Slashdot Mirror
Commercial 3D UI and for Linux
Lord Carmack
wrote in to point us to
Objective Reality's homepage which claims that it
will be releasing an OpenGL based 3D interface for linux.
There isn't much there, but it at least looks interesting-
although it also looks commercial so I'm skeptical about
how much acceptance it will see.
No, I agree with the first post, indeed there seems that alot of asholes think that just because a application is commercial, it is evil or will never make it on Linux, BULSHIT !!!, that is the kinf of shit that prevents some other companies from developing to Linux, not everything is F@@@@NG free, I use Softimage for my work, it costs 10K, I can't hardly whait for them to port it to Linux, and if they ever do it wil cost the same 10K, but I guess for most people it shuld be free, right ? well it is not, and it wil never bee, I think it's great inux is free, but whe need more apps now, and some of them are going to be COMMERCIAL ones, and I for once don't see a single problem with that, I pay 10K for SI because I make money with it, if Im in a business and I need a app. to run my business and that ap. costs money .. wel, I'm just gnna HAVE to pay and shut up about it ... get a hint people...
Yeah..why are they "charging" their os?
they are more pricier than beos.
don't gimmie "download" excuses.. when your harddrive dies.. you still can download it right?
duh!
CAVE's suck. stereo glasses suck. trust me..i have 2 immersadesks and 1 CAVE 15ft from where i work and they *suck*. i'd much rather use my 19inch 1600 x 1200 32bpp monitor than a low res backprojection system (i.e. immersadesk/cave), crappy stereo glasses which stop working when you move 4ft away from the system (it has weird RF limits which force the glasses to unsync themselves from the screen -- this makes you see a double/flickering image).
did they overhear a conversation I had...? Ever so often, it is humbling to see that other people come up with exactely the same ideas as you.
Its still available to download, and I think a new version (which allows you to run it on Irix 6.x).
Check it out here.
A while back I was going to invent the internet, but got busy and before I knew it, Al Gore had beat me to it. Bastard!
If you can move to a 3d environment and still run all your old X and tty-based stuff, then there's no good reason -not- to adopt the new technology.
Which is precisely why X has xterm.
And why a 3D environment -must- support X. Fortunately, that's not that hard - just map it to the side of something and you're there, either through GGI, or rewriting Xnest.
-Alex.
is that there is no current plans for a MacOS or M$ port :) me likes :)
"There is no spoon" - Neo, The Matrix
I looked at the screenshot along with some of the others mentioned in the posts, which tend to show traditional windows floating in free space...My reaction was "what's the point?"
I believe there are really only two reasons to advance technology: to make it easier to do "old" things, and to enable completely new things that were not possible before.
Can a 3D interface improve on our 2D interfaces (i.e. WIMP-style interfaces like Finder and Explorer)? I don't think so...it looks like the best efforts at going down this path so far are just giving us different ways to organize windows (arranging them in cubes and panels, stacking them along z-axis, etc). It feels like applying 3D concepts to today's interfaces could make the WIMP metaphor slightly nicer, but not by much.
If we're going to play with such a different technology, why not try using the tech to do new things with interfaces, beyond "pointing" and "clicking"? Even if these things don't happen anytime soon, why limit our creativity now by following the same old paths?
We might not necessarily see a need for 3D interfaces on our desktops now, but if we broaden our minds, I'm sure that we can find lots of "new" uses or methods that weren't feasible before with flat interfaces.
Acceptance of 3D really depends on 2 things: 3D i/o devices to make proper use of a true 3D interface (holographic displays? mice with accelerometers on all axes?) and a really good example of an application where a 3D interface is clearly more useful than a 2D interface. 3d i/o devices will come eventually/soon (mostly because of games and CAD), but the examples of 3D interfaces I've seen so far haven't made a very convincing argument.
-- John Truong
Frankly, if DOS is all you think of when you think "shell", you aren't quite qualified to offer an opinion.
Well, expressing data in more dimensions can be good, and helpful, but only so long as you're not exceeding the dimensions that the viewer can actually experience (in number or in type).
Three-dimensional workspaces would probably work better than two-deimensional ones, but only if the user has the capability of truely experiencing the 3-D reality.
If you have three rectangular items, it might be best to join them all together like the corner of a box, rather than laying them out side-by-side-by-side, or whatever. If all that you can experience is two dimensions (for example, you 're viewing the universe through a CRT), then you can't really see depth--you can just get an impression of it.
