Slashdot Mirror
Simon Phipps Looks At 'Looking Glass'
CitizenC writes "Simon Phipps, chief technology evangelist for Sun Microsystems, describes his experiences using Project Looking Glass, Sun's prototype three-dimensional computer desktop, in this post on his weblog. He mentions a couple of demo videos too."
I saw a 3D video display at the CEATEC conference back a few months ago here in Tokyo. Several makers were showing them off, but they really weren't anything to write home about. I guess it was cool enough that they were able to display a 3 dimensional object or scene with only 2 dimensions, but it's hard to see what the practical use of something like that would be outside of "You're my only hope" type messaging.
Likewise, 3 dimensional computer desktops are not going to be taking over the world anytime soon. It is hard enough teaching people to use the mouse correctly in two dimesions. Trying to wrap people's heads around a 3 dimensional workspace looks to be virtually impossible.
I have been pwned because my
Shame the "high bandwidth" video is in RealSpyware format. If there's an mpeg somewhere, please post a link and I'll have a look.
CEE5210S The signal SIGHUP was received.
You know, I like Java technology. If it weren't for the existence of Java, we'd never have this push towards writing user space software in higher-level languages that run on virtual machines. As a productivity aid to engineers, I think it's one of the best advances of the past decade.
However, although Schwartz demo'd some clever technology, it was not very flattering. First of all, it has a little bit of the "me too" syndrome, considering that Mac OS X already has some nice eye candy that uses the same techniques: fast compositing and scaling, to run videos in an icon; translucent windows; windows that easily shift and scale without losing clarity (Expose). Heck, Microsoft demo'd their "me too" six months ago with early images of Longhorn.
Second, was it really necessary to spend the whole time bashing the "dominant" operating system provider. Believe me, I'm no fan of Microsoft, but this anti-Microsoft schtick of Sun's is becoming tiresome, unflattering, and it's not helping their stock price.
I just wish McNealy would try to compete by being better, not by complaining or firing barbs. Frankly, Sun has not been delivering great software technology for several years, so to come at it this way seems very unprofessional. Bummer, too, 'cause I really want to see Sun (and Java) succeed.
It's largely written with Java (proving for once and for all that there's no inherent performance gap for Java applications) and makes good use of the integrated Java support in JDS.
It's nice to see that Java is to the point where Sun will use it for desktop projects. It has taken a lot longer than some of us hoped, but certainly better late than never!
JDK/JRE 1.5 should bring additional significant performance improvements...
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
When you try to display something in three dimensions on a monitor, not only does it not really exist, but your brain can't deal with it. Watch computer game novices (and some experts!) try to lean their head around to peek round a corner playing a FPS game. See how quickly most people get motion sick watching someone else play a game. It's all because the visuals are faking 3D and our eyes & brain can't deal.
A 3D desktop is not going to be a feasible reality until we have a feasible 3D display to draw it on. Only if/when hologram or 3D-projection displays become a reality will there be a useful case for a desktop to match; in the meantime, this just adds unnecessary complexity to the 2D desktop.
Kind of like a big virtual desktop, only you get to peek at what's over the horizon.
Which is surely all a 3D interface could give you, more space to open more windows, but surely there is a limit to the amount of multi-tasking a user could practically do. And even then, a better way of showing windows currently open in the task manager would be more usable and effective than a 3D desktop.
About the best use of a 3D desktop that I can think of would be to show how tasks are related to each other i.e browser tabs, related office documents, disk directories etc.
One of the programs that was really the visual inspiration behind the framework presented in my doctoral thesis was 3Dtop http://www.majorgeeks.com/download186.html
A review is given at the link I have provided, but this program was really the marvel of simplicity. It is only 356 KB (yes k), and truly converts all the files and folders on you computer into a 3D space.
If you are ready to re-INTERPRET what you normally store in a folder of file, so that it now fits this visual space, it is extrememly powerful.
For a person just starting off, the novelty of 3Dtop wears off pretty quickly because you easily get lost, and it looks just like eye candy, but if you REinterpret what the folders and files should contain, it becomes a very very powerful cognitive space.
Please don't flame me if you don't understand what I am saying, but as a last point, I would like to mention that as our cognitive space (displayed in the electronic space on the desktop) has no real correspondence to physical 3D space, it becomes hard to impose "laws" (like physical laws) and hence the electronic space almost becomes infinite, and really disorienting. One way to build laws is to recognize that (in Windows) the Desktop itself is a folder, which contains My Computer, which has the C:\ drive, which has windows, which has Desktop again ... it is this Russian Doll like relationship between two Desktop views that provides the backbone for building the laws that are equivalent of the "physical space" for the electronic and cognitive space ....
To see a world in a grain of sand, and then to step back and see the beach where the sand lies
Apparantly the "Looking Glass" demo (running on a little Sony Vaio) was one of the most popular demos at Comdex. I liked the desktop layout as well - very clean and simple, but amazingly flexible.
I think it's pretty obvious what they've done - just turn each window into a texture map, then project that in a 3D environment. That's why they can flip the windows, have multiple copies etc running very smoothly (3D accelerated), and also why you can do alpha blending very easily, or have the entire backdrop being a 3D projection (eg 360 degree world view.
This is probably using the OpenGL wrappers in Java... Sun will be feeding the "looking glass" technology into the Java Desktop System over the next 6-12 months. They weren't originally going to be so agressive, but due to the huge interest, they said they decided to accelerate the schedule. One nice side benefit of this becoming a "must have" is that the 3D cards guys will probably get more serious about doing proper, complete OpenGL drivers for Linux (the current situation ain't that great).
