Meeting Minutes
Wendy Reid: https://docs.google.com/document/d/1--4RetwmMJwNmO10lFZM77iTmWUpNhfdQWSbUsH2_T4/edit?tab=t.iikzz8wcu3g
Wendy Reid: So we're just carrying over from last week. I shared a link in the chat. Continuing the discussion around use cases and requirements, as promised, I cleaned up a new tab for the cleaned up version.
Of what we worked on last week.
where I took the kind of high-level user needs and authoring tool requirements and turned them into, I guess, like.
headings that we could then group
Use cases underneath
And it might help us, I realized as I was, as I was putting it all together, it's like kind of a helpful way also to frame.
the use cases and think about the different different possible use cases for each high level.
User need
But I'll let people get into the dock and take a look.
And for anyone who didn't,who wasn't here last week when we started this, the goal of this point is just to, think big.
And you know.
think about what…
The use cases are today for the use of authoring tools, the use…
Or even for as much of the future as we can possibly think about.
What are things that people can either need to be able to do right now or need to be able to do in the future?
And what does technology need to do to…
create those conditions.
Yep.
Sambhavi Chandrashekar: Wendy?
Since we are talking about AI generated content.
Could the use cases also cover user prompting in some way?
Wendy Reid: What do you mean?
Sambhavi Chandrashekar: I mean, we're just talking about…
Oh, sorry.
Okay, it's covered. Yeah, right, right.
Nicole Letoile: Instructional Designers & Learning and Development Teams
As an instructional designer, I need my authoring tool to provide real-time guidance on WCAG compliance as I build lessons so I don't discover issues during audit or deployment.
As an L&D professional, I need to audit and remediate existing content (courses, modules, assets) within the tool itself without exporting to external accessibility checkers.
As an L&D team, I need to enforce organizational accessibility standards across all authors so that we maintain compliance at scale.
Ned Zimmerman: I know this is already in ATAG, but I think it would be worth including in the…
Item under high level authoring tool requirements that authoring tools should be able to support languages and skips, provide means to translate or transform content into different languages and formats, or support external means to provide these features. I think it's probably worth highlighting without any loss of accessibility function.
of accessibility within the content, because I know that there's a criteria under ATAG around transformations, not like accessibility information cannot be lost during transformations, content transformations. So I think it's worth.
explicitly just adding that to the capability for transforming and translating.
Nicole Letoile: sorry, I haven't been as,involved, so I apologize for anything that's been said, or if I'm repeating, or,taking us on different tangents, but I did add some things in the chat, just from the learning and experience.
in the learning and development instructional design perspective of use cases.and then I thought I heard Dr. Sam starting to talk a little bit too about, um.
maybe I misunderstood. But there's the the vibe coding which is happening throughout the learning experience sectors and the.
the inexperience of accessibility knowledge, or, knowing the content standards, so that coming out with.
products and tools that are still not accessible, and now we're doing them at a faster pace, so…
I figured I'd prompt that.
Wendy Reid: Yeah, here. Let me share my screen to show you where I'm at.
So in here, this is like a good, it's also, as I was putting these together, I realized that this is like a really great way for us to kind of test some of these high level.
needs and requirements, because if we either… if we can fit a use case into them, good news, that's a good… that's a good, high need. If we can't, or we find that there's, situations where they don't quite fit.
Then it's either, either we need to change it or create new ones that…
fit those cases. Though I think, looking at,the ones you shared, Nicole,I just added this one here, because we have this one,called Users Must Have Complete Control Over the Content They Produce or Edit While Being Supported with Accessibility Guidance and Features to Make Informed Decisions.
Nicole Letoile: Perfect. Yeah. Thank you. Okay.
Great.
Okay, so each one of these are big, like high level. And then we're putting more detailed things. I got it. Okay.
Wendy Reid: Yeah.
Just felt like a good way to… because we were talking about, I think we can get really… obviously, use cases can get really, really specific. I think one thing that'll happen…
and actually, I think it would be really good for people to be… be as specific as you want. you can talk about, stuff that you do every single day.
and then we can take…
the large number and then try and kind of collapse them down into.
You know.
more focused.
cross-role use cases for the actual…
document, because I think, in a lot of cases, we're talking about different roles, and those roles are important, but there's going to be a lot of things where it's, the content creator and the instructional designer have very similar needs for.
You know, the vibe coder, and they…
And a teacher, or something.
