Meeting minutes
alastairc: welcome all
alastairc: does anyone want to (re)introduce themselves or announce anything?
WCAG 2.x updates https://lists.w3.org/Archives/Public/w3c-wai-gl/2026JanMar/0015.html
alastairc: I'd like to welcome Patrick_H_Lauke , who has taken over facilitation of the WCAG 2.x TF
alastairc: anything you wanted to highlight, Patrick_H_Lauke ?
Patrick_H_Lauke: just a reminder for anyone who hasn't had a chance yet to have a glance over that email
<AWK> When did the email go out?
Patrick_H_Lauke: we're getting ready to line up the next batch and will let you know about it soon
alastairc: the email is linked in the agenda item
Patrick_H_Lauke: 12th of January
<alastairc> https://
alastairc: this board shows all issues that were in the email
alastairc: the list in the email is probably the best start, rather than the project board
Patrick_H_Lauke: also everyone feel free to join our call, it's on Fridays
Subgroup updates
alastairc: in the user control group, we're still looking for more participants, our group has things like no flashing and adaption type requriements
alastairc: we've been extending out tests and methods for those
<Zakim> Rachael, you wanted to go next
Rachael: I'm leading the 'single sense' group right now, still in need of a co-lead. We've focused on definitions so far, which we'll talk about more later
julierawe: in our group we have sufficient people who speak English. We're still looking for speakers of Mandarin and Hindi
julierawe: no major blockers, a lot to crank through
<Zakim> kevin, you wanted to feedback on Sign Language
kevin: I'm leading the sign language work… at the moment we're looking to validate user needs
kevin: we have 2 people but neitehr of us use sign language
<Zakim> hdv, you wanted to report on Errors, task completion group
<julierawe> Addition to Text and Wording update: We're also looking for speakers of Russian and Arabic.
hdv: Errors, task completion & help - talked about a specific requirement: error notifications.
… looked at the wcag2 version, how useful it is, looking at other sources, and see if there are larger issues.
… also looked at definitions.
<Zakim> Makoto_U, you wanted to report on Image and Media alternatives subgroup
Makoto_U: our group meets weekly before the AGWG call, we just had the week 2 meeting today, we're discussing definitions of 'available' and 'equivalent'
Makoto_U: we identified some issues that we must tackle in the next few weeks. our discussion has been passionate and active so far
Francis_Storr: in our subgroup we started going through the feedback that came in, started to address it in the requirements
Francis_Storr: everything going ok
giacomo-petri: in our subgroup we are still looking for a co-lead. We're currently looking at definitions and have drafted most of them now
giacomo-petri: our next step is to complete the definitions from exploratory
giacomo-petri: and we're starting to identify the issues in GitHub
alastairc: thank you. Any other questions or comments about our subgroup work?
alastairc: is everyone happy, or do people still want to join any of the groups?
Definitions https://docs.google.com/document/d/1WhaesDbhuB8SmOeX05-qfo0nSvdXHugpCJUw8rA9nfI/edit?tab=t.0
alastairc: also feel free to email the chairs if you don't want to say it here
alastairc: we got the Definitions doc ready. I would like to suggest we run through quite a few of them in this meeting
alastairc: I'll say, here's this definition… I'll skip the ones taken from WCAG 2
alastairc: then let's focus on the ones that have issues associated with them. A couple of people also commented in the document
<LenB> present_
alastairc: the proposal for 'Blocks of text' (*reads out*) is slightly different from WCAG 2. I suppose 'logograms' helps to internationalise
<AWK> Are we creating requirements in v3 that require blocks of text to be defined?
<Patrick_H_Lauke> is there an actual definition of "logogram"? not something many may be familiar
<CarrieH> agree with what Patrick_H_Lauke is saying here, this isn't a common word
<CarrieH> I have no idea what it means
<GN015> what is a logogram? My native speaker colleagues doesn't know th term, either. So, should it be used as a part of a defining phrase?
