Meeting minutes
Yes, happy Global Accessibility Day
Announcements
MaryJo: welcome and Happy Global Accessibility Awareness Day, some regulars not available today.
… from GitHub, 1.4.10 reflow issue / conversation quite a lot to follow...
… i have split into four
… copied into four topics
bruce_bailey: requested the 4 items to be listed
<maryjom> 4 aspects of the SC: the definition of CSS pixels (or use of device-independent pixels), guidance for non-web documents, guidance for non-web software, and guidance for closed functionality
<maryjom> one issue opened for each aspect
maryjom: Should help us think about aspects, for example definition easily its own topic
… there remains some bugs/linkages and I will meet with Michael C to try and resolve
<Zakim> dmontalvo, you wanted to update status
dmontalvo: I am coording and comenting
maryjom: Okay, i will let Daniel cordinate Michael Cooper on that.
maryjom: I am setting separate meeting on closed functionality. People interested should have that invitation. Next Wednesday is first meeting.
… Also group working on command line issues, meeting Fridays, and those conversations will continue.
… I was not able to coordinate Reflow sub group, so we will talk today.
Project standup and planning
MaryJo: i covered most of the reorganization....
… Screen share of project board, has a number of items before next publication of editors draft
… I looking for volunteers on open ToDo items, but there may be at least one to kick back to AGWG
bruce_bailey: Are these blockers for public review, or editorial draft?
maryjom: For public review -so we can put a public draft out without broken links
maryjom: These are all blockers for first call for public review...
… many syntactical, like headings, but reflow conversation is substantive
… we need to have all the 2.1 content for this first call
WCAG 2.2 just went out for CR yesterday, so that will pause our work on 2.2 sc
maryjom: I was/am hoping to be done with new content by end of June, so that will be close
… we have a timeline starting up with EN 301 548 going through a refresh
… so this is reason to keep pushing
Mike_Pluke: Agreed this will important for next update on EN 301 549...
… but deadline still open on that as there are significant details to work out before meetings start
maryjom: Our publication takes 4-6 weeks AFTER we all agree that we are done...
<Chuck> long and deliberative publishing process.
maryjom: long publication process, and we have summer vacations and TPAC in September
… we will need 2nd public working draft with 2.2 additions, and responses to public comment.
… those are our known unknowns, so we must keep pressing.
Discussion on definition of Device-independent pixels
<maryjom> w3c/
maryjom: issue 162 , css pixel dependancy
… conversation started with 1.4.10 reflow, but that is NOT only sc impacted...
… issue 162 lists other sc impacted
… we had a good bit of consensus in April...
… but that broke down a little bit as we considered other SC
… we have a couple directions we might go in.
<maryjom> Note: Non-web software and their accompanying platform software do not use measurements of CSS pixels. Therefore, platform-specific pixel density-independent measurements should be used. Examples include: px in iOS and MacOS, density-independent pixels (DP) for Android, and device-independent pixels for Windows.
Phil: I suggest Device Independant Pixel but further discussion has clouded that...
… we probably need some physical fall back as welll
+1 for conversion tables or easy to understand examples from device specific examples, CSS pixels and visual angle
FernandaBonnin: Is compelling to keep using wcag term , and visual angle is a good way to approach , but i am not in favor of providing sample calculations.
maryjom: I tried doing a table of conversions, which requires viewing distance, but those are all RANGES in authoritative sources...
… numbers used in CSS pixel definition seems a little arbitrary
PhilDay: Impact for low vision complicates choices. Agree with FernandaBonnin to provide tables -- but numbers might not follow strictly from calculation.
More extreme examples: digital signage, and a drive up ATM designed to be used from a vehicle
Mike_Pluke: Agree that real world implications are going to make this hard. Not just mobile phone but display on a wall. It is a huge range...
… It is a major challenge, especially given intent to accomodate people with low vision. Agree with some tables.
FernandaBonnin: This issue of font size and viewing distance has been very challenging for us...
… We are looking for some medical references, and all we have is early research from WCAG 2.0.
maryjom: I would caution that we might not want to talk about font size per size. With reflow key thing is viewport which is different....
… The other SC of particular consideration is Target Size -- but that is different sort of accomodation.
The other place CSS pixel is critical is with flash rate, which is just a patch of screen.
Mike_Pluke: As such, we are not defining a character size. So I take your point of reflow being about viewport.
… But I think mathematically it still comes down to the same issue.
maryjom: I think there will be some differences in reflow.
Mike_Pluke: Possibly, but I think issues are linked closely.
maryjom: With table approach, what we should we use?
FernandaBonnin: I suggest we pick common use case first, like phone, table, display on wall.
maryjom: Do we have authoritative cites for those use cases?
Mike_Pluke: We also have things like watches -- maybe even closer than phone
maryjom: I was looking for distance for computer versus laptop versus tablet -- no success.
… For television, there are some recommend and typical viewing distances
LauraBMiller: I will do some research, I think I can find references.
maryjom: Thanks, I could only find televisions, and those made no reference back to CSS pixel.