Lisa Liskovoi: Just a clarification question. In the authoring tool requirements section where it says authoring tools must be able to support accessibility features, is that for the author or for the generated content?
Wendy Reid: That's a good question.
Thanks for the generated content, because for the author, that would be a WCAG requirement.
This is interesting. There's been a lot of discussions in AgWig lately about.
about authoring tools accidentally, because there's been a whole,back and forth about third-party content, or user-generated content.and so this, like…
And part of it is just the…
complicated relationship between where WCAG ends and ATAG starts.
And so I think if we're thinking in the ATAG context, we have to say it's the accessibility features of the created content.
Or of the, authoring interface itself.
When, when creating content?
Lisa Liskovoi: Yeah, so I interpreted that as the accessibility features of the authoring UI, but I am wondering if that's not what…
Maybe we want to be more clear here.
Wendy Reid: Yeah, I think so.
Lisa Liskovoi: because I think the use case here is that I'm just moving stuff from our discussion thread as a content creator. I need to be able to create new content.
I think what they mean here is that the tool needs to support the author and be accessible to the author. But.
I'm welcoming other reads on this.
Miriam Fukushima: Yeah, I also understood it as the features for the generated content. So if I author a video, I need to be able to put subtitles in audio description and so on. So the authoring tool needs to support those.
technologies and give me access to.
put the content into these features so that the end result, the content I create, is…
Accessible.
Sambhavi Chandrashekar: But if we look at,8.3.6.
Then there are some…
ATAG 2.0.
requirements for the author being able to use the tool.
in an accessible way, have settings, that part is also there. So that angle is also there.
Miriam Fukushima: Yep.
Yep.
Lisa Liskovoi: Then maybe what I can do is add another section that says authoring tools need to be accessible to the author.
Miriam Fukushima: Yeah, I think those are two different use cases.
Wendy Reid: I wonder too if this is also an opportunity now for us to clean up.
and maybe much more clearly define… so, remember, we're, obviously ATAG 2 is something we work from, but, the next version of ATAG is an opportunity for us to, like.
Rethink.
the relationship and, the structure and things like that. we don't, like… we can use ATAG2 as a…
As a basis, but we don't have to…
We don't have to, like…
Cleanly make it, you know, everything the same, plus additional stuff.
because I do think, it was interesting this week in the discussion about user-generated content.
Because…
There was like…
Almost an argument being made for oh, well, could WCAG.
have all the requirements necessary.
For user generated content.
And…
I was well, kind of. And I think one of the big things that's changed is how much is now possible in the creation of content.
but at the same time, how little control platforms still have over what happens during the content creation process in the sense of… especially in the sense of accessibility, because…
Even though now you can add.
You can prompt the user to make their content more accessible. You can even remediate their content to be more accessible. There may be reasons why that…
should not happen or cannot happen, or the author does not want it to happen.
And you need to respect those.
And I think that's maybe where, we come in to say, there are cases where we need to kind of defend this.
abilities.
Miriam Fukushima: Yeah, maybe we need to,
like, in the very beginning, we discussed the notes,of the beginning of ATAG, so maybe we need to expand those really to define
what ATAG is, but also to define what it is not, and how… specifically, name how it differentiates from WCAG.
And that there's the responsibility of the author during creation and hence the tools and the developers that provide those tools have a responsibility too. But in the end, that's separate from the finished product, the content.
And,whereas then the author is still responsible for their own content, but the tool is out of the loop once the… it's declared finished, or something like that.
Wendy Reid:
Sambhavi Chandrashekar: In my experience in filling the 2.0 even recently.
A.1 to A.4 completely corresponds to WCAG. We could take everything from the WCAG and populate it here. And in fact, we found two or three things we should be doing in the WCAG and went back and caught the teams and
like customizable keyboard shortcuts, some things we hadn't done, documentation, linking, and things like that. So when it comes to BE dot, probably we could think of some restructuring like you were saying, Wendy.
That we need an exactly map from 2.0 to 3.0, all the sections.
But ADOT will remain the same, I think.
More or less.
Because…
It's about the tool. And I don't know if when we say authoring tool, we mean a tool that's dedicated to authoring content.
Whereas situations like the LMS, it's a small part of the LMS.
I mean, it's just a point that we have to keep somewhere and think or discuss later. The other thing is…
in the LMS, there are 4 different tools, the HTML editor, the media editor, the interactive content editor, quiz editor, and then we have the AI-based.
