Linux

Esfera: A fascinating UI concept for window management in Ubuntu

As usual, the Linux development community has come up with another idea for the desktop that will astound users. Read on as Jack Wallen explains the idea of Esfera.

On the Ayatana mailing list recently a new concept bubbled to the top that has some serious potential to bring to the desktop a world of new possibilities. This concept, dubbed Esfera, came about when the default Ubuntu 10.4 theme moved the window buttons from the right to the left side. This move obviously freed up all that space on the right side - a side users were accustomed to using. Seeing this as an opportunity for growth, mailing user Pablo took it and ran with it. What he has proposed really struck me as something Linux CAN do and SHOULD do. It will be another revolution in the desktop thanks to the Linux community.

But what is Esfera?

Imagine you have a new control on your desktop windows. This control accepts gestures that can do various tasks. One of those tasks would be to flip the window over, giving you a new space to use. This space could be used for:

  • Notes
  • Related files
  • Links to other open windows
  • Hidden information
  • And much more

Or, instead of turning the window, imagine the window has another layer. So instead of pages within a document, it's like pages within a window itself.

Here are some other ideas off the top of my head:

  • Quickly switch windows from landscape to portrait mode.
  • Rotate windows to any angle you want
  • Shear windows and stack them like books or open them like (literal) windows
  • Shrink windows to icons (ala Enlightenment)
  • Toggle a level of transparency
  • Copyright information hidden on the back of a document
  • Back-to-front printing

The ideas are limitless.

I know this has been attempted one other time - that time being with Metisse. But that attempt was too grandiose and complex, so it was scrapped in favor of Compiz. But if this idea can retain a level of simplicity, while maintaining the ability to work with Compiz, it could revolutionize the desktop.

As a writer, I can instantly see the benefit of having access to the back side of a window. As a consultant and administrator, the perks of this idea are also readily apparent (imagine slapping a network diagram on the back of the Network Manager tool on your Ubuntu server...you would never forget where that diagram is).

Of course what this very idea does (besides open up huge possibilities) is showcase the flexibility and open nature of the Linux operating system. It is this sort of thinking that keeps Linux at the forefront of innovation. And these types of ideas will continue to make the Linux desktop the single most impressive desktop available.

I believe end users are ready for such levels of complexity on the desktops. And if they aren't, I know that users like us are. The standard desktop metaphor has grown tired. It's time for something new and something REALLY impressive. I believe Esfera is just that. I hope the Ayatana development group follows through with Pablos' idea. It will add yet another layer of WOW to a desktop already impressing users across the globe.

About

Jack Wallen is an award-winning writer for TechRepublic and Linux.com. He’s an avid promoter of open source and the voice of The Android Expert. For more news about Jack Wallen, visit his website getjackd.net.

42 comments
Hernab
Hernab

Imagine http://www.financeandmarkets.net/ you have a new control on your desktop windows. This control accepts gestures that can do various tasks. One of those tasks would be to flip the window over giving you a new space to use. This space could be used for

The 'G-Man.'
The 'G-Man.'

Don't switch users however. Actual delivery does.

hariks0
hariks0

Almost all these are possible with compiz already. But I guess this is still relevant as Gnome 3 will not support compiz and will be the default of Ubuntu 10.10.

YankDownUnder
YankDownUnder

Um...have I walked into the Twilight Zone here? "There is nothing new under the sun." With nearly ANY window manager (Gnome, KDE, XFce, etc etc etc) and Compiz, this has been done already - time and time again. Unfortunately, no one has completely incorporated all of this "pre-configured" in a "first install" basis. This is just a melting pot of already existing feature. I truly do send kudos to Ubuntu/Canonical for this concerted and unified effort - however I do have to remain cool about it all as I've been doing it since the advent of Compiz, so there is really nothing special here. On this note, though, I have to state that this is one of the draws towards GNU/linux that I've been able to use for my clients in migrating them away from Microsoft-based environments, or from "dogmatic dependence" - especially those powerusers that CAN make use of multiple desktops and window enhancements.

