RE: CFC - Target Size (2 options)

I could live with the second option, and on Steve’s point:

> The first question is simply why would a person with a disability have more success hitting a target of 26x26 with 8 spacing (pass) than hitting a 43x43 target with 7 spacing (fail)?

If you had a list of links which were plenty wide enough, but 28px high with 8px spacing, you have 36px between the middle of each target.
[Steve] Right, and that would pass.  And my example of a 43px target with 7px spacing has target centroids 50px apart, which fails.  How does that make any sense?

Touch devices have algorithms that help you select things, so spacing acts as a increase in the size of the touch target.
[Steve] Roughly speaking, those algorithms are just based on whether or not your touch patch centroid lies deep enough in the probability curve centered on the target’s centroid.  In other words, it’s more accurate to say that targets in close proximity get in the way of each other’s probability curves, and not that spacing makes anything bigger.

Having said that, I don’t think we’ve bottomed out the effects there, or come up with defendable values yet, or worked out if that really helps the author achieve certain interfaces, which is why I can’t +1 it.
[Steve] Agreed, but I’m never going to get past faulty math.  If we want to bring in a spacing option, then the SC needs to say something to the effect of: “Targets are at least X by X CSS pixels with at least Y CSS pixels from the center of the target to the center of any adjacent target, …”.

-Alastair