Uh part as well.
So, do we have 5 ATAGs, or do we have 1 ATAG, and…
make a superset of all these. If there's a failure, we mark it.
Otherwise, we ignore if it doesn't apply to one of those tools. That's another practical difficulty.
Wendy Reid: I mean, it's…
There's maybe another, I guess, an additional class of use cases and requirements because of just the nature of any of the accessibility guidelines specifications.
And that is around the reporting.
And…
I'm going through this with WCAG 3 right now, reporting, testing, and policy.
They're all kind of… just by nature of the types of specs that we… these are, they have these…
And they're not even necessarily sections within the spec themselves. They might be supplemental content, but they are part of the consideration.
For how you design the requirements.
So maybe I will add that section at the bottom.
I could spell.
All right.
'Cause, Sambhavi, what you mean, I think, is…
If I have a platform, a product that I maintain, and I have a learning management system that has individual tools for creating quizzes for.
I'm creating written content videos and these are all essentially separate tools, right? Maybe they're maintained differently. They have different pages. Do you test them and assess them in totality, or do you test them and assess them on their individual merit?
Sambhavi Chandrashekar: We we do one single testing for the Lms.
by way of external audit as well. I'm not talking about the product process which has like accessibility integrated across the whole process.
My logic is that if we sell a product separately, we create a report separately for that, even in ACR, because it ultimately has a.
sale implication in sales, right?
So, where we don't sell products separately, we have a consolidated ACR, like the mobile app and the core LMS will be in one ACR. Likewise, it's all in one single core.
So we'll have one single ATAG is what we decided.
Because we had to create the 2026 ATAG for the customer.
which we did. And interestingly, Remy and I had a very, very interesting conversation.
chat,in a separate session.
And the 11.8, which, Lisa, you pointed out in EN301549, I'm just…
Sambhavi Chandrashekar: EN 301 549 Section 11.8
Sambhavi Chandrashekar: Adding that link here.
I was asking last week if ATAG is mentioned in any regulation.
It's not mentioned in Section 508.
But in EN301549, and therefore in the Canadian EN301549, it is mentioned under 11.8.
And all of 11 section is about software.
As a SaaS company, we ignore the whole software section because that's about downloadable software.
So that's the reason why we didn't touch this at all.
So, yeah then that's why we decided to have a separate ATAG and not add it into the ACR. In our ACR, we have an EN 3.1509 section.
Under which Section 11 of EN 301, we just ignore that, we say not applicable.
Because ours is not a software in the real sense of the word, but it's a SaaS product.
Miriam Fukushima: Yeah, I think it's important to also consider that a lot of authoring tools are just a mix of or like hybrid versions of different things. Like we have a lot of tools online and they're not technically software, but we could also have them locally as software.
And,
to the… one of the main points we discussed very early on was to make ATAG more accessible, in itself for deployment, and.
As that, we should kind of give,
best practice or guidelines on how to break down if you have a software suit that has many different products, but they are all under one umbrella, or on one core,
platform that.
you… it should be possible to see those separate and break it down to make it more applicable, and not, like… if it's just,one big thing that.
developers just never tackle it because they see no end to it, so that you can say, Okay, if you can distinguish one workflow, maybe in a software and go workflow based and say, Okay.
These parts of the core platform there, the workflow starts, but then I'm only in this part, or in this part software. And, the rest I can ignore for now, but fix this workflow.
Wendy Reid: something that's,happening within WCAG 3, which is another thing that we might consider. I keep an eye on that work, mostly because I'm in it, but also because I think.
We'll want to take a lot of it for…
the ATAG 3. And that is, so there is this discussion of introducing the concept of paths as testable units.
and a path is essentially, just a subset of…
Of a product. So, you could… you would… it's kind of like how, if you do… if you've ever done, a large…
website or app assessment, you very rarely, test the whole thing in one go. You test your sign-in, your sign-in experience, you test your registration experience, your home page, your purchase path.
and breaking things down that way, which also makes things a lot more…
digestible, not only internally, but also, it may also make things more digestible in terms of reporting, because you can say the sign-in experience performs like this, and the…
And I think we can take the exact same concept, like that workflow-based thing, where it's oh, I'm creating a text document. Here's the accessibility of our text document creation process. Here's our video product creation process.
Sambhavi Chandrashekar: I'm I'm all for it. I'm totally endorsing that because I've always.
advocated for workflow accessibility.
