Always a good subject to bring up every once in awhile, I have a
question for everyone AND a thought or two on something brought to my
attention by a Delphi Programmer on one of the 3rd party newsgroups.
First the question: Let's say that we have a TAction setup named,
"StyleAct" which will when executed set an integer var to one of four
possible values. By default the value is -1. Then when executed the first
time it sets the value to 0, on the second execution the value is set to 1,
and finally on the third execution the value gets set to 2. If executed
again the value goes back to -1 and starts all over again.
This value is used to set the visual style of a form that has two
panes divided by a common splitter. Value -1 sets the splitter to a percent
of 100 so that only the left pane is showing. Value 0 sets the splitter
percentage to 0 so that only the right pane is showing. Value 1 sets the
splitter to 50 percent so that both panes are showing, in a horizontal
configuration, and Value 2 again sets the splitter to 50% but in a vertical
configuration.
Of course some additional visual changes are also made for each of
these values, such as toggling transparency on/off and locking/unlocking the
splitter depending upon whether you want the user to be able to make manual
use of it such as would be the case for values 1 and 2, but we can let these
minor attributes alone for now.
Now my question has to do with the best way of providing a means by
which the user can select and set the value he or she currently wants via
the execution of this TAction. And in this particular case I chose a single
button on a toolbar which allows the user to click thru the four settings in
turn and then back to the first again. And the visual indication of this is
provided by 4 different glyphs, or as is the case here, by altering the
TAction's ImageIndex at each click of the button. But should the glyph seen
at anytime reflect the current style that the interface is set to OR should
it reflect what the style will be IF the user clicks it? I've seen programs
that use both methods, and even a few poorly designed ones that use both in
the same application! <g> But I'd like to hear what others think about this
and what they usually do or would do in this same kind of situation. I'd
appreciate any comments on this at all!
Now for the design discussion: This concerns something that I have
been giving a lot of thought to recently, and I'll explain why in just a
moment. But was brought to the forefront of consideration when another
Delphi Programmer writing on a different venue brought up his own concerns
on the same subject. Like me, he'd been trying to better understand the way
people use their computers, most especially as is concerned with the area of
visually-guided thought and interaction.
The rectangular mainstream of visual presentation was quite
naturally dictated by the same rectangular shape of the computer monitor,
and it's very doubtful that this will change substantially until they are no
longer in use...perhaps if and when free-form, 3-Dimensional, focused
projections are commonplace this will change, but for at least the
foreseeable future and our lifetimes, the ONLY changes we can expect will be
in areas of quality, portability, and style.
Now some might argue that the more common use of multiple monitors
can provide a great and positive quality that we as programmers can use in
the design of our applications to better the way in which we evaluate and
interact with the information being presented , and that by means of
stacking and/or row banking we can allow for a lot more leeway in the design
of visual presentations then they yet have. However the ONLY way that can
be true is if the boundary edges of aligned monitors are first made visually
inconsequential! Then the total working area and it's overall width and
height could make it possible to substantially alter
presentation...especially when that presentation may call for more than one
form in view! But as it is now, the apparent edges of tangent monitors
causes such distraction to the eye that any value gained is just as
immediately lost! And that doesn't even take into account the requirements
of keeping each monitor in "phase" with each other so that slight
differences in contrast, acuity, and image centering do not further the
visual disjunction that can so easily occur!
Taking these things into account it's easy to see that under ANY
circumstances one singular monitor is by far much better than any multiple
setup possible. And if one agrees with this, then it's also apparent that
what we have to work with are but two attributes concerning visual
presentation...and they are, "aspect" and "resolution"!
Tomorrow I'm going to continue this diatribe with a more detailed
analysis of both these attributes, and how both or either can effect the
visual empathy we all hope to provide our users. In the meantime I invite
any and all comments you might have to make on this subject, especially if
you have given honest consideration to them and this consideration has
changed in even the most minute way how you design and/or scale your own
programs!
from Robert Meek dba Tangentals Design
__________________________________________________
Delphi-Talk mailing list -> [email protected]
http://www.elists.org/mailman/listinfo/delphi-talk
