New desktop design / graphical OS methods?

Well, Croquet and LookingGlass are both full operating systems themselves, rather than just hacks. Although Croquet can currently be run withint Squeak, rather than formatted onto a drive partition, it's final state will be a fully-fledged operating system. Although I can't find it at the moment, on Sun's website (sun.com), there's a QuickTime video of Sun's presentation of LookingGlass at JavaOne in San Francisco. It really is impressive, and I believe that it will be the next big operating system "trend". It's the future, to be honest. Longhorn isn't 3D, but most likely the next Windows operating system after it will be. Of course, this is all just my speculation...

I doubt we'll see 3D operating system go mainstream soon. While I wouldn't have a problem with it, much people already have a hard time using the mouse in a 2D environment, and just won't be able to interact with a 3D desktop in a productive fashion.

I, for one, while I think I would be able to use one, would most definately prefer using a 2D desktop rather than a 3D... study ? [grin]

Quote:

Original post by xMcBaiNx I doubt we'll see 3D operating system go mainstream soon. While I wouldn't have a problem with it, much people already have a hard time using the mouse in a 2D environment, and just won't be able to interact with a 3D desktop in a productive fashion. I, for one, while I think I would be able to use one, would most definately prefer using a 2D desktop rather than a 3D... study ? [grin]

What I'm thinking is, how would one design an operating system in 3D that would be more productive than a 2D design?

Quote:

Original post by xMcBaiNx I, for one, while I think I would be able to use one, would most definately prefer using a 2D desktop rather than a 3D... study ? [grin]

Oh, xMcBaiNx. You're just full of humor [grin]

I think sphere XP is stupid. I dont like how you have to double click on a window to bring it in front of you to use it.

Quote:

Original post by Katta What I'm thinking is, how would one design an operating system in 3D that would be more productive than a 2D design?

Emphasis mine.

Don't you mean a user interface? An operating system is abstract and independent of dimensions except as far as concerns user interaction.

There is no inherent usability advantage in 3D, though. In fact, there is an inherent usability disadvantage in most contexts. The fundamental issue of user input in a 3D environment remains, by and large, unsolved.

There also isn't any particular functional benefit in 3D for most things we use computers for. Aside from the types of things we generally apply 3D for today - video gaming and entertainment, visualization and rendering - 3D, structurally, doesn't improve the operating environment. Word processing in 3D? Spreadsheets, diagramming, modeling, draughting? These are all applications that derive significant advantage from the abstraction that 2D requires.

If you want to innovate in the user experience/user interface category, think about eliminating barriers to activity. Why do I have to look for a program in a menu? Is there a more intuitive way to accomplish this - automatically launching a drawing context once i pick up the "pen" or "brush" tool that comes with the computer, or when I pull the drawing slate near?

Can my direct gestures, my gaze - can these automatic functions form the basis for a new generation of user interface, so I don't need the mouse or the scroll wheel or...?

Can voice/speech input truly augument the keyboard (accept that replacing the keyboard is a much longer-term enterprise, and will be infeasible until all input we choose to enter is either automated, like "sales figures for Q3", or "natural" - no textual programming languages), such that I can dictate notation to my computer in a noisy environment?

But there are many more, less obvious, aspects of user experience. Can my entire filesystem be versioning, allowing me to look back at how my work has evolved, branch it, integrate the best features of various drafts?

Can every application provide an automation interface, so I can tell my media player, for instance, to loop from 02:15 to 04:57 of the video I'm watching six times?

Can I have a truly reflexive media environment so that I can create my own expressions based off of any media I consume, seamlessly and without needing to buy or launch separate, often-overpriced applications?

These were the questions I asked myself when I realized that my computing environment was something I had to work around as opposed to something that facilitated seamless concrete realization of my ideas. These are the questions that I have committed myself to at least trying to answer on my own time, and if I find tenable solutions, sharing with everyone else.

Quote:

So I began to wonder, am I not being as productive as I could be?

You're sitting on your ass in front of the computer, and you need to ask this question? Seriously...

I don't know about you, but my desktop is productive enough for me. There is no sticky notes on the desktop (there are some on my monitor), but I can easily open files and programs and switching applications. Here's a screenshot:



You see those icons on the bottom? Quite like Mac. I am surprised that I can drag and drop a file or picture from the Internet to one of those program to open it.

Quote:

I think sphere XP is stupid. I dont like how you have to double click on a window to bring it in front of you to use it.

In its present form I believe it is nothing more than a momentary novelty. I could see myself 'loosing' applications very quickly. And since it doesn't replace the taskbar, I am forced to ask what the purpose is. I minimize something if I'm not moving it, so...