There's no point in checking the nook and corner of every page. It doesn't make business sense either.
Critical workflows, that's when, when you test with people.
it becomes easier to do workflow testing. And one reason why I don't like.
audits in the traditional sense, although we have to do it page-wise for compliance purposes to make our ACR.
But path or workflow is the best way. I don't know if we can bring that into this.
New version.
Wendy Reid: I think we can. Yeah. I mean, especially if it's made part of WCAG 3, we can…
Sambhavi Chandrashekar: If it's good and would yeah, please do it.
I mean, we cannot ignore the page wise, but we could prioritize this.
Wendy Reid: Well, one discussion within the working group right now about paths is whether
we… or… I'm actually in favor of doing it either way. Do you define a path as a series of pages, or do you define a path as a series of components?
And I think my position on it is that.
it can be either, and it should be up to the, the person doing the testing, because on some products, it may make sense to say it's a series of pages, and on other products, it may make sense to say it's a series of components.
Sambhavi Chandrashekar: Yep.
Correct.
I agree.
That's a totally unrelated question. In the LMS, I mean, this is all practical difficulties I'm facing. In the LMS, we have a Lumi.
product, which is AI.
I mean, adding AI to the teaching learning experience. So does that have to be reported under 2.0 at all?
Because 3.0 is coming for AI generated content.
Wendy Reid: I mean, I think the thing now is that you can report on your AI interfaces.
but you do it in the sense of, is the interface accessible in meeting ATAG? Is the content that it produces. Meeting ATAG.like…
Us adding AI, there's some things we obviously have to add, but I think a lot of it is actually just being…
you can… you could read AI into a lot of these things, but we're just making it more explicit.
Sambhavi Chandrashekar: Yep.
Even 2.0, like you rightly said, has so much of auto-generated content.
Abhinav Kannan: Would it help to identify what might be significantly different?
When it comes to auto-generated content in 3.2 versus 2.2.
Would there be…
Any points that might be…
Very significantly different, and how would that be applicable to the software?
Sambhavi Chandrashekar: I would think that in 2.0, what would have been meant, I don't know, could be, scripts generating content kind of thing, where it's still programmatically generated, and not…
Through AI.
Wendy Reid: I think a big one, a big difference between automatically generating, like Simbaba was saying.
Before, it was scripted. And so you could trace back the issue. If something went wrong, you could trace back why something went wrong.
That is more difficult now because AI doesn't always leave logs behind or doesn't always explain why it's doing what it's doing.
it also, something I've noticed in…
that is, an interesting problem is that.
So, if you're using…
let's say, like the Canvas feature in something like Gemini or whatever, and it's generating a document for you, and you…
as a sighted user, I can see that there's hierarchy, and, you know, the headings are different sizes, there's different…
you know, organization to the content.
But I also have no control over that organization of the content, and I don't have ways to even check easily, oh, is this…
An H2 or an H3? Can I change that? I want to change the layout of the content.
and it can also sometimes, because you have to control everything through the, essentially, the chat interface.
it can…
you can actually cause problems by doing that.Wayne's not here today, but he emailed me about a use case he wanted to, raise around.
Challenges where tools will.
change. if you gave it something with heading, with proper heading structure, it might change it.
In ways that you're not clear about, if you thought you had a good idea of what the content looked like when you went in, but it changed it, and now you don't know what's going on.
You don't have that level of…
Editability and it might be for us to say like that actually has to be a thing. You have to introduce this as a feature. For it to be an accessible authoring interface.
Sambhavi Chandrashekar: If we don't do that, we have the experience of a voice, a speech input user.
Like, I'm editing using speech input, say Dragon or something. I'm looking at one's pillow, I want to correct it, but I can't use my hands to go and do it. It's similar to that, except the constraint is on the other side.
Sambhavi Chandrashekar: I don't know how that helps, but it just came to my mind that disability is playing out in different ways.
Lisa Liskovoi: Just to note that I've moved all the use cases from the discussion thread over to this document, although obviously feel free to move them around.
Nicole Letoile: So in the case of what we're doing, we don't want to be a specific right. We don't want to call out any. In this case we're not actually naming.
any specific tools or software. Is that correct?
Wendy Reid: Yeah.
I mean, this is not the final version. So if it's one of those things where it's like trying to describe this is doing your head in and it's just easier to say like InDesign, then it's fine because we can edit it later.