GreggVan: this definition isn't useful as in the current way this could mean entire documents. I have no idea what it means
<CarrieH> and this could be problematic for literal thinkers and individuals whose primary language isn't English
<GN015> +1 to Carrie (she refers to my question on logogram)
<Patrick_H_Lauke> agree with greggVH that this definition doesn't define the start/end point of what counts as a "block"
<alastairc> AWK - the requirements using this are in the draft https://
Rain: visual blocks of text has more to do with how things are placed together
Rain: rather than what's intended to be read together
Rain: that will need some solid visual examples to go with it
<bbailey> FWIW, currently "block of text" is "more than one sentence of text" https://
<Patrick_H_Lauke> beyond "code" semantics would also be ... logically if things do belong together or not
Rain: semantically, when you're writing code, a block could be anything from a section to a paragraph to a list item
Rain: there's a scope challenge
<Zakim> Rachael, you wanted to react to Rain to ask how best to capture the feedback
Rachael: process question… how are we capturing the feedback we're getting? is each subgroup capturing this?
alastairc: I was intending to summarise it next to each item
<bbailey> I also had to "look up "logogram" which is "Linguistics. a symbol that represents an entire word directly rather than representing a speech sound, such as a Chinese character."
<bbailey> https://
julierawe: re logograms… they are one 'picture' that describes a word, like in CJK languages
<Patrick_H_Lauke> sure - my question was more "is that definition readily available for somebody reading the spec"
<kirkwood> does a block of text need a “bounding box” to be a “block” Thus, would point to having “space” around it?
<SydneyColeman> present
alastairc: for some definitions it will be useful to see them in context… you can find this in the spec preview in the list of definitions, each definition shows where it is used
<kirkwood> often addressed in a style guide
<alastairc> q/
joryc: Rain hit a lot of points I wanted to make… wanted to say, there's three versions of 'block of text'; the code base one, visual blocks of text and blocks that are a complete idea
joryc: depending on which requirement is referencing this, we may mean a different thing
joryc: we need to be clear about which it is
<Zakim> AWK, you wanted to say Are we creating requirements in v3 that require blocks of text to be defined? Perhaps we should avoid doing so..
AWK: I see we have 3 criteria in WCAG 3 that use 'blocks of text', I wonder if this term is problematic enough that we maybe should not use it. In some cases it may be easier not to have a definition at all
<kirkwood> +1 to not using it
alastairc: then we a way to differentiate wrapping from non wrapping text… which isn't the distinction exactly either, but conceptually what I mean
GN015: coming back to the term 'logogram'… it's a kind of character that represents a word. I think we don't need the word 'logogram' as we already talk about words
alastairc: what about languages that don't use words?
<kirkwood> +1 to removing logogram
GN015: a word is not defined as a sequence of letters
<bbailey> +1 to AWK to avoid "block of text" -- and thanks for reminder that we avoided term in developing 2.1 !
GN015: it exists in all languages incl Chinese and sign language
<CarrieH> +1 to removing logogram. Also language doesn't equal words, there are people in this world that are "non-speakers" that use AAC devices..
<Zakim> hdv, you wanted to ask if checked with CSS
hdv: Even in Mandarin, a set of characters could be a "word". Would love to hear from some CJK speakers here. I would be +1 to removing "logogram". We might find something of interest in the DOM and CSS specifications RE how to finesse "blocks of text".
<CarrieH> More about what AAC devices are/what they do... https://
Jennie_Delisi: wondering, do we consider like speech and language pathologists, it would be nice to align with standard definitions for things like sentences
<GreggVan> +1 Kanji are jus differet characters. words in japan can be made up of both kanji and hiragana
<bbailey> Per Hidde, also: https://
<Makoto_U> +1 to removing logogram (at least for Japanese)
alastairc: a lot of this is about how text flows, not sure if getting in semantics of language will help too much
GreggVan: wanted to read out loud Makoto_U's comment '+1 to removing logogram (at least for Japanese)'
alastairc: ok, the next one is 'body text: text comprising the main portion of a document'
hdv: +1 to your point; I think there are too many interfaces that WCAG applies to, or is going to apply to, that have different sets of what body text could mean, or does not have body text, so it seems to me that is not a helpful distinction to make.
alastairc: I'm not sure about body text, as there are too many different kinds of interfaces
GreggVan: I don't think we should have 'body of text' as something to make rules about
GreggVan: like, can we say that is the only part that's important on the page?