If we take a plane (or the [inner] side of a cube) and rotate it, so that the surface becomes more parralel with our line of view, in two dimensions, everything on that plane appears to shrink along one axis, and either shrink or elongate along another axis, which can be... unpleasant; in three dimensions, we see that things are moving away, or toward us, and the mind adjusts the perceived sizes of everything, so we don't get that strange, irritating, stretchy-shrinky effect.
The `non-flat planar' layout of workspaces is the only advantage that I see, in a 3-D workspace, with regard to presentation of data that doesn't naturally occupy at least three dimensions--being able to place more items side-by-side, but have them occupy the same amount of (or less) space, is great, but being able to push and pull 2-D windows to give a better `document P is make occult by document Q' feel is rather worthless as far as anything besides aesthetics goes, and may actually act as a hinderance to `productivity', by adding complexity and time-consumption to the navigation of the workspace. This is big problem that I've always had with mice and other scrolly-pointy-clicky things--in order to perform an action, you need to move to a specific location, and, to move to another location, you need to traverse all of the space between it and your current location. Three-dimensional touch-sensitive displays would probably speed things up;), but I think that I'll be sticking with my keyboard until I have wormholes in my workspace, or the workspace responds to thought....
-rozzin.
We have already witnessed the (r?)evolution from CLIs to GUIs. The world is 3D, so why shouldn't you interface with a computer the same way? I see this leading to more intuitive UI's. Maybe this will all lead to holographic computer displays, or fully immersive virtual reality. As far as pointing devices, we have gone from keyboards, to mice, now what? I can't remember where I saw this, but I have seen a sort of "3D trackball..." The ball from a trackball on a curved surface, that is used to manipulate machinery in 3 dimensions. If someone knows what these things are called, please reply.
:)
Basically, this is just progress. This particular 3DUI probably won't be very good, since it is the first. But *this* is the future.
So, who is going to be the first one out with an open-source 4D UI?
You obviously don't live on Bainbridge Island like I do, where the ferrycam.com is a lifesaver. Bainbridge is beautiful, but when you want to go any further than downtown seattle on a regular basis, you're screwed. By the way, why the heck are you talking about this here?
SpaceTec's "SpaceBall" is a neat solution which might fit this well. Rather pricey, at least the version sold to CAD Engineers is, but I think there is also a cheaper version, which for all I know varies only in the included software. (I really am not sure).
cheers,
timothy
Apparently Objective Reality is suffereing from the same illusion most GUIsers do. That M$ created the Universe, Al Gore created the Internet, and that efficiency comes on its own.
.25mm fab in under $1Million and leave the FAD, and FUD to marketroids.
I used to love how Wing Commander ran under 640k in the early days.
And have you checked emulators yet? The gigantic storylines in games like Final Fantasy I fit in 256k and the emulator itself is hardly 100k.
Seriously write a How to build an Open Hardware
The ship sank. Get over it. (This sig was cut out from another's shirt and painstakingly hand-posted)
Honestly, I can't see any possible advantage to a 3D UI other than the coolness factor. I mean, come on! What kind of pointing device would be required to work in a 3D environment? Do you really think that it would be more intuitive than a standard "2D" environment? And 2D in that sense is really something of a misnomer. Afterall, can't you stick one application behind another already? It's only 2D in the sense that you are always looking at your applications straight on. And that's a GOOD thing. Why on earth would you want to see the side of an application?
And think of the resources that would be wasted on a 3D UI. Imagine the processor power that would be required to do even the simplest tasks. What is the point?! How would this be better than the GUIs we already have? Imagine how difficult it would be to use a mouse to select an object in 3D space. Why, why, why?!
OK, I'm starting to rant. But the major criticisms I've seen in the other posts I've read is that it's not open source. Nobody seems to have mentioned that it's an incredibly stupid idea. OK, it looks cool, but that's about it.
You can take my command prompt when you pry my keyboard out of my cold, dead hands.
I am not an idiot. Please use my name to email me.
"That's right, I'm quoting myself."
-Upsilon
there's a movie called "hackers"?
wow, i bet it's incredibly bad!
i wonder if the only redeeming feature is angelina jolie.
Back to remedial training with ya...
--
Ok, a 3D UI *sounds* incredibly cool, but how will it actually improve the UI? The only 3D effect in the screenshots was showing windows tilted at an angle... that just makes the windows harder to deal with!
Seriously, what can a 3D UI do that a 2D UI can't? (Not counting stuff that makes the UI less useful, like windows at 45 degrees.)
This will come out for BeOS too. And LinuxPPC. But no plans for mac and windows http://www.oreality.com/synapse_pr.html
---
We do have stereoscopic vision, but it's rather crude -- limited to close range, and with rather poor resolution. I don't think it offers much, especially since the necessary goggles would be a PITA.
