unity-design team mailing list archive

[Ed Lin <edlin280@xxxxxxxxx>]

On Sun, May 22, 2011 at 6:02 AM, Ralph Green <sirable@xxxxxxxxx> wrote:
>  If this is some kind of principle that guides the decision to use
> the global menu, then it is time to change.  I am quite experienced
> with the mouse and how to use it.  I use as much acceleration as is
> reasonable.  It would not be an improvement to have the mouse go
> twitching away out of control, which is what happens if I speed it up
> more than it is now.  So, the distance is not irrelevant.  If Fitt's
> Law is considered important, how can the distance be irrelevant?  It
> is one of the variables that most drives the result.  If you ignore
> distance, you are not using Fitt's Law, but rather an arbitrary
> derivation of it.

I claim no expertise in Fitts's Law or mathematics but I'll try...

Because one part of the equation becomes infinity the distance can be
ignored (distance/∞). On very large screen it's likely faster to
access a target not only directly under the pointer but very close to
it than one at a screen edge, not only because of mouse (and
arm/wrist!) acceleration but because you first need to "find" your
target with your eyes which it's outside your field of vision. A
corner will still be faster to access because you don't have to
search, focus and correct at all. This however assumes that you really
want to activate your arm and quickly move the mouse. So besides
Fitts's Law ignoring field of vision it also ignores our human
laziness...

I'd also argue that while mathematically sound (at first glance)
literally putting in infinity doesn't give you a realistic result,
think divided by zero which can break an otherwise perfectly usable
formula. Put in a ridiculously large "D" and with infinitive "W" you'd
get exactly the same result (you'd calculate with zero) which doesn't
take in account b, the inherent speed of the pointer. Clearly the law
fails here. Just see the derivation: natural number =∞/2 ??  Someone
with a bit more math background, please chime in.

>  Concerning my query about Fitt's Law, I note this quote from wikipedia:
>  "It applies only to movement in a single dimension and not to
> movement in two dimensions (though it is successfully extended to two
> dimensions in the Accot-Zhai steering law);"

 Mouse movement is one-dimensional, approximately because you don't
move the pointer in a perfectly straight line. But this is taken into
account as noted.



On Sun, May 22, 2011 at 7:28 AM, Ralph Green <sirable@xxxxxxxxx> wrote:

>  I spend most of my time running
> virtual machines or vnc sessions to machines on my network.  The edge
> of their screens is somewhere in the middle of the screen on my host
> machine.  So, global menus may hurt me much more than a typical user.
>  I do this even on an isolated machine.  I like the ability to
> checkpoint the filesystem on the virtual machines and isolate any
> errors to a VM that I can replace or roll back.
>   I am guessing the average user does not do this as much.  I setup
> client machines this way fairly often, but I don't remember running
> across anyone else who did it on their own.

ssh and automation with scripts ;)
But really, this is nothing the designers of Unity should or could
take care of. I'd put that into the hands of the devs of your favorite
VM and remote desktop solutions.

You know you can also trap your mouse in a window or run another
system on a separate virtual desktop?