GreggVan: I have seen things like 'this should be in the same font size as the body text'… so basically it's used as a reference point for what some other thing should be
julierawe: this is about text related requirements… there may be a way we can avoid the text 'body text' altogether
julierawe: we were just reusing what was there and can look at alternatives
alastairc: I think one I saw was about text sizing, eg thing could be relative to the body text rather than starting point
Jon_Avila: I think blocks of text and body text could be useful in other places like text contreast
alastairc: if it was needed could it refer to platform body text?
alastairc: ok, next up is 'clear language'
<julierawe> We can remove "quickly"
alastairc: one issue could be that it relies on quickly doing something
hdv: We discussed it in our subgroup. Our consensus is that "clearly" isn't helpful...
… because it isn't helpful. I think this is the same with this phrase as well.
… Its hard to measure but still a goal to strive for. The word "clear" can be really tricky.
<kirkwood> maybe plain language definition: clear, concise communication designed for readers to quickly find, understand, and use information easily, avoiding jargon, complex words, and convoluted sentences
GreggVan: there shouldn't be definitions for things already in the dictionary. 'clear language' is a phrase so that's fine
GreggVan: because definitions are normative, they have to be objective
GreggVan: so the definition has to use objective terms.
GreggVan: this is a critical requirement, but we're wrapping our head around the treshold of it being objective… if we aren't going to use it normatively we don't need to put it in the list
julierawe: the history from Silver days… the reason why 'clear language' has been used for a while, it's similar to 'plain language', but we needed an alternative term as some countries have laws around plain language
alastairc: do you know if it is used by any requirements?
julierawe: it's currently a section heading and used in an assertion
<Rachael> +1 to "clear language" being deliberately chosen to avoid overlap with ISO and legal requirements. For WCAG 3 we may be able to reassess that
alastairc: the way it's used at the moment I'm not sure we need a definition
<kirkwood> yes it is
GN015: I seem to remember that it is defined under reading level in school years
<kirkwood> in the US. plainlanguage.gov
GN015: so 'clear language' could avoid the legal implication
alastairc: and there's a thing that it doesn't have to be 'plain' it just has to be 'clear'
Bryan_Trogdon: our understanding how we look at it through the @@@ scale… it's a way you can test and evaluate that
Bryan_Trogdon: 'plain' feels actionable and 'clear' more how you interpret it
<kevin> +1 to being removed based on current use
Heather: we probably want to get to the spirit of what we mean by plain language
Heather: then the requirements under the heading are what becomes testable
julierawe: a couple of folks were comparing it on reading level… I remember people had concerns about reading levels internationally, as it is different in different countries
alastairc: ok, next one is 'complex numerical information'
alastairc: this seems to be a relative rather than an objective definition
julierawe: I believe this one is at exploratory because we're going to figure out more
<kenneth> corroborating Julie, it's used in numerical alternatives available which is exploratory
alastairc: i'll leave it there
alastairc: then 'content'
alastairc: why did you delete the later part of the definition?
DuffJohnson: to disambiguate it from markup
alastairc: that seems to make it very wide
hdv: I am worried if we separate content and markup and content means something different than what it meant in WCAG 2.
… I am afraid it will confuse people moving between versions.
… I wonder if that is helpful or not.
<kevin> +1 to Hidde's point
DuffJohnson: the subject of empty markup and whether that's conforming… in other to conduct that debate one has to distinguish between content and markup
DuffJohnson: in my space, PDF, we need to distinguish this
jkatherman: content and markup aren't the same and not disambiguating them isn't helpful
jkatherman: I think separating them is helpful
<Jon_Avila> In WCAG 2.2 the definition says "including markup" so even today it's not limited to something created by markup.
hdv: I want to note that while I agree academically that content and markup are not the same.
<scott> does wcag really need to define what markup is though?
<bbailey> +1 to AC to keep "by an interface" at the very least
hdv: there are practical consequences but I think we need to keep the definition the same,.
<janina> Is it helpful to say 'content is what's rendered'?
<Zakim> kevin, you wanted to comment on separating markup and technology agnostic requirements
<hdv> s/that while I agree academically that content and markup are not the same/
kevin: re Heather's point. In assertions we may need a disassociation from markup as an idea to keep things tech agnostic.
<Jon_Avila> I think the key is "communicated to the user" and "information and sensory experience".
alastairc: I think it better we can keep our definitions tech agnostic, where we can. There will be some assertions that will need to be a bit more specific about things like markup, but for general discussions of content as a concept we can be agnostic
jkatherman: The markup isn't so much what we are worried about so much as the distinctioin.
jkatherman: we are not as worried about the markup in text and wording, more the content itself. When we talk to users we see different roles for content writer vs engineer. This distinction between content and markup better aligns with users
… when we talk with users there are diffierent roles for each.