Note that any data-representation will still be essentially 2D -- you can represent a 3D surface, but only a surface. You cannot represent the interior, because the surface will hide it. Again, we only see 2D, because a surface is just a twisted plane (the twisting has some informational content, but not a whole dimension's worth of it).
It is amazing, when you think about it, that we understand three dimensions at all. It shows there's a hell of a lot more going on inside our heads than our I/O can express.
it would be efficiant to move around like I do in a quake game. Mouse for y axis and spin and keyboard for z and x access. when you want to click on something, just click the second mouse button to take it out of move control and render a 2d curser. use this curser for fine control of objects.
I responded to someone wanting to add 3D to Gnome a while back. They felt you could express more information with a third dimension, but they overlooked the fact that human vision is inherently 2D.
In an already 3-dimensional context, it would be helpful to have a 3D extrapolation of the interface -- something that the aforementioned 3dwm seems to be trying to do. But to put a 3D interface onto a 2D display is just glitz.
The human mind does have a proficiency at creating an internal model of a 3D situation, even though it is only perceived with two dimensions. However, while this is useful for understanding inherently three-dimensional situations (as in CAD, for instance) it is not a good way at dealing with other information.
People naturally organize things in a two-dimensional fashion when given the choice. Be it shelves, stacks of papers, tabular information, etc. It is easier and more accessable.
While there are certainly more innovations left to be made in interface, the new directions are much more subtle than 3D.
We don't nearly as much press (in fact, almost none to date) compared to some of the other X projects such as Gnome or KDE and no external funding, but we do progress.
I think the difference is that we have basically gone through the low level architectural stages. We have picked a base design which is powerful and yet flexible and now that that's done, developmental pace is starting to pick up.
Of course, a little PR never hurt and we could certainly use more developers and some funding.
I believe in the end, X is doomed. It's controlled by a closed consortium, has 15 years of baggage and is the essense of monolithic programming.
It provides very few of the modern GUI services that Windows or MacOS provide, so what you see is window managers and toolkits taking up the slack attempting to add these services, basically fracturing X the say Unix was fragmented. You get programs which have been reimplemented to work on a specific toolkit or specific window manager over and over again. Duplication of work is common, and unfortunate.
As far as 3D interface, it has been spoken of. Unfortunately there has been no clear model for 3D interface for the 2D desktop world we live in. Microsoft actually has shown a 3D interface demo which was quite interesting as have a few other companies, but none of them were complete.
--
The world is neither black nor white nor good nor evil, only many shades of CowboyNeal.
this looks very similar to the Harmony Hyper-G browser I played with a few years ago
Take a look through this: RealPlaces.
It's very interesting.
-- John Truong
If it ain't free then it can be bought; I expect this company to be bought by Microsoft!
The future of Linux, of which I doubt this 3D Operating Environment is, should not be based on a commercial product that can be usurped so easily.
I laud them on their apparent hard work but... unless they are extremely dedicated to Linux and BE they are targets for assimilation!
Codifex Maximus ~ In search of... a shorter sig.
What does 2D GUI components mapped into 3D space provide, other than the backside of a window, that cannot be done in ordinary 2D environment? This whole concept sounds like a slashdot kiddie-bait.
Antti S. Brax - Old school - http://www.iki.fi/asb/
Of course they're biased...
against anything not Linux...
against anything not open source...
idiots
Ok, but you're still talking 2D. I'm not saying CAVE is greater than a 19inch 1600 x 1200 32bpp monitor. You're talking about technology and you have to accept that real good 3d environments will exist soon. In 3D environments you want 3D widgets.
>beos will be the alternative os.
Want to see the future of beos? Look no futher than the Amiga. The BEOS guys are walking into the exact same minefield that the Amiga crowd did with their obsession with mulitimedia.....
>Man, i was using a 3D UI when i was h4x0r1ng gibsons back in the day.
Actually, it's nothing but Microsoft Bob revisited. I shudder to think of the kinds of industral disasters this kind of interface is going to cause at say a chemical plant when something goes wrong and people start to panic and begin fumbling around within the UI.
Can you imagine an ATM with 3D widgets :) It doesn't look open source so I doubt it will get on the desktop much but it still looks useful for commercial companies who have money and need a cool UI for a specific application.
Ever since I saw the movie "Jurrassic Park" I thought a 3D GUI would be neat. It is taking the inefficiency of a high-graphics UI another step further and making it a high graphics 3D UI, that can't be good for performance. Also, it becomes difficult to interact with a 3D UI on a 2D screen with standard 2D input devices, I'd somwhat like to know what tecniques they intend to use to solve this problem.