Tidux
Tidux

Fluxbox and other window managers have had the ability to tab windows together for years, and (Fluxbox, at least) to script things like "if you open this program, open these other programs in a tabbed window". Tadah! All of that "grouping" junk, without the extra shiny doodads. I agree that the desktop metaphor is getting old, but I think that the real UI revolution is going to be dropping the mouse in favor of touchscreens, and creating new window management and application GUI styles within that paradigm. RIP, rodent.

saundersp
saundersp

This is not a fascinating concept - leave the buttons where they are !!! Your supposed to be trying to win people over to Linux NOT push them towards Microsoft. This proves more and more that Ubuntu is the best friend of Microsoft, by pushing the fans away. Like the grub screen - why are there so many options on a duel boot system, what is wrong with Linux and MS option only [or what ever the OS's are].

saundersp
saundersp

Why change the buttons - I moved to 10.4 and the buttons are changed, so its back to Windows - they do not confuse users as much.

aroc
aroc

Sounds similar to Sun's LookingGlass windowing experiment - kinda cool as I remember from a demo. Sorry I missed the earlier post re LookingGlass, but just want to highlight with a subject line referernce.

jeromemueller
jeromemueller

Instead of building more complexity in to a twenty some year old windows and pointer UI why not spend the energy building a future UI. Gestures is a good analogy, but how about gestures of the hand, face, eyes, etc. How about mind spaces we can switch between as we switch threads of thought now. Spend the energy making the UI simpler, not more complex.

andrewgauger
andrewgauger

Accessing the back of the window should be made easier: just click and hold for 1 sec should bring the back of the window in focus. The "crescent right to left" gesture would never be "accidentally" discovered (as is the way I end up finding most of my UI).

madmalc567
madmalc567

Next time you want a ham sandwich why not try putting a slice of bread between two slices of ham - the possibilities are endless

Riofl
Riofl

Would be nice. Sun has had something similar in java for a few years now with their Looking Glass 3D desktop concept. It is quite nice. Last I looked at it was a few years ago. At that time they had reversing windows with notes, and other features, transparency, book shelf stacking etc... so although not completely new it would be nice to see something like this done in c properly. It was too "alpha" at the stage I checked it out to be able to run it seriously which was a shame. If it was more mature I would have switched to it immediately.

Murfski-19971052791951115876031193613182
Murfski-19971052791951115876031193613182

I'm not too fond of the idea of using mouse gestures, because I use a trackball, which is a bit less amenable to gesturing. If you're going to use gestures, though, why not carry the idea out to its logical end? Using a touch screen, thereby enabling one to use his finger(s), would make it very intuitive. That might make me give up my trackball.

brian
brian

I just want a full-gui linux server....how long do I have to wait????

mariuslv
mariuslv

If you want to manipulate windows with mouse gestures, why not make it simple and just use the title bar itself, rather than introducing a new button? Some use of gestures, such as dragging to the side to dock a window on the edge of the screen, or dragging up to maximize, have already been implemented quite well both in KDE and in Windows 7, but without introducing a special "gesture button". I'm not dismissing the idea of using gestures in new and creative ways, in fact I think this sounds really cool. None the less I remain unconvinced that the ideas in the Esfera concept necessarily depends on inventing a new button.

carlsf
carlsf

All I got was garbage ......

sonicsteve
sonicsteve

Or maybe even during setup of ubuntu much like IE faces during windows setup on EU versions. Choices could be, Would you like to try the new Esfera window controls? Then give a description with the choices; 1. No thank you keep my Ubuntu the same as usual 2. Yes but give me the normal window controls also. 3. Yes, use efera as the only window control. Though I don't see the Ubuntu development team making a drastic choice like this as the default window manager. Nor do I think they would want to force users to make a choice. Which means we will need to install it after setup is done. What I would like to see is a choice during setup of what apps we would like ubuntu to install with. Everything from advanced things like LAMP servers to the mundane like installing thunderbird, or Gnome Color chooser. Since we now have an app store like Android, and Apple, we should use it during the setup for users who have Internet access during the install.

samnets
samnets

Potential for enhanced e-book application -- with association capabilities. Could be enhanced by "merger" with Semantik for visualizing networks of concepts, designs, social and geographical relations [yes, Metadata VISUALIZED]. Further value in providing ability to show DYNAMICS [both by animation and simulation/input data presentation]. GREAT IDEA for a new UI "frame" -- could finally move us beyond icons and simple stacking to INTEGRATION of data and concept abstraction with visual tools. Anyone who works on complex problems [especially in teams] has either "invented" pieces of this or wished he/she had. Sam Leven