Nicole Letoile: Okay. Yeah. This is a specific to Grackle Docs. When you make copies of Google Docs after you've made changes with Grackle, it doesn't copy to the copy, and teachers use copies all the time. So I wanted to call that out.
Okay.
Thanks.
Wendy Reid: Yeah, we'll find a better way to describe it.
I don't want to get ahead of ourselves either. But as we're thinking about these use cases, this has come up a few times already, too, is I think probably the next thing we'll have to do is think about definitions.
And terminology, because we've already run it.
ATAG 2 has some terms that need to be amended or updated, and we also probably need to start introducing new terms.
with, I think one example is, ATAG2 does use the word prompting, or, prompt, but that has now taken on a completely new meaning.
I'll have to think about it. Abhinav, go ahead.
Abhinav Kannan: Yeah,I did see some chatter on…
So there has been… there has obviously been a lot of talk on…
Prompting strategies.
What… so there's a rule-based prompting, an examples-based prompting.
some of my co-authors and I have written a paper on prompting strategies, but there has been some chatter on LinkedIn on prompting strategies for generating accessible.
For accessibility, for generating tagged content.
So, I can look into…
how these prompting strategies are successful versus how much they're failing.
And probably revert.
back with what my findings are, and we can probably… if that might be helpful, we can have a little bit of a discussion on that.
Nicole Letoile: I wrote about that. https://www.letoile-education.com/blog/83-rounds-what-ai-assisted-accessibility-testing-actually-looks-like
Abhinav Kannan: I'll go down that rabbit hole for a bit.
Miriam Fukushima: I was just kind of wondering, because here in the document we have some overlapping things on how we should sort this, because, for example, in user need cases, very specific ones that.
For example, as a teacher, I need to…
Create content to be.
without barriers of the interface, but that is more generally already in.
Authoring tools must be accessible to the author.
How do we go about, sorting these kind of things?
Wendy Reid: I think we'll have to do another pass.
Once we have… I want to, do, the maximalist and then go back over it and, clean up the…
Clean things up.
Sambabi, go ahead.
Sambhavi Chandrashekar: Yeah, Lisa, correct me if I'm wrong, but in the Canadian standards work, we usually.
borrow these definitions if it's available anywhere else, and then only if it's inevitable, we coin definitions, right, Lisa?
Lisa Liskovoi: That's a good… yeah.
As long as they're available openly, yeah. So no ISO definitions if we can avoid them.
Sambhavi Chandrashekar: Right.
So should we look for AI related?
standards, open standards.
Wendy Reid: we can try. I mean, probably one of the first places to look right now is… I don't know how many AI-related standards there are currently, but,
Probably a safe place to start would be some of the legal.
things that have already come out, like the AI… EU AI,legislation, Canadian… even if it's… Canadian… Canadian one has not… has passed? Has not passed. I cannot remember. It was in the news.
We can look at different countries and just see, I think, the kind of language and terminology they're using.
And then maybe look at… I know IETF is doing a lot of work with AI stuff, so if we can make sure that we're using similar terminology to them as well, it probably wouldn't be a bad thing.
Sambhavi Chandrashekar: This sounds like a mini research project. Do we crowdsource that or does somebody get assigned?
Wendy Reid: Let's do research.
Sambhavi Chandrashekar: The Canadian standard is there, published.
Sambhavi Chandrashekar: CAN-ASC-6.2:2025- Accessible and Equitable Artificial Intelligence Systems
Wendy Reid: Thank you.
Sambhavi Chandrashekar: I'll add it to the page.
Wendy Reid: I have so much content already. Good job, everyone.
Lisa Liskovoi: Do we want to make a list of the definitions that we want to compare across these documents? Because I'm thinking there's a lot of definitions in all of these that we're not going to use, so bringing everything over is probably overkill.
Wendy Reid: Yeah.
Prompting for sure.
Miriam Fukushima: We won't…
We wanted to define the term content.
Because we had different versions, like the content that could be generated while,
content creation, for example, from an AI as an output to like as a resource for creating the content.
but also the end product, the content that is created and is the finished product to present to the public.
So the term content you wanted to define again?
Wendy Reid: Yeah, I'll have to come up with that.
Oh yeah, the other one was,
We were talking about this a few weeks ago.
Deterministic.
They're talking about the difference between like.
essentially deterministic accessibility.
like testing in general versus.
