Re: Knob styles (was: AU host properties)
Re: Knob styles (was: AU host properties)
- Subject: Re: Knob styles (was: AU host properties)
- From: Kurt Bigler <email@hidden>
- Date: Wed, 21 May 2003 13:23:04 -0700
on 5/21/03 12:10 PM, Marc Poirier <email@hidden> wrote:
>
But the problems with software "knob" interfaces go deeper than the issue
>
of circular motions with mice. Even if you allow linear control, the more
>
ergononic way, they're still a software interface abomination because the
>
action does not correspond to the view.
Look deeper again. When you turn a round knob (particularly a small one),
your fingers are probably making linear motions. Try it. It is possible to
make "round" motions, but that is much more difficult, probably involving
the whole arm rather than a simple twirling between the fingers.
For me when I use thumb and forefinger on the right hand the fingers are
moving in opposing linear directions, on a diagonal axis a bit
counter-clockwise from vertical.
This suggests to me that the linear+knob combination is not as far off as it
may appear.
To make it truer to the physical world a possibility to consider might be
for the knob to respond to the mouse movement in an similar way that a knob
would respond to a single finger twirling it. (I know things similar to
this have been discussed and dismissed, but bear with me.) Basically you
have a linear motion mechanism whose directional sensitivity is set by the
position of the mouse relative to the knob. It is hard for me to imagine
how this would feel to use, and I suspect it would require an experiment to
find out, and there are probably several possible variations in tracking
that should be tried.
However one thing is clear: the novice user approaching the knob would
probably find the immediate response more intuitive, and would not be
shocked by how the knob responds. Meanwhile the experienced user would be
able to get the same linear response they are used to from the current
linear+knob implementations, just by clicking toward the edge of the knob
that gives them the desired direction of control.
It would probably be useful to have an option for the advanced user to
enable a click-position-independent linear control of the knob. Also,
possibly clicks well-inside the knob might enable this behavior, whereas
clicks near the edge or outsie the knob might enable the new-user-intuitive
behavior. After all no one interacts with the center of a real knob, only
the edges, so really this would yield no surprise. Or the center could
behave as an interior extension of the top edge, or of the right edge of the
knob, as a user-selectable option. Either approach would yield a wide range
of click positions that would yield the desired behavior for an experience
user, while most likely giving a no-surprises response for the novice.
Maybe the directional sensitivity should be set by the position of the
initial click relative to the knob center. Or maybe it should constantly
track the knob position. In any case I think it would be a mistake for the
sensitivity to track the distance from the center of the knob, because you
want a large click area within which the response is the same for the
advance user.
In any ase it occurs to me that providing user options that default not to
surprise the novice is a good way to go, as long as there is a way for the
advanced user to select a less-physically-true but more streamlined mode of
operation. Also, recognizing that the pure-linear implementation for a
round knob is really just a streamlining of an already-existing physical
modality may make it seem less objectionable to some.
To me, user interface abominations are the things you can't adjust to, not
the things you can adjust to in a few seconds. I thought the same thing
when I first used a linear knob, and had a really strong reaction. Within
60 seconds I liked it better than any alternative I could have imagined.
You just have to worry about the new user who _doesn't_ adjust to it for
some reason. The criterion should be whether it is genuinely unnatural for
the user of various experience levels, as opposed to intellectually
unnatural to the mentally stubborn. People with lots of computer experience
can sometimes quickly adjust to things that boggle them briefly, as long as
they don't get into an intellectual argument with the software. :)
After a couple years of OS X, I still spontaneously try to use command-N to
create a new folder in the finder, and it still doesn't work. THAT is an
abomination for the experienced MacOS user: something that is not
spontaneously adjusted to after a relatively short time. Of course, maybe I
am being stubborn.
-Kurt Bigler
_______________________________________________
coreaudio-api mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/coreaudio-api
Do not post admin requests to the list. They will be ignored.