IndianArt
IndianArt

We need a pioneering spirit to innovate.

aroc
aroc

Odd how I see folks at work and in cafes with the latest note/net book with a touchpad... and a mouse plugged in. I have worked with several touch screens on notebooks/tablets and standalone monitors, as well, as a dozen or more PDA/HPC/phone devices. I have found that the bigger the screen, the more fatiguing it is to use for any length of time. There are issues with support for the scribing hand (esp for us lefties), and for shoulders (and even backs) with all that reaching if we have any joint problems to begin with. Our physical "configuration" and limitations dictate certain "classic" interface solutions such as keyboards and mice (although I prefer a trackball if I can find one with just the right layout, and durability, and/or a Trackpoint instead of touchpad when I am traveling mouseless). To each their own.

filker0
filker0

The location of the "buttons" on the title bar is configurable; if you don't like where Ubuntu has them, change the default.

ron.carlton
ron.carlton

Windows doesn't change and confuse users? Have you seen the "Ribbon" in MS Office 2007? It drives me nutz and there is no way to turn it off and have ordinary menus that I have been using for a couple of decades.

Slayer_
Slayer_

Why change something that works good? There is a reason that the brake pedal is in the middle and the gas pedal is on the right. It worked well once and we never changed it. Changing it now would be disastrous.

rwparks.it
rwparks.it

Innovative controls can be seen where developed for those with disabilities. i.e. the Windows 'sticky key' setting. Applying new psyche to these actions such as holding down the mouse button for over 1 second opens new UI abilities. But be careful that it doesn't counter existing applications or hinder those unable to use it (i.e. challenged with fine hand motor control). Holding down the mouse button spuriously or for a long time should give consistent results. Be deliberately considerate.

terry
terry

Given a decent GUI on a server would have me converted yesterday.

PhotoIT
PhotoIT

New interface designers are looking at creating interfaces that work the way we work. If you want to turn a sheet of paper on the desk, you grab a corner and flip the page. This gesture would emulate that to a degree. Much like the touch screen interfaces in our lives today, gestures are becoming less of an exception and more the norm. While this would be an example of a mouse gesture, and not a touch screen, we need to begin somewhere. A foundation must be created to build improvements upon. The blue button is merely an IDEA. Click here, make a movement and get the desired result. It is a concept in draft form. Perhaps it may end up as a simple little circle in the vein of the existing navigation design. Maybe a triangle. In the end what this little ball represents in positive change is far larger than just the little ball. If the final result IS a blue ball, then it's still better to have SOME improvements, then no improvements at all. In invite us all to look at the possibilities rather than looking at something minimal in the draft that does not meet our expectations or approval.

terry
terry

Rotating 3D objects makes a lot of sense but why rotate 2D objects (because you can)? Semi transparency...well??? But flipping the window and making notes, that is great. The audience agreed. As a developer two, or even, three screens is very useful. Maintaining those EASILY on one physical monitor would be great. A desktop free of clutter is good.

filker0
filker0

The "back" of the window concept is nice; for one, I'd love to have a layered HTML editor that had a WYSiWYG view that I could look behind at the underlying HTML. I view it more as layers, so there might also be a DOM layer. Transparency on the presentation view to see the mark-up and/or document structure is just one area that his might be useful. But this is an application model, not a desktop model. If the desktop toolset provides an easy way for the compositing of several views into a single display window, and the default (absent the desktop extensions) simply creates a window per layer without the application having to support a separate model for each, then this may be a good thing. Mouse gestures are difficult for some people to master; I have trouble with them mostly because I'm a motor dyslexic. There should always be an alternative way of triggering any action that you assign a gesture to. When I was working on a new GUI design in the days before the X project at MIT, or even the original Macintosh was released, we had a rather different set of operations; you could resize a window so that it showed more of the document, resize a window so that it showed the same portion of the document in a larger size, and resize the contents of a window to zoom in or out. We had scrollers and pagers; the scroller was like the scroll bar today, the pager moved between pages within the document. All operations that could be done with the mouse (or any other pointing device) could be done from the keyboard (though not as easily) by using chords of the arrow keys and a few modifiers and other keys; it was a distinct mode. We supported multiple locator devices, either joined or separated; you could use a mouse or trackball for UI operations and a tablet for drawing and they did not affect each other's pointer location. These are all features I'd like to see in a desktop UI. I believe that this might require extensions to the X windows system messages, it should not be impossible.