Like some of the others here, I do wonder just how productive it would be, but it didn't seem hard to use at all. It does give Linux (and Unix since it can run on Solaris too) a very nice wow factor - the Sun guys gave it a kinda "who cares about waiting a few years for Longhorn, here's what you can do today!". Will help dispell the bad perception that Linux has for desktop use.
PS The original demo was written by a guy in Sun Japan in his spare time. Yep, a real demo...
You're not exactly right there.
It's only hard to place things in 3d if there isn't a clear system for it. If you look at the demo video, it looks like the intended use of 3d-ness of the desktop is actually just being able to turn your application windows sideways to sort of file them away. Certainly, windowshades has much the same effect, but this way you can select what you want faster, both because it's a bigger thing to click on, and because you can visually identify your application instead of having to read the title bar.
-S
I've seen this in the flesh and was lucky enough to play with it. (Friend is an employee of Sun)
It does sound like some of you didn't even bother to RTFA. One of the points of the 3d desktop is to make it easier to find your windows. Your not going to lose your windows with this technology.
I think it also shows that Sun is still trying to innovate, frankly that should be applauded. Besides, it looks way cool!
It's largely written with Java (proving for once and for all that there's no inherent performance gap for Java applications) and makes good use of the integrated Java support in JDS.
;-) you can see the CD selector from the screen-shot to the right in action, and I can imagine all the other experimental 3D apps I've enjoyed using (photo gallery browsers, SQL database explorers with 3D visualisation, etc.) finally making it to the real world.
That tells you nothing about Java performance. The performance critical portions of Java3D and the operating system's 3D drivers are written in C and assembly language. Even JavaScript and VRML manage to render 3D scenes fast.
At one level it provides a 3D windowing environment for existing X applications (interesting enough in its own right), but at another it introduces the ability to create 3D applications where you interact spatially to explore data. In the demo video (starts a little way in, persist or fast-forward
3D data visualization is an old hat, as are 3D user interfaces and mapping 2D window systems (including X11) onto 3D surfaces. There are even a bunch of open source projects around, including 3dwm.
What if windows were translucent so you could see the multiple windows you're working on at the same time? What if you could tack a note to yourself right on the Web page you're viewing? What if your CD or movie database became a 3D jukebox, where titles were joined with images to make finding what you want easier than ever?
Translucent windows have been done many times, as have annotations. 3D representations of physical objects as user interface metaphors have been done numerous times (and those kinds of interfaces generally belong into the Interface Hall of Shame), and "titles" can already be "joined with images" in some MP3 players, including Windows Media Player.
Sun Microsystems' latest innovations by its Advanced Software Technology Team will make the above scenarios a reality for the desktop of the near future.
There is nothing wrong with tinkering with old ideas and trying to integrate them into a nice system. But, people shouldn't repeat old mistakes and they should give credit to the people who came before.
Referring to such tired old ideas as "innovation" either means that Sun is ignorant or that they are deliberately misrepresenting their work.
I think 3D GUIs may have some good purpose, but the problem is that until now they have mostly been applied to things that just work better in 2D.
The clearest example of this is exploding menu-style choices from a 2D list that is easy to scan quickly and accurately with your eye into some 3D "infospace."
Imagine going to a restaurant where, instead of a paper menu of the food options in front of you, all of the waiters in the restaurant, each holding a big sign with the name of one menu item on it, form a big circle around you and you have to turn in your chair around to view each one.
Restaurants have been around for a long time and I don't know of any that work that way.
Syntax error: loose != lose, affect != effect, then!=than
You should watch the video. It certainly looks nice, though how many features you'd actually ever use is another matter entirely. I suspect many people would turn off all the chrome, and use it much as they use window based systems today.
They have miniature versions of all the running application at about 45 degrees angle from the viewing plane in a row at the bottom of the screen, so there is no chance to lose your windows as you suggested.
Most people I know do have difficulties getting used to and navigating simulated 3D environments. For example, beginning FPS players seem to have a lot of problems learning their way around new maps.
3D is a rather difficult problem because it'll have to be done in a way that models itself to the end user. Right now, many people have problems with operating a mouse, moving and resizing windows, etc. If 3D doesn't find an extremely intuitive way for managing these sorts of resources, it won't be particularly useful.
Also, although we live in a three-dimensional environment, the best science I've read so far suggests that our perception of it is largely two-dimensional. For example, when humans remember how an object looks, they don't memorize the 3D shape of the object, they memorize 2D images of how the object looks from various angles. There is research suggesting that when we memorize a truly 3D environment such as an office building, we really create a series of linked 2D 'maps.'
This situation becomes very clear in video games. Most folks don't have a hard time at all getting used to FPS games, where the action is really mostly two-dimensional. A lot of people have a slightly harder time with flight simulators, but can get the hang of it fairly easily. This still isn't really a 3D problem though - it's more of a "2 1/2 dimensional" because it's still largely 2D action that is augmented by "up" and "down", but freedom to move along the up/down axis is severely restricted so fluent thinking about the game really only requires a 2D internal model of the action combined with details about whether something is above or below oneself.
3D space combat games are where the situation really becomes clear. My own experience with them is that when I was first learning them I really didn't keep a good idea of the 3D action in the games at all. I kept an idea of where various objects were in relation to me, and didn't maintain a good concept of the overall layout of the gamespace in 3D. Not exactly Khan and his inability to strategize in three dimensions, but certainly not a fluent ability to plan and strategize in irreducibly three-dimensional environments. Watching other people learn to play these games suggests that their experience is similar.
I certainly agree with the arguement of '"3D" on a 2D space,' but in addition, I'm also wondering why we're trying to this right now anyway. In my opinion, we have a long way to go before we even have 2 dimensional desktops mastered, why are we trying to go 3D already? I think it would be better if we mastered what we already work with before we move on to bigger and "better" things...