<Zakim> GN, you wanted to say that in the past the term 'author' used to be used to refer to the responsible(s) of the content including its technical representation including markup
GreggVan: My concern is that things like submit buttons and such (i.e. things that don't carry a sensory experience) I think we should clarify that with "Sensory experience and interactions to conveyed"
<Jon_Avila> I agree to include interaction.
GN: Historically we have included the text, and interactions. I like the suggestion of referring to an "interaction"
alastairc: we still run the risk of confusion for those moving from WCAG 2 to 3.
<DuffJohnson> +1 to Gregg. We are in the WCAG context, so don't need "by an interface", IMO.
GreggVan: We might want to delete "by an interface" in favor of something like "by a page" to avoid questions of what is an interface.
alastairc: is it clear we are discussing something digital just from the context of it being WCAG?
<DuffJohnson> I'm not sure than an "interaction" is also "content"...
GreggVan: Yes, I think so. I don't think we need the word "interface", we can just discuss content
alastairc: the way we use content encompasses "everything" including interactive elements.
DuffJohnson: I agree with the current edits
<GreggVan> +1
<kirkwood> “interactions” is misused in this definition
<Jon_Avila> Sensory experience includes the audio and other experience
ljoakley: content is not interaction itself. Content does not actuate itself. I don't think including interactions and sensory interactions as "content". I don't like this defintion.
alastairc: we need a term for "everything on the page"
<kirkwood> interactions is not correct
ljoakley: I'd like to not carry the WCAG 2 defintion into 3. We need simplicity.
<bbailey> Can "includes code or markup that defines the content's structure, presentation" can be a note?
<Zakim> bbailey, you wanted to ask if "includes code or markup that defines the content's structure, presentation" can be a note?
bbailey: we can include a note saying that we are not changing the big picture idea
<DuffJohnson> Content: information to be conveyed to the user (even simpler, per Lori).
stevekerr: Are we already covering the idea of "everything" in the idea of web app or page, rather than needing "content"
<Jennie_Delisi> Content: information and experiences for use by humans.
GreggVan: We are missing the noun here. Should it be "that which conveys information"?
<hdv> +1 Rachael
<kirkwood> +1 to “interactive elements” or i”nteractive components.”
Rachael: I think information and sensory experiences are the noun. Content is the thing contained by something else. Is there a different term from "content" that allows us to get away from the WCAG 2/3 conflict
<kirkwood> +1
<kirkwood> Rachel
<Jon_Avila> information, sensory experience and interaction conveyed to the user
janina: Let's keep it simple. It may not be useful to worry about interactive vs. not. Its about what the content creator is trying to convey. It is what is rendered, not the rendering mechanism.
stevekerr: I'm wondering if we need to mention "decorative" content. WCAG 2 calls-out sensory experiences and decorative content.
<DuffJohnson> +1 to stevekerr. Perhaps consider 2 definitions, one for "substantive content" and the other for "decorative content"...
<Jon_Avila> for use by human may cause issues - does that mean not processed by machines like assistive technology.
alastairc: If we are splitting between informational vs decorative content we will note that 1/3 of all SC's reference content
<janina> +1 to Hidda that lm can consume content
hdv: why limit content to what human's experieince, what about LLM's consuming content?
<kirkwood> some of the cognitive accessibility protocols are to use technoloogy to simplify thus “humans” would be an issue
scott: I want to remind folks about the conversation about content from the 2.2. updates. The LLM doesn't pull the interface items.
scott: are we removing a defintion of markup? Isn't there already a understood defintion?
<bbailey> Wilco's reference to Understanding Reflow which distinguishes (text) content from web page: https://
alastairc: we only need to define markup if its used normatively and in a way that might be different from the defintion.
Jennie_Delisi: Ultimately humans experience all this content, as distinguished from computer to computer conversaiton.
<Zakim> alastairc, you wanted to comment on the humans bit
GreggVan: +1 to removing "the user" so we don't have to define who the user is. We can talk about conveying information to the user agent which will eventually be communicated to the user.]
joryc: I want to bring up training LLMs. They are consuming what interactions are available. Turn interactions into intent.
