W3C

– DRAFT –
ARIA Authoring Practices Task Force

04 November 2025

Attendees

Present
arigilmore, CurtBellew, jongund, Siri
Regrets
-
Chair
-
Scribe
arigilmore

Meeting minutes

No meeting next week because of TPAC

Publication planning

Matt_King four PR's in the milestone for publication. none are technically ready yet

Matt_King: we will not publish tomorrow, but SkipTo PR is close

Matt_King: reviews pending from Adam_Page and arigilmore for PR 3213

the link checker is failing, but not sure why

jongund might need to do a merge to fix it. Didn't touch those files

howard any idea why this is failing?

jongund can try submitting again to see if it will resolve

@howard a rerun should work

jongund maybe the link I am using just has a hashtag and I need to update it

jongund does the link checker check if a link is just a hashtag?

<Siri> can you add link to the PR?

w3c/aria-practices#3213

Matt_King does the skipTo script end up creating a link with just a hash on the actual rendered page

jongund yes if you are in IOS or android

jongund reran it and the link checker passed

jongund talkback and VO would be able to activate the link off the page and using javascript I can open it

jongund this only happens in the popup mode

jongund for some reason it does see the link, but it just won't activate normally when off screen

Matt_King just looking for the pending reviews. This will be 1 of the 4 pending for publication

Matt_King will change to later date for publication, but not sure when

jongund would be nice to check on android too

jongund will find someone else with another android advice

Siri has a pixel and can test the SkipTo feature

jongund turn on talkback with chrome and use swiping and see if you can double tap to open it and make sure you can activate the menu's

jongund I did notice a big difference between android and IOS. Android with a keyboard willl let you use cursor keys

Matt_King the arrow keys correct ?

jongund yes

Matt_King on IOS you can only use tab keys

Issue 3193: listbox example with aria-actions

<Matt_King> github: w3c/aria-practices#3372

Matt_King didn't see any commits yet. how's it going?

CurtBellew waiting for something, but can't remember haha

Matt_King: need to summarize the changes

Matt_King: in ARIA we had a separate meeting

CurtBellew has a list of things to implement. Wants to use aria-live when going up and down. Add favorited or starred to the actual name of the option

CurtBellew how will the buttons get read out

Matt_King this is a browser bug

Matt_King concrete changes are changing the name when the item is marked as a favorite and announcing when the item is moved up or down

Matt_King: you can model from listbox from rearrangable options, which has move up and move down

Matt_King: aria-live is included in that example which has strings that we previously discussed with the exact announcement we wanted

jongund: one more thing with the left right keyboard interaction

Matt_King: during TPAC we will take a look at aria-actions. We will use the PR preview for the meeting

Siri: question: these interactions are only on hover or focus?

how does the user know if they never use it?

Matt_King: this is a discovery thing just like gmail

Matt_King: normally if you are gonna interact with the list then you will eventually see these interactions

Matt_King: the goal here is to help people show how aria actions can help them

CurtBellew thinks he will have time before TPAC

Issue 3377: Spin button pattern home and end guidance

<Matt_King> GITHUB: w3c/aria-practices#3377

Matt_King: there is something about the spinbutton pattern that turns out to be problematic, but we only discovered when writing screen reader tests in the ARIA AT screen reader group

Matt_King: in the spinbutton pattern is that Home should set the value to Min Value and End should set the value to Max value

Matt_King: however if you can type into the box then home/end have specific interactions

Matt_King: support standard editing keys for single line input and i've listed the four lines that discuss this

Matt_King: the last note states we should not have any javascript editing the values

Matt_King: historically we thought of this as a slider, but sliders are not capable of typing a value

jongund to me it seems like the home/end key should change the number rather than the cursor, but one has to win right?

Matt_King screenreaders are expecting home to do what it normally does with those keys

Matt_King lets say the spinbutton was a large value with incorrect value you could arrow left, but home should do that for you to go back to the beginnning

jongund: we should adopt patterns consistent with how screenreaders use the keys. We will have to wait along time before those get updates

jongund: honoring the path of least resistence then we should go with that

Matt_King i don't think screenreaders will agree to change the home/end in cusotm spin button because then they will have to have different behaviour in custom spin button vs standard html with input type=number

Matt_King browsers don't provide any special home/end behaviour

the home/end move the cursor to the beginning of the input

Siri but here you are saying when you click home then it will change to the min value and not bring it to the beginning of the input

Siri it is changing context without informing me

Matt_King: we should remove that home/end guidance

Matt_King: the people spinner you will have to page up and down to hit the max/min values

Matt_King: in a spinner that takes a long range of values then you can just type the min value

Matt_King would anyone be opposed to removing the home/end guidance

jongund: thinks its fine. The most logical thing to do

<Siri> +1

jongund depend on the range there should be some guidance on page up and down

jongund maybe page up and down should move to a certain percentage of increments. then eventually go to max/min values

jongund if the range is 0 to 10 then page up/down should go to 0/10 and then if it's like 100000 then we should increment by percentages

Matt_King page up/down should move at least 10 steps in each direction if the spinner has a very large range then maybe try dividing by 20?

Siri thinking from user perspective. How will end user know what will page up/down will bring them to. Can we assign a key for min/max

Matt_King you can type in the min/max. That's how they are different from sliders

Matt_King: i think i have enough information to at least unblock the ARIA AT project and propose changes to the existing spin button

Matt_King: that will effect how we change page up/down. We should test this in ARIA AT and will need to update after the spinner changes

Issue 3373: "Read this first" uses client-side JavaScript

<Matt_King> GITHUB: w3c/aria-practices#3373

Matt_King the "read this first" content on the top of each pattern page and other pages and it's being done by client-side javascript. This should not be done by the client side anymore

Matt_King: we want the content fully controlled not in the build repository

Matt_King we should control what the content is and where it goes in the repository and the rest is in the build

Matt_King we already have a template, but i also noticed that when I do a local preview of the example pages the content shows in local preview. That one is being done in a way that the w3c team is happy with, so how is it happening if it's not client side javascript

howard: the javascipt file that's there for it to render locally it does reconstruct the html and build it on the page

Matt_King what's the difference between the read this first and the usage warning?

Matt_King: see's the usage warning locally in the example file

Howard: not for me...

Matt_King: is in his local then goes to content then patterns, accordion, examples in a windows. I have skipTo, related links, design pattern, heading H1. in vscode it shows it but not in windows directly

Matt_King should we do the read this first the same way as usage warning

Howard: yes, but there is another comment asking if we are comfortable with contributors viewing this

Matt_King we should do it the same way

howard: what controls what page it gets used on on for the usage warning

Howard it's probably being done through App.js

Howard: App.js is on all the pages

checks for a certain path

Matt_King can we have a separate file

Howard: yes

Minutes manually created (not a transcript), formatted by scribe.perl version 248 (Mon Oct 27 20:04:16 2025 UTC).

Diagnostics

Maybe present: howard, Matt_King

All speakers: howard, jongund, Matt_King, Siri

Active on IRC: arigilmore, CurtBellew, jongund, Matt_King, Siri