Re: I must be doing something(s) wrong

[David Ross <david netwebconsultants com>]

Thanks for your prompt reply.

On Wed, 2004-12-01 at 13:35, Alexander wrote:
> David Ross schrieb:
> >       * zoom - 100% magnification is about half of actual size.
> 
> Actual size of what? Having 1600x1400 on a 15" screen causes small objects, 
> too :-)

That's an important point. One that I overlooked even though its been
staring me in the face :-). I think tuning display setting for graphics
is very important. I don't recall seeing any documents or references
about it, but I could have overlook them too (see above and below :-)).

I like a lot of "bandwidth" whether its on a monitor or diagram (see
below). The aspect ration for monitor scan area and desktop need to be
congruent to avoid distorting the image. All the mucking around I do
with the display adjustments could be part of the problem.

> >       * grid - Have not found a combination of settings that are easy to
> >         work with for object sizing and layout. Stable granularity at
> >         different zoom levels.
> Again, settings to do what? Snap-to-grid-granularity? Try 
> Object->Align[vertical/horizontal]

Point well taken. I have had more difficult with setting and maintaining
object size.

> >       * Connection points - are not visible most of the time, have not
> >         been able to easily connect, add or delete. visibility is part
> >         of the problem, but positioning the pointer directly on then
> >         hasn't worked.
> Connection points are little crosses on the object's borderline. Drop the 
> arrow/line on such a cross->connected. To add new connection points (to 
> some figures)->object's context menu->"add connection point"

The connection points aren't visible most of the time. More so at higher
magnifications. I have had difficulty using the context menu not being
able to point the point (hmm... to the point, but probably confusing).
There seems to be no effect if the pointer (ad infinitum, ad nauseum
...:-)) is position exactly on the blue "x" when visible or on the exact
spot according to the rulers when its not. The menu command seems to
work when the pointer is about a centimenter away.

> >       * garbled views - view images seem unstable. Objects change for no
> >         apparent reason (size, text). Refresh usually clears it up. This
> >         can indicate that a crash is immenient.
> Sometimes I have thicker outlines/borders but I also never used images and 
> never hat changes which I didn't cause. On Windows or RedHat?

Not certain, but my impression is Windows/Redhat is similar if not the
same.

> > 
> > So far I have mainly worked with shape sets included in the
> > distribution. The collection has a lot of good stuff. The models I am
> > trying to develop are large and complex. Graphical layout, format, etc.
> > is essential for an effective presentation. The inability to adjust
> > format options is frustrating.
> Less is more. I never presented a diagram containing more than 10 objects. 
> Depends on personal preferences, I'm quite normal and can keep only 7 facts 
> in mind :-) What options do you miss, font style/size/type, thickness of 
> lines? Use the object's context menus, configuration files (mentioned in 
> online help).
> > 
That's a good point. I will have to give it more consideration. Managing
complexity is critical in developing architectures and designing systms.
I address this through standard sets simple patterns. They documented
separately to server as a modeling palette or vocabulary. Layout and
other visual cues are effect ways of layering details in a visual space.
I prefer this to flipping pages, but my audience may not.

I would prefer making font style changes globally or on sets of objects,
possible through a style. I guess that would apply accross the board to
all graphical elements. As long as you asked :-), styles could be
attached to layers. Some would be fixed as part of the symbol
definition. The remaining ones could exist in layers which could be used
to build views.

Thanks again.


________________________________________________________________________
david

email:    david netwebconsultants com
phone:    (631) 224-1244