aroc
aroc

I agree wholeheartedly about The Ribbon (of insanity), however I did find out about the Quick Access menu/bar, that lets the user customize their own 1-bar set of favored controls. It can take some time to learn which of the many controls are the ones you want handy all the time, but it is worth the time and effort to tame that Ribbon-beast. I would hope any new windowing innovation will follow M$'s "lead", and always provide options to customize the interface (not really new with them, but just a little prod there ;-). No one size fits all.

Slayer_
Slayer_

It's been in the same place for both mac and windows for the last 20 years. People have come to expect it.

davidsaintamour
davidsaintamour

You're kidding, right? We should never change any technology or innovate because Windows never changes? This comment makes no sense.

rwparks.it
rwparks.it

The standard gas pedal has worked well for automobiles. But what if you can't use your feet? I knew a parapalegic who designed and built a joystick-controlled van back in the mid-1970's. Today, can have hand controls hooked to the gas pedal so that one can drive the car both ways. That's not disaster, that's creative evolution. Thus, the standard brings a simple consistent approach, and exceptional needs drive innovation and progress. Keep developing new ideas for UI.

Slayer_
Slayer_

Why are we not in caves anymore? Because of women. Men are always satisfied with good enough, women are not. Ever modern advancement has been for women. So why are we not advancing the OS as well? Why bother, its good enough, and there is almost no women to tell us otherwise. The small advancements that do happen, all have female influences. And that's my nugget of BS for today

PhotoIT
PhotoIT

If neither you or your wife change in life, much less improve yourselves, life would get very boring and you would rapidly fall behind society. The world will continue to move forward through change. That is just the way it is. If the world is moving forward and you are not, then you are essentially moving backwards. The same is for all technology. Move forward or hit a premature EOL. If we as humans did not embrace change, we would still be living in caves and grass huts. The apple OS has successfully implemented the controls on the left of the screen for years. 10.4 has already moved those controls, except for a minimise menu button. Why not take advantage of the unused navigation area to make usable and useful improvements. Just because you or I many not embrace the suggested change, does not equate to most people not embracing it or having a need for it. Heck, applications have a help menu. One that a large number of users never seem to see or use (based on the number of help calls I receive that are easily resolved via the help menu). Shall we remove the help from the menu because countless thousands don't use it? By your incessant need to pushg your limited vision for the future, I wonder if your comments are made out of intelligent analysis of information or simply out of fear of something new that is born of a process you were not part of...I'm sure this thread just can't wait to hear what you post next.

Slayer_
Slayer_

Perhaps you prefer to venture around, pick up a few STD's and say its a fact of life, but I prefer to stay home with a single woman who ages gracefully.

davidsaintamour
davidsaintamour

Everything changes, including by the way, Windows. Just because they moved a button, I'm not going to run crying back to M$. It's your life, do what you want, but the idea that a UI change will cause lots of people to go back to Windows makes no sense to me.

Slayer_
Slayer_

Why move a button that has been on the right side since the inception of the modern GUI, to the left side, to make way for a button that many won't even use? What was wrong with putting this new button on the left side, where it won't bother anyone? Do these obvious solutions completely allude people?

rwparks.it
rwparks.it

Good word choice - 'Extend'. Extending capability so others may use it better. Needs drives improvement, and broadens the ability for more to use. Another example - power saws for left- and right- handed people (Lefties - have you ever had trouble using a standard circular saw?). Traditional power saws are made with a right-handed perspective. Adding a handle on the front of the saw makes it a 100 times easier to use from a left-handed profile, while not taking away from any standard (righty) use. The addition extends usefulness to more people, leaving the choice to them. For GUI --- simply extend abilities allowing customization to better serve one's individual needs.

Slayer_
Slayer_

So your saying that because the odd person needs custom controls, we should all have them?