https://docs.google.com/document/d/1ouVFq4w-i0rchNHtTAG_JoRwHfYm9mN2MkxFBct1JSI/edit

touch target

Jake: summary from Tuesday – Lots of issues

<JakeAbma> https://docs.google.com/document/d/1sszSUKB8t3VuRzxHtOjLfQZjNYCw-xr_EbuMwW7WiGc/edit#

Jake: Notes at the end of the discussion, but not complete. This is a summary. There's more in the minutes
... the issue of making links even smaller to create the eight pixel margin
... overlapping for different reasons – popovers, drop downs, the index, sticky footers, headers, if overlap there's not an eight pixel margin and way more difficult to define
... problem with if you have five but next to each other 150 wide and 30 pixel tall why would we require a pixel spacing, because They are 150 pixel wide already
... there was no intent in the understanding document.
... also a question if we have research or guidelines from the big IT companies – Apple, Google, Microsoft
... I think Kathy had it somewhere, just not in the document

Kathy: it is in the document at the top it's an editor's note

Jake: okay – that's the guidance from the platform providers, I can take this one out
... Whole discussion about zooming, that's not sufficient solution, also pinching, not everybody can pinch. We need to flesh it out more
... and of course the whole automated testing will be one which is only possible if we clearly set the boundaries, so what exactly are the concrete boundaries for testing
... so this is a Short summary of what I remembered at the end of the comments. There might be more in the minutes

Kathy: if it meets the other guideline than it doesn't need spacing. So the question is if we have 150 x 30 doesn't need to have eight pixels horizontally. So that's just an update to the understanding document to clarify that point.

Jake: I don't think that'll be accepted as it's not in the normative text. There is a clear wish to split it up in the normative text,not talk about 44 x 44 if that's not what we wish for

Kathy: I see, so they are asking to modify the actual SC test
... so then we have to say the touch target pointer has a dimension of 54 x 54 pixels including the size of the target plus the spacing around it. So we can change it in that way. And then that would be in the normative text. Because then it doesn't matter – the calculation is whatever the target is, if you had 150 x 34, or whatever, then you would have to have the spacing between it
... the only thing that that does is if we did that then were having more than eight pixels of spacing around it were basically saying we want to have a 54 x 54. That's goes beyond what android. That's where that math came from.
... I wonder if we change it to if the target pointer is less than 48 x 48 CSS, including the spacing between targets, then – I'll write it out.

Working in the SC document

Jake: interesting gap – making the touch target smaller because It makes you pass the success criteria is also an issue

Kathy: I wonder if we simply have something in here that says – keep in mind this is adjacent touch targets. We are saying that that plus the spacing between the targets is at least 48 x 48 and just forget aboutwhat the 44 x 44<
... it depends what research we want to go by. A lot of the research out there that says if you have a touch target that's 48 x 48 pixels and it's adjacent that it is sufficient based on people's average size of their finger and being able to hit the touch target
... if we want to change it and go off of the other one we can do it. If you feel like we should increase that we can do it, we can change it to 52, that would be in line with what we had before

Jake: it is 51 because success criteria only Kicks in when you have 43 pixels not 44, it's less than 44. So at 43 we demand the eight, which is 51. And then it doesn't matter if it's eight on top of the left one or the right one because if you have 8+ the target which is 43, that's 51.

Kathy: what I'm saying is if we go by the research the recommended touch target is somewhere between seven and 10 mm which is essentially 48 x 48. I can have a very small – 30 x 30 touch target and I can pass this – we're not saying the touch target, which is our AAA would have to be that size
... looking at the Microsoft information because they had a lot of research and sound backing behind what they were

Jake: then you are trying to fill in the gap of the AAA

Kathy: the AAA says the target needs to be 44 x 44. This SC says you can do that with both the touch target and the spacing between it

Jake: this is from a different perspective, it doesn't feel okay because you can get links in there like 26 pixels high +8 pixels on top and at the bottom you will not even get to 48. So you will not solve it there to get more difference between those links.
... if you have spacing of 20 pixels on each side and a target of eight pixels you are fine. If we take a more extreme example so you still have 8 x 8 pixel touch target and there's 20 pixels in between, would that be okay
... those were also some comments, so I'm giving some examples – 5 x 5 target and lot of margin is that fine? If it's about spacing between targets than making the space bigger – is that exactly what we want while not increasing the touch target size?
... and also the idea, the success criteria how it was was you can make the touch target smaller and still pass the success criteria – is that what we want?