-- The act of censorship is always worse than whatever is being censored. Always.
Damn once again I wish I would act on my ideas earlier. I had this idea about 4 months ago and was already to code it to until a stupid assignment got in the way. Oh well... I hope they make it as cool as I thought it would be : (
Benno
Check out the 3Dwm website for something quite similar... This is a three-dimensional window manager for X with OpenGL support. It's still in early development, though. Another cool app is GLACE, which supports running X applications on 3D surfaces.
IMHO most 3D widget sets in the near future will probably act like their 2D counterparts, just with 3D special effects such as maybe windows coming towards you as they are maximized. In these cases it would seem to be easier just to double buffer, scale, and modify the widget data while the effect occurs instead of using a real 3D widget set.
Whoa... look at the pic 'bigp.jpg' in that same directory
Ah-nold Tuxenegger!!!
Skip
--------------------
flifson@csdotuctdotacdotza
Skip
--------------------
"To create an apple pie from scratch,
you first must create the universe."
This looks a lot like the 3D ui demonstrated by the GGI folks:
. jpg
http://www.ggi-project.org/images/cube_withxggi
From what I've read about Berlin and its predecessors, it seems very well architected and worthy of development. What amazes me is that it continues to progress, albeit quite slowly.
-- Too lazy to get a lower UID.
Well, it's nice I guess.
So lets take the low emory footprint Linux and overlay a 3d interface on it.
How smart is this?
It's a thankless job, but I've got a lot of Karma to burn off
I'm sure that a lot of people already know this, but for those who don't....
http://www.berlin-consortium.org
The Berlin Consortium appears to have similar
goals( altho it's a little hard to tell from the
somewhat vague PR by Objective Reality ) w/ the
advantage of being GPLed. Berlin has the
disadvantage of being rather in young in
devel. terms tho.
Berlin will be a window system, somewhat similar to X but drawing with OpenGL, and with a single built in abstract widget set with swappable implementations. (hence getting both consistency and user-configurability)
Initially as I read the site it will be optimized for drawing 2d wigets onto a 2d plane, but I see no architectural feature that would prevent the widgets being swapped for a "theme" that draws them in true 3d.
I'd be curious to see how they manage windwos 3d accelerated applications. It's not an impossibility, but I can imagine that having the accelerator render both the window manager and an application at the same time will steal clock cycles from both and cause an ovreall system slowdown.
Man, i was using a 3D UI when i was h4x0r1ng gibsons back in the day.
In our current 2D environments we sometimes have to use the "Lower Window" command, which un-obscures other windows which can can then manipulate. Not as effective as having 3D input, but it works. Further, some users aren't going to be comfortable using a 3D cursor.
So you start with what everyone already has and is comfortable with, and then you add additional input options as time goes on.
who wants to pay $100 for a GUI that isn't compatible w/ X... I'm sure one of these open-source X-based 3D GUI's will beat it. How cool would it be to be using normal X apps in a 3d windowing system? :)
-ethan
Spent time on something more worthwhile
i don't see any problems.. :)
on beos platform that is
3D UI??
You have a screen (which is a 2D display device) and a mouse (which is a 2D pointing device).
How exactly you make a 3D interface that is more useable than a 2D one is beyond me. Remember that the 3D is all in the mind, and that conventional windowed displays aren't much less '3D'
p.s. any of this remind you of that 3D cube demo that the GGI guys did -- interesting, but hardly what you'd ever like to use (though I must admit that a rotating cube would make a nice window transition effect)
John_Chalisque
Posted by MaverickPl:
Why not get ttf fonts working for your computer? You can do it with either xfstt or the defualt xfs that comes with RH6. Also you need is a tool called ttmkfdir and so ttf fonts. It is very easy infact.
Also if you are looking for some cool fonts go to *shameless plug*
http://www.da3.net/fontworld
such a work to get the browser workin heh?
god.. I'm going surfing.. u have fun tinkering your linux.
You guess ;-)
MS Research showed a 3D UI prototype called the Task Gallery at the WinHEC conference in April, running live & working Windows apps inside a 3D environment. They're betting on bringing the legacy apps along and then augmenting the environment with new uses for 3D. Nothing public yet on the MS Research website, but it was featured in a big figure in the Wall Street Journal's special feature on managing information overload last week. ("Managing the Mountain")
I wonder what kind of carpel-tunnel syndromes the repetitive tasks of 3D pointer controls will cause.
But anyway...