Quote:

Origional post by Oluseyi Stuff.

Yeah, I guess that's what I meant. I was pretty groggy for some reason when I wrote the post.

I really can't see how a full 3D desktop could help someone much, besides those few people using highly optimized work environments.

What I mean is, I suppose, can 3D be used successfully to augment and improve the current method of desktoping, and if you were designing a UI from scratch what would you do different.

But yeah, your post has a lot of good points. More intuitive UI's would bbe nice. Move your mouse toward the Photoshop icon, and a pull out menu appears with a list of options on how you could open PS, etc.

Quote:

You're sitting on your ass in front of the computer, and you need to ask this question? Seriously...

Eh? You are too, I would suppose. :)

Quote:

You see those icons on the bottom? Quite like Mac. I am surprised that I can drag and drop a file or picture from the Internet to one of those program to open it.

Well, that's a feature I never knew Windows had. Heh.

Where's your taskbar?

I have found that increasing my desktop realestate has dramatically improved my efficiency. The sphere thing sounds like it is an eye candy vertion of multi-virtual-desktops. I would be all for a replacement of the mouse being an eye scanner. I find that I can be most productive if I dont move a hand off the keyboard... Which is also why I have no desktop icons or start menu. I simply have a large list of hotkeys. I know this isnt ideal for most people, but for developers, or anyone who spend a lot of time on the keyboard, it seems to work well.

I think that we are also fairly limited by our displays. I can't wait until I can get a 4' OLED screen... I guess most people dont need so much screen space either, but I find that one thing I do which is inefficient is management of the many windows I tend to have open.

From my posting this, it seems to me like the requirements of GUIs of the future 'normal user' and 'programmer/keyboard/multitasker' are going to branch out pretty far apart.

I think a dream GUI would be:
HUGE screen with rediculous resolution
keyboard with htokeys for everything
eyescanner replacement for mouse.

I'm not sure it could get much better than that.

...

Dwiel

Quote:

Original post by Tazzel3D I think that we are also fairly limited by our displays. I can't wait until I can get a 4' OLED screen... I guess most people dont need so much screen space either, but I find that one thing I do which is inefficient is management of the many windows I tend to have open.

A stopgap measure is available today. It's called the Macintosh.

Windows applications, even if they are MDI, all have a background "application window." Microsoft also decided that floating toolbar palettes were a Bad Thing™, so every app is basically designed to take over all of your screen estate.

You know what X Windows feature I miss more than any other on Windows? Being able to switch input focus without bringing the window to the top of the stack. For me, I often find it useful to keep an eye on one window while working in another. Having every application independently implement "Keep on top" isn't a robust solution.

The stopgap on the Mac is first the fact that all applications share the Menu bar, its contents varying with the window/application in focus. Second, applications don't have that background window to occlude all other document windows. Essentially, document windows float. This allows you, to a certain extent, to keep an eye on another application's window while working.


On a more philosophical level, the ideal approach, in my current opinion, is to view the desktop as the "application" and each application as a "plugin".

I think there might be advantages to 3-D user interfaces, because it saves screen real estate. Example:

I currently have 5 icons at the top of my browser: Back, Forward, Refresh, Stop, and Home. I can picture a replacement of this using a single sphere. Clicking on the sphere and dragging left goes Back, right goes Forward, up to go Home and down for Refresh. Just clicking is the same as Stop. Moving the mouse over the sphere would unwrap it, showing you the icons if you forget. This would save 4 buttons worth of screen space.

I would also like it if every window in the UI could hide the menus at the top and show them as overlays when you alt-click, like in Maya.

This is a bit of a stretch - literally - but if you could warp windows it would allow you to make some useful, wacky effects such as magnifying just the text around the cursor in your word processor so it becomes more legible while still being able to see the page layout.

Windows can pop or shake to get your attention. Windows could change colors.

Most of this can be done with pseudo-3D, but it'd be easier with real 3D and you'd need the accelaration anyway. If it were up to me, I'd put all UI processing on its own chip so that CPU-taxing applications don't prevent me from using my desktop.

Tom

Quote:

Original post by ParadigmShift I think there might be advantages to 3-D user interfaces, because it saves screen real estate. Example: I currently have 5 icons at the top of my browser: Back, Forward, Refresh, Stop, and Home. I can picture a replacement of this using a single sphere. Clicking on the sphere and dragging left goes Back, right goes Forward, up to go Home and down for Refresh. Just clicking is the same as Stop. Moving the mouse over the sphere would unwrap it, showing you the icons if you forget. This would save 4 buttons worth of screen space.