Kathy: I think it's a matter of figuring out

Jake: the distance between them can be bigger than the eight pixels we started with, but it can fail

Kathy: I get what you're saying, I'm not sure the best way to word it

I need to think about this one. It's probably best just to revert back to what we had, because this is definitely not going to work.

Jake: I could have missed some of the details, solutions. Writing them down, listening, not thinking of summarizing them. Just thought of summarizing near the end of the discussion not when they started.

Marc: I'm thinking carpentry terms – you must have a total of 2000 pixels in your touch target. I'm not sure the exact math.

Jake: if you have a list of Very long links I'm not sure if 2000 will fix it – if there are only 26 high the problem of clicking on the top one when you mean the middle one still a problem, because they are pretty wide. And also the calculation will be even more hard I think

Marc: I have a Google link, it's 1000 pixels and total but only 22 high, so they are stacked and you can still click the wrong link. So you solve one direction but not the other one

Jake: Andrew suggested trying to take the width and height apart near the end of the conversation

Possibility from Kathy: For adjacent touch targets that are less than 44 by 44 CSS pixels then there is a minimum of 8 CSS pixels for the height or width that is less than 44 CSS pixels except when:

Another possibility from Kathy: For adjacent touch targets, if the width or height is less than 44 CSS pixels, then there is a minimum of 8 CSS pixels between targets except when:

Kathy: I'm saying if the width or height, and then minimum spacing between targets

Jake: you'll still get the feedback we have 30 and need margin okay, we'll just change it to 22

Kim: if it's less than 44, then the total needs to be 43+8

What is 14 mm in CSS pixels, because that's really what were trying to hit, the 14 mm size in general

Kathy: we're trying to hit 14 mm, doing the calculation, we intended was 44+8, converter, 52, 53. So I'd feel better saying 53
... Saying 52

Jake: define height or width

Kathy: If we look at 14 mm around is really 52 CSS pixels by 52 CSS pixels that you have between different targets, the space plus the target is that – if we have that you've got to right next to each other. We'd want 52 x 52 around for one of them. How we've written it now I'm not convinced that we've actually got to that.

Jennifer: Where did you get the 16 from

Kathy: if I had two targets, eight CSS pixels between the two – that is what I intended

Jennifer: can we reference other success criteria within a success criteria?

Kathy: no

Jennifer: If the success criteria which is AAA were to change I wonder if we would have to change that in herto – that something we don't have to worry about right now
... bizarre edge case but that's only 30, and button on either side. There are two buttons that are compliant and then middle button has a spacing of 20 pixels on one side within a spacing of zero on the other side.

Kathy: if you have 20 pixels on one side and zero on the other side, you're fine, you still have enough – you would still target on that item you might go over. You still have that with and that's what we were generally trying to get to
... so it doesn't matter if you have all of it on one side and the other target is right next to it
... as long as it Combined is 52 CSS pixels overall
... still have to change other language

Kim: It looks like the SC works now. We still have to change the other language to match the SC. We'll pick this up in two weeks – next week is Thanksgiving.

https://docs.google.com/document/d/1sszSUKB8t3VuRzxHtOjLfQZjNYCw-xr_EbuMwW7WiGc/edit#heading=h.mntlv4jvrc29

This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Default Present: Jennifer, Kathy, MarcJohlic, Kim
Present: Jennifer Kathy MarcJohlic Kim Jake
Regrets: Detlev
No ScribeNick specified.  Guessing ScribeNick: Kim
Inferring Scribes: Kim
Found Date: 21 Nov 2019
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)