign to the table other than this bad attitude that reeks of a sense of entitlement.
That is exactly why we are weary of Looking Glass, Halb. Free/Open Source got to where it got, because it was not dependant on proprietary key infrastructure or a single company. To adopt any technology that could replace a FREE key piece of infrastructure, (read Desktop Environment), with a piece of technology reigned by the whims of one proprietary company, would be extremely foolish.
As a technology this is interesting, but if it is not Free, we don’t need it as a contaminant on a strategic position. The same problem lies with Java and Mono. It is a possible venue for proprietary control to encroach upon Free technology. To protect what we’ve got now, FOSS needs to keep the key infrastructure “clean”.
Proprietary only mixes and matches with FOSS in niche applications. To think otherwise is to sacrifice the sovereignty of FOSS in the long run. This may sound extremist and you can label me a zealot because of it, but remember, you cannot have a publicly accessible house if it is surrounded by impenetrable military-grade fencing.
If Java and related technology cannot be free and FOSS needs that kind of functionality, FOSS should replace it.
Development at http://www.3dwm.org/ has been dead for a while.
Even if Looking Glass is Open Sourced with a Free License, the key technology is still encumbered. FOSS does not hold any free license for the use of Java and/or it’s related patents. A Free Looking Glass remains (mostly) proprietary through it’s underpinnings.
Maybe Sun should lower its focus and try to make Java win the
battle for creating ordinary gnome desktop applications
before rabbling on about the future.
And how is focusing on the future a problem? 3d is not a vision or a dream, it is a reality in our future. There will be innovation. Why is that a problem?
Oh, and so I don’t neglect Gnome; Gnome will be fine. There are some very capable leaders guiding Gnome development. That of course is an entirely different subject. If you honestly feel that they are leading that project wrong, then you yourself should probably speak up, and not require an executive of Sun to devote his efforts to something unrelated to his specific job.
As a general remark on this thread:
I’ve been reading these negative remarks. I just can’t believe so many people are so readily discouraged about a word or two regarding something new. You all sound very similar to my Papa when I told him about “something new.” He hardly let me finish the word “new” before sighing in disregard.
Hey now, Papa, back to earth! We’re evolving down here.
Anyhow…
These ideas are very interesting. The presentation in the Looking Glass demo is somewhat prepared of course. Anybody is capable of building a presentation based on a few ideas. It will be of more interest to me to see their innovation truly take on form.
You seem to have some problem with your father to sort out 🙂
Just kidding. Your comment was just and fair.
I can’t believe the criticism here. A company is trying to do something new and possibly innovative.It seems like a worthy effort to me. Adding 3D to the desktop will certainly make it a pleasent environment to work in. If it is not more functional it certainly seems AS functional as 2D. There is much to be said for aesthetics in computer use. This is being developed for Linux, but targeting this at business is not productive. This effort needs to be squarely centered on the home desktop user. This would surely hand the desktop crown to Linux.
Open source apps can be developed on closed source languages. The purist attitude is not practical, because it is corporationns that will bring linux too the masses. It’s not just for geeks anymore. There will always be Debian for that.
I just had a filosofical insight. Every license type
is trying to build its own universe. Microsoft is taking
third party programs, rewrite them and distribute and therefor
expanding their license universe.
FSF people is doing the same, I can’t use that, its XXX licensed or depending on XXX license.
Where does it all end? Every piece of software duplicated
in all flavour of licenses?
Maybe its just my old-man way of thinking as someone just commented.
A comment on myself:
But then then there is true innovation.
For XXX licensies to get angry that aplication AAA is
only available under license YYY, aplication AAA has to have
been innovated in the first place.
Sorry for the bull. I will be quiet all evening now.
“Our cognitive psychologists ran user studies to see whether the TaskGallery successfully engaged the knowledge and abilities people use to navigate physical space. Study participants told them that it seemed more natural to them to place tasks on the walls of the Gallery rather than on the floor or ceiling. Users were also very adept at remembering the depth ordering of tasks. This suggests that people feel compelled to treat the Gallery like a real-world environment, rather than just as an interesting desktop theme. In general, users preferred the TaskGallery to the existing Windows interface. Further studies are ongoing and will attempt to tease out the various advantages and qualities of using a 3D environment for normal productivity tasks.”
That’s an investigation on TaskGallery, not Looking Glass! I’ve seen both demo’s and TaskGallery is certainly not the very same…
(For the record i do find these projects interesting.)
3D WM’s / DE’s
Freemind: http://freemind.sourceforge.net (not exactly 3D, but a different approach)
3DWM: http://www.3dwm.org
3dDesktop: http://desk3d.sourceforge.net
Threedsia: http://threedsia.sourceforge.net
Zion: http://www.talisman.org/~erlkonig/z/img/z/z-tty-early.jpg
3D Filesystem Browsers
SGI Buttonfly (similar program used in the movie “Hackers”)
SGI FSN (for 5.3 IRIX versions and earlier only; used in “Jurasic Park”)
FSV: http://fsv.sourceforge.net (FSN for Linux)
TDFSB: http://www.hgb-leipzig.de/~leander/TDFSB
XCruiser: http://xcruiser.sourceforge.net
3DOSX: http://www.acm.uiuc.edu/afs/sig/macwarriors/www/projects/3dosx/inde… (for MacOSX)
as a system administration tool *g*
http://www.cs.unm.edu/~dlchao/flake/doom/
That info comes from Microsoft. Microsoft has been promoting its OS as becoming easier and more user friendly while the truth is that it is becoming more cluttered and requiring more effort to do simple tasks (ie: they continue to bury things deeper and deeper in cascading menus and views, as well as outright hiding things under the assumption that they are providing less clutter).
Microsoft does not seem to do legitimate research. Either that or they are not seeing the results for what they are. They are seeing only what reenforces what they believe and want to believe. I’m not going to do a point by point analysis here because no one is paying me to spend the time and effort.
You don’t have to trust me, though. You can do your own research. The Internet is filled with university research results. Specifically, research into cognetics, GUI design, human-manchine interface design, human sight, instructional design and psychology, etc. Just avoid anything that comes from a commercial source (or at least take it with a huge grain of salt).
If you want some quick reasons without the references to back it up: 3D interfaces are inappropriate for any task that is not a real world task. LCD and CRT screens cannot produce 3D visuals. 3D representations on a 2D screen are fine for games and design, not for interfacing in tasks where the majority of content consists of 2D materials (like text, video, pictures, etc). A 3D GUI is wasteful of resources (including visual space). etc, etc, etc, etc….
Purist Thinking is what made FOSS possible in the first place.