I keep my most updated list of publications at:
http://www.interaction-design.org/references/authors/miguel_a__nacenta.html
You can download most of my papers from:
http://hci.usask.ca/people/viewpublications.php?id=16
and
http://sachi.cs.st-andrews.ac.uk/people/faculty/miguel-nacenta/
Dr. Nacenta,
I read the recent piece describing your work in New Scientist and Googled Fat Fonts hoping for a download. Found your web sites. Scalable vector graphics are fine for design, but do not a usable font make, at least not in a transparent manner for the Mac OS. Maybe another day.
I’m a retired professor of audiology with continuing interests in perception, technology and what I term “communication engineering.” In the old days, I messed with perception of synthetic speech. My last experimental work was a study of the effect of coded electro-tactile stimulation of the tongue as an ancillary cue to “lip-reading.”
I applaud your Fat Fonts idea as a continuation of the work of Edward Tufte (please, please cite his work). Fat Fonts issues seem to include: no zero (OMG, Roman numerals redux), nesting limitations, and unclear compatibility with metric anything (i.e., base-10 progressions, exponential notation). Your AVI’12 paper addresses some of these in creative ways. But I write for another reason.
You elected to scale digits on the basis of area and a linear proportionality linking cardinal number size and ink area. Your personal web page also notes an interest in psychophysics, which I heartily commend. From that statement, I assume you are familiar with the classic laws of Weber (1834), Fechner (1860), and Stevens (1957, 1975). The later, a power law, suggests that for square shapes, perceived size grows with area raised to the 0.7 power, times a constant. Work by others questions the underlying veridicality of Stevens’ conclusions about prothetic perceptual (“how much”) continua, but the power law remains a useful description of the relation between stimulus and perception for a host of applications. Fechner’s law has similar pragmatic merit (e.g., in photography the log progression of exposure durations and aperture).
To my eye, your number 7 looks too big re digits 1-6, and too small re digits 8 and 9. Here’s an experiment to consider: Compare your existing font to a new font in which the proportionality ratio of ink (area) is not 2, but the power 0.7. The dependent variable could be the time (in seconds) required to translate a complex image into a string of conventional digits (efficiency). Another DV could be the percentage of digits correctly translated (effectiveness). Independently run 30 subjects in each of these two conditions, counterbalancing order. If you get a significant result, do another experiment, adding the independent variable of complexity (number of digits OR nature of the event depicted by digits). If that runs out to be a significant, go back to Stevens’ work on the area of squares and expand it to other shapes, including amorphous, irregular shapes.
Now you’re doing psychophysics. Next stop might be computer simulated whiskey tours.
=
Cheers
Mike Chial