I would like to propose a project for those who'd like to be involved
with the art project but who would prefer not to create artwork to
increase the usability of the projects:
(1) Create sheets of the currently-designed icons as-is and with various
filters applied them to synthesize how they might look from a distance
or by visually-impaired users.
(2) (Even better) Create a plug-in for the GIMP or Inkscape (similar to
the icon view too) so as the artists work on the icons they can preview
how their icons would look with the same filters applied to them within
their graphics creation tool.
The most recent issue of Boxes and Arrows has a very well-written
article by Matt Queen about how to analyze and improve icons for
discernibility.
The two main vectors by which we perceive differences in visual
information are:
- P pathway - color & detailed shape (parvocellular)
- M pathway - gross shape, luminance, motion (magnocellular)
The more differences between icons in *both* of these aspects, the more
discernable the icon will be. If icons all share the same color scheme
and shape, for example, you're losing a lot of bandwidth for making the
icons discernable.
To make icons more discernible via the M pathway:
- most simple method: 'squint'
- better method: spatial frequency filtering. The author suggests using
the R statistical package [1] (yes, it works on Linux as well as OS X
and Windows.) He outlines the methodology quite well in the article. You
can see an example of the output here:
http://www.boxesandarrows.com/files/banda/icon_analysis/Icon_Analysis-02b-4.jpg
Matt's suggestion for how to apply this for the creation of better icons:
"I just mentioned in an email to Dustin Hamilton that he hit the nail on
the head, distance filtering of icons is an evaluative technique—not a
design technique. My knee-jerk reaction of how to coerce this method
into a design tool is to create a plug-in for Photoshop or GIMP
(preferably) that would allow an auto-updated window of the design space
filtered at set distances. That way all design decisions (small and
large) could be made in the context of the filtered frequency views. I
imagine there is someone out there with enough talent and time to create
such a tool. It seems like it would be useful."
Thus my suggestion #2 above. Queen points out a paper with the equations
upon which such a plug-in could be based on:
http://retina.anatomy.upenn.edu/~bart/scriptie.html
(see the paragraph that starts with "Later theories of perception yield...")
If anyone is interested in this project please reply so we can organize.
Also, for the record & to reduce potential confusion, I would like to
make clear a non-goal of the Echo icon theme:
- The goal of the Echo icon set is *not* to serve as a high-contrast and
simple-shape icon set for blind and low-vision users. (The goal is to
serve as a more general-purpose icon set.)
Thanks,
~m
[1] http://www.r-project.org/
_______________________________________________
@xxxxxxxxxx
http://www.redhat.com/mailman/listinfo/