See also: IRC log
<Kathy> https://github.com/w3c/wcag21/issues/60#issuecomment-277159966
<Kathy> The size of the target is at least 44 by 22 CSS pixels for pointer inputs except when a link or control has an alternative link/control that is at least 44 by 22 CSS pixels, or is part of a block of text.
Kathy: it can be either way, 44 x
22 or 22 by 44.
... if you have in-line links on a block of text you can get
links stacked on top of each other. if 44 x 44 there is overlap
or spaced out too much
... if 44 in at least one direction okay. If you had one
in-line link and another in-line link right below 22
... the thought process is generally speaking 44 x 44 but
exception for in-line links 44 x 22 because that's what we can
do realistically
Shadi: for links it's different than buttons. Buttons are the bigger issue
Kathy: the only challenge that I
see is discussion about users being able to zoom to do it but
within blocks of text if you zoom you then lose the context and
you have the same problem that low-vision users are saying with
the fact that you have to then scroll horizontally and
vertically. Maybe that's not such a big issue because if you
are magnifying the screen you probably want to be...
... clicking on that link so you're going somewhere else
anyway
Shadi: if there are many links in
that paragraph it's probably a bad paragraph – an issue
somewhere else. But if it's a list, a table of contents or
something that's maybe one of the issues that I can
imagine
... forms on mobile – keep expanding to hit the button and then
making it smaller again to get an overview of where you are in
the form – that's tedious, that's my most frequent. Table of
content I can imagine that that might be an issue – but if I
expand once then click to the place I want to go through that's
fine. So I'm thinking this is a good compromise.
... form – when you have a lot of buttons close to each other
could be radio buttons and links. Very often sometimes you have
the label which is linked and then the side of it might be a
help page, a link inside the label sometimes so if you don't
hit the label correctly or if you want to hit the link and not
the label or the label but not the link it causes a bit of a
challenge
Kathy: we really shouldn't have things within labels
Shadi: or even close by
Kathy: so in general we think that 44 x 22 is a compromise we can live with for in-line text links
Shadi: I'm happy to be used as a use case – want to make sure not to generalize, could be other cases
Kathy: probably not many scenarios where we are going to have overlapping text, but may have a bigger issue at that point with too many links anyway
<chriscm_> Kim: Make sure language does not imply one particular platform.
https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/WCAG_2.0_understanding_that_needs_language_update
<chriscm_> Kim:The second example includes "mouse" which is very desktop centric wording.
<chriscm_> Kim: Guideline 1.1 "...link to an audio clip..." in mobile all you get is the audio player, so considerations for mobile are different.
<chriscm_> Kim: This serves as a list of items that are directly applicable with updated language.
<chriscm_> Kim: Items are marked fixed in the document. You might take the suggestion and fix it differently.
<chriscm_> Kim: Can leave notes that "change not going to be made" in the document. Might be a better method than pull requests.
<chriscm_> Shadi: Might be best to coordinate with the chairs on workflow. Overall makes sense.
<chriscm_> Kim: Workflow within the group, need help reading techniques and spotting these.
<chriscm_> Kim: Fine just saying this needs a fix, without a fix suggestion.
<chriscm_> Kathy: We need to coordinate efforts. Will try and help.
<chriscm_> Kathy: There is an August deadline for success criteria to make it into 2.1.
Kathy: anything you can do to
push the conversation further on these to get these to a
finalize point is going to be key
... we also have some of these with no manager
... that's why I've been spending a lot of time talking to
people – for example trying to get touch target size to the
point where it can be accepted
... August deadline even though it's not going to be out until
later. In order for the working group to have time for comments
they need those success criteria to be locked down. That means
they have to go and satisfy all of the requirements for a
success criteria – so it has to be testable, implementable
across different technologies and there has to be technology to
support it so it has...
... to be doable by the developers etc.
... so that's where we need to make sure that we are pushing
this. Over the next weeks I plan to try to get at least one or
two of them that we are working on and try to see if we've got
some things that we can do to address people's comments to try
to move this to a point where we are not taking the success
criteria in a place we don't agree with, but we are
compromising and looking at...
... all the different concerns to see how we can address
those
... that's where we are going to spend some time. There's going
to be a coordination process – I'll have more on that next
week. For now just know that we really need to be spending some
time pushing these getting people's feedback trying to get
these to the point where were actually finalizing them
... I've taken the priority order we did on the call and
started working on them from there. If you have some extra time
and you can start driving the conversation in these, putting
comments, suggesting if you see something or we need to drive
the conversation further if you can help do that
... we've got a lot of work to do before August
This is scribe.perl Revision: 1.152 of Date: 2017/02/06 11:04:15 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Present: Kathy chriscm Kim Shadi Regrets: Detlev Henny Jonathan Jatin No ScribeNick specified. Guessing ScribeNick: Kim Inferring Scribes: Kim Found Date: 30 Mar 2017 Guessing minutes URL: http://www.w3.org/2017/03/30-mobile-a11y-minutes.html People with action items:[End of scribe.perl diagnostic output]