RE: Target Size SCs issue feedback


On 26/05/2017 15:53, Repsher, Stephen J wrote:
> Josh,
> 
> ØSteve Repsher said:
>>> The little testing it has undergone has turned up several issues, 
>>> namely focus highlights and >overlap, Which have the potential to end up creating significant inaccessibility.  However, if the technique is to use a 44 pixel line, then I’ll support it.
> 
>> Thanks for the suggestion. In terms of your request for clarification but I see that Kathy did reply.
> 
> To add some context to the sentence you quoted me on, I was talking 
> specifically about the CSS padding and negative margin technique that 
> had been proposed to handle links within blocks of text.  Yes, Kathy 
> replied that this would no longer be used for the AA version, but my 
> objection is for the AAA version.  What technique do we plan to 
> document to make 44 pixel targets within blocks of text?
> 
> I find it hard to believe I’m the only one not concerned about this, 
> but perhaps I’m missing where on list or on GitHub the issues with 
> this technique were thoroughly discussed and resolved.  Yes, it’s AAA, 
> but it’s still an SC that needs accessible techniques.

The hardline answer would be: you want AAA? Avoid having touch targets that are too small.
[Steve] Again my concern is the technique, not the SC itself.  If we want to say AAA conformance means 44 pixel lines, then okay.

If you have content where there are lots of links in inline content which may potentially overlap, this may require you to either create some alternative (like a separate list of all links, as a conformant additional alternative), or to rethink how you present your content, or...
[Steve] You can't have potential overlap if padding and margins are positive, and that's my point.  Putting aside the unresolved issue of focus indicator confusion which creates inaccessibility, do you really expect authors to think about and test their content so much that they will consider all possible viewports, zoom levels, text adaptations, etc. when analyzing for potential overlap?  That doesn't seem feasible to me unless we're also going to produce a tool to do that for them, which I'd argue is practically (and perhaps even theoretically) impossible.

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 Friday, 26 May 2017 18:21:17 UTC