Sambhavi Chandrashekar: Probabilistic.
Wendy Reid: We're going to have to come up with something.
Sambhavi Chandrashekar: It's not that it's working.
Ned Zimmerman: I've usually just seen it as deterministic versus non deterministic, but I know defining in a negative is not as ideal.
Abhinav Kannan: Yeah, deterministic versus non-deterministic is what we commonly hear, but non-deterministic might not be the most suitable.
Lisa Liskovoi: We'll need to define at least one of those.
Wendy Reid:
Yeah, I,
I know at least in standard space, it's why we have normative and informative and not just normative and non-normative because we have to be precise. If we're not precise, someone's going to get mad at us.
Probabilistic or something along those lines is probably.
Or another turn, because basically what we're, I think, based on the conversation we were having then, the discussion really was.
There, you know, in… back… back in the day,any… any kind of accessibility assessment that was done with tools was deterministic, because it followed very strict.
You know, contrast does not meet this because you're comparing this to this, and you have an exact number, and it's very — which severely limited accessibility testing because it could only be deterministic.
Tools that.
can do things that are not done in a deterministic fashion. The problem being that it doesn't mean they consistently do the same, achieve the same results.
So how do we differentiate the deterministic from the non-deterministic?
Abhinav Kannan: probabilistic is probabilistically vague ...
Wendy Reid, Lisa Liskovoi, Sambhavi Chandrashekar:😂
Sambhavi Chandrashekar: Usually we use probabilistic for user testing sort of things.
Not exactly where, like you rightly said, it doesn't.
It's not a zero or one.
Ned Zimmerman: Stochastic?
Sambhavi Chandrashekar:👍
Abhinav Kannan:💯
Sambhavi Chandrashekar: Cognitive, for example.
Sambhavi Chandrashekar: They will all be probabilistic.
I mean, I was… I was part joking, but…
It is a term we have used.
Wendy Reid: If we use stochastic, does that mean we get to refer to my favorite AI paper ever, the stochastic parrot?
Sambhavi Chandrashekar: Oh, yes. Oh, that is, oh, that is some of the parrots, Stokakic parrot or something.
Did I hear sarcastic parrot?
Wendy Reid: They're the stochastic pairing.
MJ - Mary Ann Jawili: Hey, this is MJ.
I was wondering if it would be… if we should,
Define authoring tool, or is that pretty clear?
Wendy Reid: we have, an attempt…
definition?
that we coined a couple of weeks ago.
Or I say weeks, more like months at this point.
Thanks to Lisa asking me for it a few weeks ago, I remembered where I put it.
Because I realized I had forgotten.
It's this one
You don't want to cut these.
Miriam Fukushima: We also had the different roles in the authoring workflow.
that we kind of defined in earlier sessions. I think the developer of the authoring tool, the user of the authoring tool, and then the consumer of the content.
And that sometimes even different.
users and roles can be under the umbrella of user of the authoring tool.
Wendy Reid: I have to go back in the note to pull out the full list. I know it's somewhere.
Abhinav Kannan: I'll just quickly add,it's… would the… we are looking at author, developer, consumer, editor.
Can we also add the platform that would host?
the platform on which the content is being shared. For example, we might create… somebody might create an accessible document. They might share it on WhatsApp, they might share it on Facebook, they might share it on Instagram.
Whether these platforms mess with the accessibility tagging or keep them intact, that could also come into play.
Wendy Reid: I'm trying to think of the word that I want to use for this.
It's buried.
That's a good one, Abhinav.coming from publishing, reading systems,play an interestingly chaotic role in…
EPUB content.
Miriam Fukushima: We also wanted to redefine the term of user interface, that it's not static anymore, but it can contain,content itself,especially in.
AI form and that resources and.
that kind of stuff is part could be part of the authoring process and of the user interface, and that the user interface is not the fixed.
line, or taskbar buttons anymore, but can be can come in various forms nowadays.
israel: Yeah, so,just more like a question for things to define contents and the clarity around the states of the content. When you say finished product is the…
finished document itself or the cost to be published? And my question is right to when an author uploads a doc, for example, and the platform converts to a more accessible way for XML to be.
Then finally publish in a different way like HTML. So is the upload of documents like doc PDFs.