Why not just implement that as a square on a current 2D browser interface? Click and drag up, down, left, right... yep, all doable right now.

That's not a 3D interface.

A real 3D interface is not the use of some DirectX/OpenGL to add a little flash to a traditional interface. A real 3D interface involves operating volumentrically rather than in a single plane.

Quote:

Original post by Oluseyi Windows applications, even if they are MDI, all have a background "application window." Microsoft also decided that floating toolbar palettes were a Bad Thing™, so every app is basically designed to take over all of your screen estate.

This is not true. Any Win32 app can open up as many windows as it wants to (Instant Messaging apps do this, for instance), but it would be running counter to the Windows philosophy of "let the user know which app the document is running in". I find that dozens of floating documents clutter up my workspace. However it would be nice to be able to detach windows from one "app window" and regroup them and their associated toolbars into functional groups. Unfortunately COM just won't be able to handle this anytime soon.

I'm not trying to flame here, but I have no idea how any of what you just said relates to the original topic.

Tom

Quote:

Original post by Oluseyi Why not just implement that as a square on a current 2D browser interface? Click and drag up, down, left, right... yep, all doable right now. That's not a 3D interface. A real 3D interface is not the use of some DirectX/OpenGL to add a little flash to a traditional interface. A real 3D interface involves operating volumentrically rather than in a single plane.

I agree that it doesn't have to be drawn in 3D to work, but it *could* encapsulate 3D information if it uses the intuitive human concept of sidedness, which would (I am arguing) make it more accessible to a less sophisticated audience. The major problem with new computer users is that they don't know what elements on screen need to be played around with in what ways, and that can be the difference for instance between an alien-looking scroll bar and a more natural Acrobat-style grabbing-hand interface.

Tom

Quote:

Original post by ParadigmShift This is not true. Any Win32 app can open up as many windows as it wants to (Instant Messaging apps do this, for instance), but it would be running counter to the Windows philosophy of "let the user know which app the document is running in".

I'm sorry, did I give the impression that it was impossible for the application? I assumed that everyone here would be familiar with the fact that Microsoft aggressively pushed that style for the majority of applications with the Windows 95 Logo Compliance initiative.

I apologize for the confusion.

Quote:

...lit would be nice to be able to detach windows from one "app window" and regroup them and their associated toolbars into functional groups.

That's actually a very nice idea.

Quote:

Windows can pop or shake to get your attention. Windows could change colors.

*twitch*

No, not that, anything but that.




I like the Pseudo 3D all-in-one-icon idea for a web browser though. While it isn't true 3D, it would save a lot of screen estate. This could also be integrated into a hybred of those circular context menu's, and mousegestures, thus eliminating the need for those buttons on the title bar at all. Edit: though, this may not be completely efficient, until you get used to it.

While doing some outside work, I had the idea that when you push the mouse over to the far left and far right, the desktop would 'jog' in that direction. Much like an RTS, I guess. Rather than having 'hide away' toolbars, the desktop would actually be bigger than the screen, and everything could be neatly hidden to the left, right, top and bottom. Just push the mouse up to reveal, or , push with a combination of middle-click (that's what I would use).

'Twould be a great spot to store the AIM buddy list, and Winamp.

@Katta:
My taskbar is hidden at the bottom.


Normally when people are working on a desk, they have stuff randomly placed and not interfering with one another. I do agree that applications in Windows, or even Mac, sometimes occlude other applications, giving it a sense of single-tasking instead of multi-tasking. You can rearrange the windows, but most applications take up the entire desktop, unless you resize the windows there is no way you can keep an eye on three or more applications at once.

So IMO instead of putting all running applications in one place and hide its interface (Windows in taskbar, Mac in menu), they all should be placed on the desktop. When application is minimized, its "application window" is resized to a much smaller size but still keep it visible so that user can still monitor what's going on. These minimized application windows will be floating around the desktop (can be moved around or arranged like icons), maybe giving them 30-50% transparency and stays always on top. Something to this effect:





3D interface is not necessary until we have developed holographic operating systems. The screen itself is 2D, it really is not going to increase productivity of any kind. What xMcBaiNx said, some people even still have troubles operating the mouse in a 2D environment. Why add another dimension?

Quote:

Original post by Tazzel3D I would be all for a replacement of the mouse being an eye scanner.

I don't have much to add, but I would like to say that eye scanners will probably never be particularly robust input methods. It would probably be limited to the way RTS games behave when the mouse gets near the edge of the screen. Extreme looks will cause pages to turn or whatever, but slight looks will do nothing. Your eyes do too much on their own to be much more precise than that.

CM