I don't think a pointer is a 3D thing at all and shouldn't moved into a 3D interface. I think a cross-hair is more appropriate. Currently, We move objects on our desk but our "view" of the desk is fixed. The alternative would be a fixed position pointer (crosshair) where mouse control moves the desktop underneath. (This is like panning across the desktop when you run virtual resolution higher than your physical resolution)
People find this kind of view moving annoying in 2D but are more accepting of it in 3D.
Once better input devices are made, then work on controlling both view changes AND a 3D pointer should be put in.
The truth is more important than the facts.
-Frank Lloyd Wright
3D is better than 2D, just slightly less than 2D is better than 1D ... it is all implementation, and I have not seen any good implementation yet. Three dimensional displays have to be displayed *in* three dimensions.
(one dimenson being a row of pixels)
--Chris Brown
The only way I can see this as being overcome would be to have the mouse pointer constrained to 2D, while a second controller (keyboard maybe?) actually spins the 3D space to come to the mouse. This would seem to be very time consuming and awkward. I'm curious to see how they solve this problem.
-- Give him Head? Be a Beacon?
-- Give him Head? Be a Beacon? :P)
(If you can't figure out how to E-Mail me, Don't.
imagine working with 3d interface to get your job done faster.. without going thru the shell.. i just can't imagine going thru dos shell in the future.
... insted u going thru dos shell....The technology now a these allows us to take advantange of the faster hardward..what's the use if you got a super fast computer and still tinkering with your old dos prompt?
unless you are not looking forward on voice command technology where your computer will talk to u
why are linux peeps's mind are so 2-d.. god I wish I knew.
Beos platform also?
are you guys bias?
X-windows are slow and clunky..
beos will be the alternative os.
Linux is wayyy far on the gui enviroment.
beside.. x-windows's interface is very messy and looks very cheap.
Example: http://www.ferrycam.com/livepush.html or anythign else the requires plugins which makes up a huge sector of the WWW. This is not acceptable at all.
Gotcha -- MS Windoze user "criticized" Linux with "facts" entirely from his wild imagination.
The page uses netscape-specific server push that happens to be implemented in exactly the same manner on all platforms. Simple attempt to load that page from any Linux box will show that, and our smart Anonymous Coward obviously never tried to do that.
And if sir Anonymous Coward cares, no one uses nonstandard plugins (ones that don't correspond to known and supported everywhere MIME type) anymore -- even on Windows they impose more security risk than what they worth.
Contrary to the popular belief, there indeed is no God.
You can simulate a 3d interface on a screen by bordering. Color the top and left sides of a windows with white, then color the right and bottom side with black. You can also make it appearn the button are raised or lowered that same way. :P
I've been playing with a 3d UI for a little while; apparently a lot of us have. I'd like a lot more than just a 3d interface though. Something that brings people into the same environment, something where object have permissions so sharing doesn't have to be total, something where I can work on my house design while watching the netrek freaks blowing each other away overhead...
Here's a screenshot of a tty running in my current software, with sheep wandering below. From here I can run commands from the shell level that produce output back here in my space, like a life simulation I wrote in Perl that communicates over TCP with my Z program to produce a life grid floating a little ways back into the trees.
Oh yes, the sheep bleat when you click on them.
I got into this idea at Origin games when I ended up with a $350k SGI reality engine on my desk. Now if I just had more time...
We need a shared, networked, distributed, permission-equipped environment -- all I've seen is internal-object-only, unnetworked, undistributed, one-user, no-permission environments. Even XEROX PARC can't seem to get two applications running in one space, according the little posted on their website , at least. Don't you want the tension of trying to read your email, or do some work inside of Quake 4?...
-Alex
check out the redhat web site..
"although it also looks commercial so I'm skeptical about how much acceptance it will see."
If someone wanted money for the work they did, that must make them fucking evil and they have no place in linux. I hate zealots on either side of the fence. If its good I'll use it.
Hold on there! The poster did not advocate zealotry, only pointed out that a commercial product in the linux world of mostly free / Free software would have a built-in limitation to its appeal. He doesn't claim that wanting money for work is evil or anything of the sort. I even get the feeling that he might regret the tough time that this product / it's creators will face because they are at such a self-imposed disadvantage.
... how many people use it? A fair number, I'm sure, but no question the fact that there is free competition in the form of XF86 which works well for the Vast Majority means that, just like a commercial GUI, the appeal is limited. So what?
Remember, there's MetroX
As you say, those who think it's the best tool, and can afford it, will use it. Good!
Best,
timothy
although it also looks commercial so I'm skeptical about how much acceptance it will see."
If someone wanted money for the work they did, that must make them fucking evil and they have no place in linux. I hate zealots on either side of the fence. If its good I'll use it.