Consider a finished product as well
Wendy Reid: You've nailed exactly why we probably have to come up with definitions for this, because there's essentially, like…
Uploaded content to a platform. That might transform us.
we'll come up with much better terminology for this, but yeah, that is a big problem, is
that is there such thing as a finished product? Is there a final version? or is there just a version that exists.
israel: A workable version. Yeah.
Wendy Reid: You know, a version not actively being edited.
israel: Yeah, yeah.
Yeah. Yeah. Because I'm in the I work with the publishing power form and have the issue.
a lot. So, user can upload a PDF, a doc, or now a XML file, so… and then everything is merged into a new format to be published to users. So, that's my… that's why I'm asking that.
And right to the rows here, the editor, I agree. I think it's a great idea to add editor here.
But maybe you can even create for editors because in some platforms they have.
different roles for editor, for copy editor, production editor,copy setting editor, so just to put, editors, because it's different roles in the same one. Sometimes the actor is also a reviewer too.
Yeah, so sometimes the editor is also a reviewer. So editor can be different roles, all of them in single person, but different specific ones here.
Lisa Liskovoi: Just on the editor comment, I think we probably want to go pretty broad and say something anybody who modifies content other than the original content creator, so that would include.
you know, any kind of… any of those editor roles would fall into the broad category of modifying content once it's been sort of initially created.
Wendy Reid: Interestingly, in EPUB world, we had a lot of back and forth over.
what to call…
what's it called? The kind of primary driver of the content? Because very rarely is, EPUB, especially in a publishing case, is it one person? It's, you know, a team of people.
Possibly even multiple companies. And we ended up landing on EPUB creator. But over the years, we had referred to them interchangeably as author, content creator. We had a lot of terms flying around for a while before we landed on EPUB creator. And that might be something we'd consider too, is document creator.
Or, we're using Content Creator in a few places, because.
arguably content, I mean, like we have to define, there's a bunch, content.
It can be videos, it can be audio, it can be text, it can be…
a video game. It can be all sorts of things.
Abhinav Kannan: And that's… that's great to see, because I've… I'm struggling with this in my own work as well, on… on trying to define content author, content creator, designer, all these terms are a little bit interchangeable in my…
In my own writing, so it's good to see that I'm not the only one struggling with this.
Sorry, the others wanted to speak, I just jumped in.
Lisa Liskovoi: So one thing this is making me think of is that we might want to consider defining the action rather than the role, because, as you were talking about it, Wendy, I was thinking, well, the content author could be their own editor also. They could be both the content creator. And then.
we can think about them even editing content, but potentially in a different format, or sort of downstream enough that it's a different process than the content creation process, but it's the same person. So rather than.
defining author, we can define authoring or editing.
Nicole Letoile: What about someone who needs to verify that tools support accessibility checking and compliance reporting- do we need that role as well
Wendy Reid: I kind of like that.
Miriam Fukushima: Yeah, I also just wanted to add, for example, we have a content management system. And the output is basically websites. So that's also kind of a different thing. But we also have, for example, tools that.
are just the back end to other tools that then users use. For example.
Like,similar to quiz… to quizzes, if you have the,
the back end that makes people put in the quiz questions and the answers and materials to study. But then the actual tool.
that gets used is the quiz tool then by students. So sometimes it's like a chain of authoring tools that work together. And I can be a teacher that inputs the questions, but I can also be the user when I use it with my class in front of the class.
And,same thing, I can be the developer of the backend of the whole thing, but I can also be the developer of the frontend, and I can be the tester that uses it for testing, obviously. And,
So yeah, they're very different scenarios.
Wendy Reid:
Yeah, I like this idea of framing, or maybe we have both, at least for now.
We are over time. Israel, I'll get the last word and then I'll wrap things up.
israel: Yeah, just quick here. I like the idea from Lisa about the actions over the rows here. And I just put a question here. Okay, I'm annoying people about metadata accessibility here. So where it goes, the metadata accessibility is authoring, editing.
Or a different action role. More like question for that because sometimes neither do these two people are doing this work for adding accessibility metadata to the information.
Ready for the goal.
Nicole Letoile: So for the tester, more specifically, the role verifies the VPAT and ACR with user testing to ensure accessibility claims are accurate.
Lisa Liskovoi: I wonder if we should add maintaining to the definitions
Wendy Reid: This has been great. Feel free to fill things in in your spare time.
Otherwise, we will pick this back up next week and probably think more about.
Things like definitions and research. So have that in the back of your minds as well. But have a great rest of your day and weekend, and we'll talk to you next week.