Reducing the width of a sequence diagram

Question

Does anyone know how to reduce the width of a reverse-engineered sequence diagram?
Studio sequence diagrams are very wide in relation to their height. This is because horizontal messages are separated by only small amounts, but the vertical lifelines are separated by large amounts, especially if the objects have long names and class names.
This feature causes problems when trying to export diagrams to A4 landscape pages.
In other UML editors, I could overcome this problem by arranging the objects as in two rows of "bricks" along the top of the page. The first object would be in a "normal" position. The second object was positioned high enough to partially overlap the first object. The third object was placed in the "normal" position, but partially below the second object. And so on.
This allowed the width of a sequence diagram to be reduced to almost half of its original width.
This is not possible in Studio, because:
1) The objects cannot be moved vertically.
2) Even if they could be moved vertically, they will not allow themselves to overlap another object. This is because the bounding box of the objects extends right down to the foot of the lifeline. (I don't know if "bounding box" is the correct term. I mean the box with the sizing handles.) When one object is dragged to the left so that it overlaps another object, any objects to the left of the dragged object immediately jumpfurther to the left.
Another possibility to reduce the width would be to allow the object name and class name to be on two lines. Such as:
objectName:
ClassName
This happens by default in some UML editors.

Answer 1

petersl wrote:
Found a better way to duplicate the sequence diagram
as a workaround for the Save As function.Great suggestion! Thanks!
Because of the estimated large amount of sequence diagrams that I need to do, your tip will save me hours of work. Even if "Save As" is implemented, it will not be much faster than your method.
Here is an interesting feature of your suggestion.
All my experimentation is done using the BankAccount tutorial. I have several "copies" of the same operation. Each copy was produced using "Reverse Engineer Operation" on the class operation in the dependency diagram. I then modified each "copy" beyond all recognition.
If I now invoke "Create Diagram from Selected Elements" on the sequence diagram interaction node for each copy, each new diagram is the same as the original reverse-engineered diagram :)
BTW, here is my non-techie solution for copying. No LOL please.
In the project file location, there are two files for each sequence diagram. For example:
mainSD_1168604900517.etld
mainSD_1168604900517.etlp
These files can be copied. However, they must be copied to file names that are registered by studio.
1) Copy the two files, using arbitrary file names.
1) Reverse engineer a new sequence diagram simply to get two more file names.
2) Rename the two copies to these new file names :)
3) In the project tree in studio, this would produce duplicate diagram names. So in the ETLP file rename the diagram using the <diagramInfo> tag.
You might wonder why I go to so much trouble to produce a copy, rather than simply reverse-engineer as many copies as I need. The answer is that on the first sequence diagram, I would like to make changes that persist in every other copy of that diagram, even after messages and lifelines are removed to split the diagram.
Compared with my method, your method will allow me to take a couple of months off, and still produce the same amount of work :)