Re: How many use the diagram tree?

["James K. Lowden" <jklowden schemamania org>]

On Mon, 11 Aug 2003, Alan Horkan <horkana maths tcd ie> wrote:
> On Sun, 10 Aug 2003, James K. Lowden wrote:
>
> It is not just Visio it is just about every vector graphics program I
> have seen.  I guess handles must have seemed simpler to program at some
> point.  I guess the problem is the difference between wanting to
> simpley move or change the shape, the handles make it clear that you
> want to resize rather just select and move.
> Certainly being able to join one line to anywhere on another line
> or circumferance would be nice

It's true I was talking about the connection points and I'm sorry for the
confusion.  

WRT handles per se, however, I note that no window manager uses handles to
facilitate resizing, and I really think resizing a shape on a diagram is
no different from resizing a window on a desktop.  In the WMs I know,
except TWM, when I put my cursor along the boundary, it changes shape to
show I'm about to resize the window.  That's feedback enough, and easier
to use than handles.  

> > To clarify what I meant by "tabbed diagrams".  I've found that even
> > for very complex diagrams, only a few named subsets were worth
> > maintaining. Each one is a kind of sub-diagram, really another
> > perspective on some of the objects.  Suppose each named subset would
> > appear as a tab in the diagram's window, similar to the way tabs work
> > in Mozilla.  I suggest this
>
> Mozilla tabs are showing multiple seperate documents, I was trying to
> make the minor distinction of a document with multiple sheets/pages
> making up a workbook each of which would be displayed in a seperate Tab,
> like in Excel, PowerPoint, Visio, OpenOffice Draw etc.

So we have two ideas: (a) "views" of subsets of the diagram, and (b)
"related" (probably linked) diagrams.  

I would say a drawing program is different from something that deals with
text or pixels.  The notion of "a subset of objects" doesn't apply to
browsers or spreadsheets or text editors.  I can only tell you that as the
object count approaches 100, it becomes more and more useful/necessary to
partition it, and the boundaries don't normally fall at the page breaks. 
Tabs seem to me like a natural way to render named subsets.  Separate
windows managing a single document strikes me a hard to use and to
implement.  

> The real benifit of Tabs for Mozilla was speed

Perhaps.  I was only trying to illustrate what I meant by tabs.  

> It might also be possible to achieve what you want by having a object
> say a Rectangle that you can click on and when you do so you are zoomed
> in and get to see  the parts that make it up.  This is more a case of
> better handling hiding details at distant Zoom levels.  I am not sure I
> can explain this particularly well.

Well, I think we've bandied this about before.  I hope one day we have an
object that acts as a link to other diagrams, to support "drill-down" for
such things as data flow diagrams.  It could also act as an "edge
connector" to other diagrams managing related domains.  

But I think Zoom is an orthogonal notion.  The idea of revealing objects
within objects by zooming in is appealing in a Star Trek kind of way, but
I think it would be unwieldy in practice.  Better to restrict Zoom to
magnification for the user's convenience.  

> sorry if i sounded dismissive, I am a bit grumpy at the moment, not
> getting much of anything done.

FWIW, I didn't think you sounded dismissive.  If I may say so, you're
certainly capable of sounding grumpy, but I find that's mostly a matter of
tone; the text of your comments is usually thoughtful and engaged.  

Regards, 

--jkl