I'm working on a WPF desktop application. A stakeholder requested window docking (like Visual Studio 2010) for managing multiple windows inside the application. This option makes sense for the design problem, but I am unsure as to the usability of that type of window management.
Can anyone provide strong data regarding the usability of such a window management system?
I think it really depends on what sort of content will be represented in each of the document windows. Docking works well for Visual Studio because most of the windows that are being docked are text editors, or text representations of items (tree view, list view, grid view, etc).
Our software currently uses a dock manager control, even though the majority of our windows are not document windows. It works, but it never feels like it is the optimal solution. The interface can become cluttered very quickly, especially if a user decides they need to show almost every window at the same time.
The primary issue is scaling the content of each window to size appropriately. Again, a text editor scales perfectly, adding scroll bars where required, however that doesn't work for a complicated form with a mixed layout of controls.
To attempt to fix the scaling issue we had, at one time, implemented a control on every window that would dynamically scale the size of all of the controls (fonts, buttons, everything) however this too was never quite right. Forms would be designed with a certain size in mind, and in reality it was never actually displayed at that size.
Persisting, and restoring layouts is another consideration. The dock manager we use is able to serialize the layout, although we still needed to handle when/where/how that is used.
Support is another consideration. Since the layout is infinitely customizable, it becomes harder for support to help users navigate. We end up relying on the "Window" menu for everything.
"Find the Error window... nevermind... Go to the window menu and select Errors."
Potentially presenting an alternative that provides enough flexibility would meet the stakeholder's request.
Window docking is never easy for users to understand or to use.
However if someone uses an application all day, then it lets them configure the application in the way they wish to use it.
Therefore clearly it must come down to your users and how much time they will be using the application. Don’t let a few “expert users” make it worse for the rest of your users that may outnumber them many fold.
I was wondering if there is any new research on this? It looks like Adobe XD was built without re-sizable and dockable widgets so I'm wondering if they learned some lessons?
I've been an Adobe Photoshop/Illustrator/Premiere user for a long time and despite having knowledge of the software with more features got added I got lost with docking widgets, having to reset them once in a while. That stopped with Adobe XD because they focused on fewer features, direct manipulation of content and contextual UI (you don't see typography options if you click on a button for example).
I tried before to find some articles or posts regarding this matter, unfortunately I didn't find any.
But I don't think that you need some statistics or surveys to know that they are very bad and unusable. I don't know anyone who like it, instead every one hate it and talks about its bad experience.
I see the docking system as the last resort if you didn't find any thing better than it for your application, and that depends solely on your application nature. For example, the best option for IDE applications (e.g. Visual Studio) is the docking system, and that is because the user (i.e. developer) spends allot of time on each tab and don't switch frequently between tabs.