<LauraBMiller> From Bard (I will have to check for sources:) National Institute of Standards and Technology (NIST). The NIST study found that the optimal viewing distance for self-service kiosks is between 18 and 36 inches. This distance allows users to comfortably view the screen and interact with the kiosk without straining their eyes or neck. Other studies have found similar results. For example, a study by the University of California, Berkel[CUT]
maryjom: We talked about visual pixel being an arc , does that impacted by flat screen -- or it just arc inside eyeball?
PhilDay: I added some typical distances. From freeway sign (maybe too niche) but ATM will be at arms length, so there are some numbers.
<LauraBMiller> One study, published in the journal "Human Factors", found that the optimal viewing distance for self-service kiosks is between 18 and 36 inches. This distance allows users to comfortably view the screen and interact with the kiosk without straining their eyes or neck. Another study, published in the journal "Computers in Human Behavior", found that users were more likely to complete tasks successfully when the kiosk screen was loc[CUT]
<FernandaBonnin> https://
FernandaBonnin: Waynes research seems to provide some documented research we migh use.
maryjom: How can we ask for that?
Mike_Pluke: I got credible results from chat GPT
Mike_Pluke: But some results clearly not credible.
<Mike_Pluke> The typical viewing distances for different devices can vary depending on individual preferences and usage scenarios. However, here are some general guidelines for typical viewing distances: 1. Mobile Phone: The viewing distance for a mobile phone is typically around 12 to 18 inches (30 to 45 centimeters). This range allows for comfortable viewing and interaction with the touchscreen interface. 2. Laptop: The viewing distance for a la[CUT]
<Mike_Pluke> Laptop: The viewing distance for a laptop can vary depending on the screen size and personal preference. Generally, a distance of around 20 to 30 inches (50 to 75 centimeters) is common. However, some people may prefer to sit closer or farther away based on their comfort and visual acuity. Desktop Computer: The viewing distance for a desktop computer is typically a bit farther than that of a laptop. A distance of around 24 to 36 inches[CUT]
<Mike_Pluke> Smartwatch: Smartwatches are designed to be viewed up close since they are worn on the wrist. The typical viewing distance for a smartwatch is around 6 to 12 inches (15 to 30 centimeters), allowing for convenient interaction with the small display. It's important to note that these are general guidelines, and individual preferences may vary. Factors such as screen size, resolution, and personal visual acuity can also influence the opti[CUT]
LauraBMiller: Please see what I put in above as well.
maryjom: We have conversation about people with low vision complicates factors
LauraBMiller: Bard resouce i pasted in did provide some citations.
… they do reference viewing cone and viewing distance
BryanTrogdon: Can we approach from human factors to give less brittle advice?
… As long as, from my perspective, it is certain relative size, text is accessible to me.
maryjom: This is where I think we run into issues because WCAG SC do not include "size" even though 508 and EN 301 549 do -- and they are not an exact match...
… Again with reflow, SC is not about character size. What exactly are we measuring? What does CSS pixel mean for viewport...
… Does view port for testing have to be a certain size?
… some platforms will not use CSS pixel as all, so what are implications from that....
… when I tried to create some tables, I got objects from colleages because pixel density is an issue
3/16 of an inch for upper case "I" is what LauraBMiller was referring to I think
… see section 402.4 https://
LauraBMiller: Section 508 gives a minimum font size for closed products, so that seems like it could be a reference...
… If there is zoom involved, that solves problem, so that feaure can meet need. If measured, does zoom or default size the one to use?
Sam: Laura's reference 704 and gives a mm / in size. I would recommend giving choices, but context matters...
… If a touch target, we have a measurement for that which is agnostic to readability and font size.
Mike_Pluke: People only zoom to get to a legible size for them, but that not be possible depending on the technology....
… the red flash is really about area to the eye, so that is very dependant on distance.
maryjom: When one increases character size, that might not effect Red Flash Threshold, but reflow SC does not specifiy need to support "blocks of text" ...
… but that context only in Understanding. SC is a little fuzzy, and all the worse as we try to address for WCAG2ICT.
… Hard enough challenge for web content, so non-web content becomes even more tricky.
… Please help, provide research if you can.
maryjom: We do have option of formula, but event with that approach there is fuzz in the assumptions...
… wrt Reflow does it even matter if the math is loose. Reflowing text is reflowing text...
… But then with target size, that is a real world metric -- big enough to tap with finger.
maryjom: I don't think we will be able to come up with hard numbers, so how do come up with ball park figure?
PhilDay: Thinking out loud, but instead of numbers can we explain with intent?
<Chuck> I need to depart for another meeting.
PhilDay: so with reflow, the requirment is reflow and not a measurement.
Mike_Pluke: The only reason for Device Independant Pixel was to have a term which does not use "CSS" at all for the technology...
… That only helps with a narrow scenerio, so we might still keep that. Even though DIP does not solve all our issues...
… but we can define the term and use it where it makes sense.
maryjom and Mike_Pluke agree that having term still leaves the underlying problem.
Sam: For target size, we can derive from "at arms length" we have prepoderance of literature and give a specific number...
… for reflow, the wcag2ict guidence will be less exact.
Sam: I more concerns with term in other SC
maryjom: For CSS pixel it is going to be an estimate.