[comp.edu] Why Icons?

kent@lloyd.camex.uucp (Kent Borg) (08/26/89)

While following this discussion, I have started thinking about the
value of visual icons.  I don't think it is in their universality, I
think it is in their simplicity--which certainly helps their
universality.

I see icons as attempts at very simple ideographic language.  

I think icons get their power from the fact that our eyes and brains
are very good at quickly and effortlessly recognizing familiar images.
Letters themselves are simple too, and can be recognized quickly, but
*text* requires an enormous extra decoding step and much extra effort.
Icons give us speed and ease, but at the expense of generality.  To
extend icons for more expressive power, they start to pick up grammar
and grow in numbers.  Eventually you might have something with a
structure like written Chinese.

To see when icons are well applied and when text is better, looking to
the real world can help.  As I sit at my desk I can absently mindedly
reach for my tea or my stapler.  I can find one of them by looking in
the physical place where I expect to find it, and by what it looks
like.  Now I imagine removing those clues and needing to find the name
"Tea" in an alphabetical list--or was it called "Mug".  I would rather
know that it is white, plastic, and hot at this moment.  I reach for
it often enough to make that worth learning.

To contrast, the phone number of Ethel Brewhopper is not something I
use often enough to make it worth learning, or learning an icon for.
Rather looking for her number in the phone book works very well.

Icons work well when they are used for a small number of frequently
used items.  Then the reptilian portions of our brain can be trusted
to recognize them, leaving the newer brain parts free for other tasks.
Text works well for things which need its generality and flexibility.


On map reading:

In article <1359@cbnewsd.ATT.COM> tainter@cbnewsd.ATT.COM (johnathan.tainter,ih,) writes:
>...
>direction.  The rest is map related.  A person reading a map
>probably has it layed out relative to the ground so up on the map is
>forward, left is left, etc. (regardless of ESWN)
>...

It has seemed to me that there are two different ways people read
maps.  Some imagine the map expanded to match the world while others
shrink the world to match the map.  The `map expanders' are the ones
who insist upon turning the map the `right direction' to match the
world, while the `world shrinkers' are insist upon putting North to
the top and they will match the world to that in their heads.

I have always preferred the `world shrinking' approach.  I imagine
telling a little person on the map which way he should go, then I
follow those instructions myself.  When I try to pretend the map is as
big as the world, it looses its advantage over the world.

Kent Borg
kent@lloyd.uucp
or
...!husc6!lloyd!kent