- From: Patrick H. Lauke <redux@splintered.co.uk>
- Date: Thu, 20 Jun 2019 18:28:14 +0100
- To: public-mobile-a11y-tf@w3.org
On 20/06/2019 17:24, Kim Patch wrote: > *MATF Minutes June 20, 2019 link: [...] > <kim> TOPIC concern about touch target testing > > Jake: Comments were pointing out that if you define layout via > percentages of flex not css pixels you might get different distances so > space will depend on viewport width etc. > ... so measurement would involve things like screenshot, programatic > workout of positions seems not feasible > ... inspector can in those cases not work out the distance > > Kim: So is this a roadblock - can we get around it? > > Jake: Cant see a solution apart form scoping to cases where offsets are > defined in picels > > Jon: Is this the same issue as with touch target size? > > Jake: No, that one is always measurable in pixels > > <jon_avila> I have to jump off. sorry > > Jake: Distance, if not defined in pixels, you can only guess - no > margins, paddings defined > > Kim: How often does this happen? > > Jake: The more these Techniques are used the more difficult it gets > ... If it is not 100% testable we don't even need to ask that question - > asking Wilco will probably have the same result > ... you can measure every pixel on screen but there are many variations > ... screenshot-based will be very awkward > > Detlev (describing way of measuring) > > Jake: When elements are close it could be 5,7 8 pixels, difficult to judge > > Detlev: Use a ruler? > > Jake: give it a background, see the touch area - it is an option but not > easy > > Detlev: was linked with target size - uncoupled it is less valuable (not > as important for large targets) > > Jake: We can look for elements without background padding positioned via > flex / % we can ask people if they think it can be measured and discuss > > Kim: Will we ask Wilco, will this be useful? > > Jake: We can ask him > > Kim: would be good > > Jake: Can ask him, send an email In the context of sites that may arrange things responsively/flexibly/based on viewport dimensions/etc, you really end up having to set a minimum spacing, and then require auditors to test essentially every possible variation of viewport width etc to find if at the point where targets are at their closest, their distance is still above the minimum threshold. Makes testing this non-trivial and potentially impossible to ascertain for sure (though in real world you'd test most common device viewport sizes and go by whatever breakpoints are defined in the page's styles). Also, on a generic point again: it shouldn't be just "touch target sizes / spacing". It's not just touch specific. Problems can also arise for mouse users with reduced dexterity. P -- Patrick H. Lauke www.splintered.co.uk | https://github.com/patrickhlauke http://flickr.com/photos/redux/ | http://redux.deviantart.com twitter: @patrick_h_lauke | skype: patrick_h_lauke
Received on Thursday, 20 June 2019 17:28:39 UTC