… LLMs do consume interactions as intent
<scott> i was using the LLM as an example - but my point was that people don't commonly think of the globally repeating parts of a page (its interface / navigations, etc) as the "content" of the page
<GreggVan> +1 and it is robots that feed LLMs and the robots do indeed interact
<Jon_Avila> What if it isn't provided by the author - but by something else.
GreggVan: We can define user in a wide variety of ways, but we get in trouble re-defining things. If we want to change the meaning we should add an additional modifier to "user".
GreggVan: Conetent may not actuall be from the author, it can be third party content.
<kirkwood> end-user is term we sometimes used in gov’t
<kirkwood> (tech dept)
julitrawe: our subgroup has noticed that there are some requirements that use "text content" as a way to be more specific. We need to define that clearly. This may be helpful.
<Jon_Avila> Experience is just as important as information
alastairc: I think we have gotten as far as we can on this. Can subgroups review instances of use of the idea of "content" and see if it would work. Report back on where it does work or not work.
<bbailey> Per julierawe , further down in the document , Text content: text and formatting including semantic or hierarchical structure that communicates a shared understanding of the overall content and meaning
joryc: We are talking about information. when we talk abotu content are we really talking about units of information?
joryc: I suggest using "information" vs content. Not all content is information. All information is content.
<janina> Seems music is content, but not information
Music is definitely information!
<janina> Something to discuss someday!
<bbailey> From https://
If I hear a Gb5 I get information in the form of tension that it wants to resolve to a C
alastair: the requirement should have the exception written into it.
<alastairc> "there is no way to carry out the function without doing it this way or fundamentally changing the functionality."
<Zakim> GN, you wanted to point out a typo
brucebailey: in 508 we have similar terms to "essential"
GreggVan: "Fundamentally change" the functionality could just mean adding lots more content. We need to define "essential" somewhere.
alastairc: this will be the place it is defined.
<kirkwood> wait… should we not be redefining vocabulary words as we said before?
GreggVan: Let's keep "unavoidable" in there as an alternative to consider.
<bbailey> 508 (and ADA) has concept of "fundamental alteration" , which is a an example of "essential" : https://
<hdv> +1 Rachael
Rachael: If it has to be an adverb maybe "critical to carrying out..."
<kirkwood> “essential function”? not using complete sentences in a defintion is a no no
hdv: I prefer "essential" to "unavoidable" to avoid confusion from WCAG 2 to 3.
<AWK> unable to be changed without fundamentally changing the functionality?
<Zakim> GreggVan, you wanted to say absolutely necessary or required in order to provide the same functionality
alastairc: in WCAG 2 it is usually in the content of "unless the thing is essential"
GreggVan: What about "absolutely necessary or required in order to provide the same functionality"
<Zakim> Rain, you wanted to clarify that it is about the purpose of the functionality
<Jon_Avila> if removed would alter the essence of the content or functionality.
Rain: Functionality... people often understand it as the interaction themselves, but we are using it in terms of ensuring that the purpose of the UI is met.
<ljoakley> +1 Rain
kirkwood: Can we agree to use complete sentences?
<bbailey> What is concern with wcag2 definition for essential? I get that it is weird for a definition to be a test.
<Zakim> bbailey, you wanted to ask for concern with wcag2 definition for essential ?
bbailey: I feel like "essential" works as a test.
<Zakim> kevin, you wanted to say that it isn't necessarily the case that it has to provide the same functionality
<bbailey> thank you julierawe for "drag and drop" example
<Zakim> GreggVan, you wanted to say "absolutely necessary or required in order to achieve the same result"
kevin: the use of "essential" is saying that there it is essential to the function, it can't be replicated another way.
<kirkwood> +1 to Gregg
GreggVan: 1) definitions should not be sentences, they should be a phrase that can be substituted for the word being defined. 2) Essential means that its not possible to replicate in another way 3) This is a good opportunity for a note of clarification
<bbailey> same result as what?
alastairc: I think most groups have used essential, please check how you have used it and if it needs to use the new essential definition.
alastairc: In response to Bruce asking "same result as what"... it depends on the context
alastairc: please take a look at other group's definitions and add comments
<alastairc> https://
<kirkwood> Greggs number 3) The sentence is a a good opportunity for a note clarification
alastairc: The doc to review is https://
alastairc: let's get comments in before the next meeting so we have time to resolve them
<alastairc> WCAG 3 preview: https://
alastairc: We need to make sure things are marked as exploration or developing in a way that reflects their state