19

UPDATE A new viewing procedure. Brief description and links


Here you have a new demo with the characteristic KDE flavor. Ironically, when executed under Linux it works ridiculously slow. If you can, try Windows; this way works fine.
http://www.mediafire.com/file/qownin2qj ... erDemo.jar


On the other hand, if you want to try a lighter demo, I recommend you to proove this one, where you can only see a graphic structure. However, it also presents an error -will be corrected in the future- when running under Linux.
http://www.mediafire.com/file/ytogjzyzm ... erDemo.jar

You'll have to use a mouse with a mousewheel.

I propose a new viewing procedure that enables the user to navigate the desktop easily. This procedure includes an autonomous movement of the cursor and allows a simple navigation with respect with the classical pan-zoom procedures.

You can find the documents you need at this website: http://buginabox.net/

A desktop separate of the screen frame. Theory.

1: Desktop enviroment:
Desktop enviroments offer a number of elements that combine to allow the user the management of multiple applications. This text referrs to three of these elements: work surface, window and pointer.
a) The first one, the work surface, is the stage on which the other elements are shown. This stage is set to the limits of the the screen, adapting its appearance to the screen resolution. Any element outside the screen frame will be at least invisible.
b) A window is an element capable of hosting an application. In general, the windows can be resized and dragged over the work surface, giving the user the ability to distribute them according to his convenience.
c) In order to facilitate their manipulation, the desktop environment offers a marker called pointer, which moves in the foreground obeying the wishes of the user. The pointer has the property on indicate a particular point in work surface.

2: Overlapping windows
To assist the work over a particular window, the environment gives the user the control of a signal called focus. When the user gives focus to a window, it overlaps all the others, turning the hosting application on receiver of its actions. Because the work surface has a limited space, it is common that the user is forced to deal with piles of windows that overlap in the order established by the possession of focus over time.
In order to simplify the extraction of a window was established a one-to-one correspondence between the set of windows and a set of symbols, thus giving focus to the window was reduced to the identification and handling of the symbol. Similarly other needs were solved, such as hide or close.

3: New paradigms
Having grown in complexity and enriched in various graphic aspects, the current desktop setting retains, however, the paradigms of traditional desktop environment. The issue to avoid windows overlapping is still a problem for the user, problem which has been moved to the application developers.
The desktop environment proposed in this paper aims to resolve this inconveniente review three traditional paradigms:
a) The work surface is no longer binded to be adjusted to the screen frame.
b) The resolution of the work surface is no longer dependent on the resolution of the screen in wich is displayed.
c) The screen frame becomes a viewer through which the work surface may be seen.

4: Features
The exposed desktop environment does not impose to traditional desktop environment, but rather complements it: it can be used in the orthodox way, thus ignoring the use of the rising paradigms. However, the use of these paradigms can be a source of emerging characteristics:
a) The distribution of the windows on the work surface drives the use of multiple sessions, each of them having a particular configuration for different activities or users.
b) A wider space allows the user to dispose the desktop elements in a meaningful spatial configuration.
c) A broader perspective facilitates the exchange of data between different applications or instances of the same application, allowing a more effective document management and replacement of operations such as "cut and paste" in front of other as' drag and drop .
d) Separation of the work surface with respect to the screen frame may represent an useful evolution in desktop environments.
d)...

Finally, I must say that, after having grown up this idea I came across a similar one: please follow this link: http://www.google.com/patents/about?id= ... ws+desktop

First: Your idea has something what is still possible, to have a virtual resolution what you can then zoom in/out and combined with 3D effects and acticities in KDE Workspace.

Secondly: "Modern operating systems usually offer a so called desktop enviroment,"

The OS does not offer any user interface. It only offers interface for the developer. The interface is just a API. The usually interface what user see, is stand alone software, ran by the OS. The interface can be as well the hardware. Like elevators buttons and the floor screen. The interface is as well the keyboard, mouse and the monitor itself. OS does not include user interfaces at all, only interfaces (API's) to user.

KDE SC is not different "OS" to normal person than GNOME is. Even than normal person would believe so.
The simple truth (what is tried to hide by some parties) is that the Linux kernel is the operating system. It is not a microkernel but a monolithic. Monolithic kernel is the original structure of the operating systems. After few decads, there came other structures as well like server-client (microkernel + servers/modules, what should not be mistaken to daemons) and layered architectures.

Once readed and understood your post, Fri13, I edited the text. So, where could be read:
"Modern operating systems usually offer a so called desktop enviroment, which has a number of elements that combine to allow the user the management of multiple applications"
now you simply read:
"Desktop enviroments offer a number of elements that combine to allow the user the management of multiple applications"

Thank you for your comment.

I've though about designing a window manager to do this - one desktop that tiles the windows around a virtually infinite plane. The view area will pan to show the entirety of windows as they gain focus. The desktop and panel would remain static (and move under/over the windows) as the view area pans. That coupled with KWin's native compositing/3d would be a great combination as it would enable zooming out to show all of the windows (already in a present/expose-like arrangement). Panning could be configured to happen whenever the mouse touches a screen edge or whenever a window is requested which is not entirely inside the current view area.

Hello airdrik: The idea of creating a separated desktop is not new; you see: two guys from IBM obtained a patent on the idea more than ten years ago. The issue is precisely the method of navigation. This is my original work. Having devised a procedure to rapidly navigate large surfaces, I thought it would be an interesting idea implementing it in a desktop environment. If you want know more about this procedure, you can take a look to this presentation:
http://www.slideboom.com/presentations/ ... ESENTATION
You can also see a slight example:
http://buginabox.net/web1/files/viewerDemo.jar
(it is a java executable)

Re-editing: I do not know anything about the structure of kde, but I think it would be more consistent decoupling desktop and screen, remaining elements (windows, icons,...) related to the "desktop surface".
About the task panel, I would personally remove it -occupies space-. (if we know where each window is, What would we need the buttons for?)

Text has been shortened in order to make the update possible.