Meeting Minutes
Wendy Reid: I think for today, we were just gonna continue the discussion about.
Definitions and audiences and.
Who we're aiming this at.
I'm just refreshing my own memory.
I did some cleanup of the authoring tool types and functions. I'll share my screen
So I realized as I was looking at the list that
There's kind of groupings of tools.
And so like as we initially had these kind of
All the sub-bullets were kind of already were listed on their own. And I was, as I was looking at them, I was, wait, there's like there's themes here. And so like I've done like web content creation and editing tools, like things like WYSIWYG interfaces
Content management platforms.
Educational content, code creation, marketing.
Jutta Treviranus: Yeah, after last week's meeting, I was thinking that it might be good to create something about the forms of interaction or the design patterns for the UX. So we talked about a single prompt.
But then there's also the back and forth, the conversation, or what we the ATAG form of prompt, or nudging, or influencing, or guidance, or whatever we wanted to choose as a term. But there's
certainly the AI, and even without AI, there seems to be more than a single,
interaction that there's like back and forth, and there's.
A.
like some guidance. It's accepted. Then there's a multiple additional.
things that can be done following the guidance, etc.
So it can be much more interactive than I think it was back when we were looking at it, or there's more forms of interaction.
Wendy Reid: Yeah, we almost have,
gone from
like a static interface to a more
static with some interruption to a much more dynamic interface, I guess.
Jutta Treviranus: Yeah, and there were the guidance, but then there were also things where there were certain patterns, even back then, where you couldn't proceed. And we actually discussed that at length because invalid HTML in a number of tools, you couldn't proceed.
But, it was decided that it was never going to we weren't going to
delve into that, but we made one of the I think there was, at some point, several authoring tools which did that, that you couldn't the the image would not be
added, unless you actually put the alt text, etc.
Wendy Reid: Trying to think of head off.
Jutta Treviranus: So I'm thinking, yeah possibly definitions, but also some discussion as to what
what we want to include.
Wendy Reid: I'm ecstatic. I'm trying to think.
Jutta Treviranus: I think one element is whether there is it's how many
Yeah, whether how many steps there are to it. So whether there's a prompt response or nudge response, or whatever. And then.
Does it,
Yeah.
I think it's not necessarily static, but
Jutta Treviranus: That there's multiple steps to it.
Wendy Reid: I was trying to think of, I think what I have in mind is, there's almost different levels of.
I was using this word last week intervention, right?, if I'm
Jutta Treviranus: Yeah, right.
Wendy Reid: If I'm in a
developer environment, or if I'm in, Word.
I and I'm typing, and the squiggles appear, right? The squiggled underlines. That is a signal that something is wrong, but I don't have to do any, it's not interfering, it's not stopping me from doing what I'm doing. I'm aware of it, but I'm not
And of course, like.
There's that really low level of intervention. Then there's, of course, you go to add something and say a modal appears and it's, hey, you're doing this, we need you to do, we need this.
You need to add this or whatever. Then, of course, there's the next level, which is, say, I'm trying to save the file or I'm trying to, in a code environment, I've been spending a lot of time in Visual Studio this week.
It's, I go to
commit that work, or I go to build it, and there's errors that appear. And some of them are you have to fix them before anything works.
Jutta Treviranus: Yes.
Ned Zimmerman: There's also a, I was just gonna say, I think there's also like a
There's also something that you were getting at, I think, Uta, which is the
Conversational interface that can be
I'm just looking at your Google Docs share, Wendy, and thinking
the chat that you can pull up within a Google Doc, whether it's in comments on a particular review or.
I don't know if you can actually get a full
Can you get a full — yeah, you just get a full comments thread, and then you can see where the comments are attached. But those sorts of interfaces already existed.
in earlier forms of these tools, but I think I saw a demonstration which I might have linked to or dropped in the chat in an earlier meeting.
of someone's approach to reviewing content for accessibility, where they basically pull up a Claude session in their browser using the Claude plugin.
and have a sort of ongoing conversation with the agent.
in the context of the webpage that they're reviewing. And so, yeah, those things that are
engaging the interface that are not part of the core document authoring interface, but are sort of attached to it, or I almost think of layers on top of it. I don't know if that's a good way of thinking about it, but
sort of similar to your levels of intervention, there's layers of the interface that interact in different ways.
Jutta Treviranus: Yeah, and then you were also intimating degrees of —.
Like, how strict or how
how much choice the author has.
Ned Zimmerman: Flexibility or rigidity or something like that.
Jutta Treviranus: Yeah, exactly.
Wendy Reid: Yeah, I like the idea of like layers too. Like there's, there's.
Jutta Treviranus: Yeah, the default authoring layer, the ability to add additional.
assistance, et cetera. Yeah.
Ned Zimmerman: I mean, it's also interesting to think about, sorry, go ahead.
Jutta Treviranus: No, go ahead. Go ahead. No, you go ahead.
Ned Zimmerman: I was just gonna say, it's also interesting to think about what I was just talking about in the sense of the
The chat interface attached to
The authoring view.
Sort of like the side conversation.
which doesn't necessarily also need to just be talking about AI tools.
Right, because there are so many more multi-user authoring environments than there were in 2015, where you can have real-time collaboration, including people participating via audio or video.
And so, I mean, I think this is an area where obviously the
impact of AI is central to what we're trying to look at here, but that's also something that the existing version of ATAG doesn't necessarily talk about in depth, that is more prominent now and is more
common, so I think that's another interesting piece to get into here. I mean, I was just looking at some of the
Tools you list here, Wendy, like Figma, I mean, Figma, like anything where you see the cursors of other users moving around
You know, comments appearing in real time those sorts of things are important to include, and it's people as well as
AI tools that have that.
Jutta Treviranus: Yeah, and that gets into the whole area of accountability and responsibility, because one of the biggest discussions back then was social media, where the contributions by multiple authors.
Ned Zimmerman: Impact.
Jutta Treviranus: Outside of the organization, who's responsible?
And that, of course, was WCAG discussion as well.
Wendy Reid: I was thinking, I've been doing a lot of AI LLM stuff this week for work, and something that I ran into a couple of times that I realized was kind of interesting is, like.
I think it's an interesting contrast to, say, a group of people working in the same Google Doc.
is
What a group of people are working in the same Google Doc.
for the most part, you're aware of the changes happening., you can see them, right?, Google has this whole
uh UI, UX, for you see the you see highlighting, you see different colored cursors, you see someone's name attached.
And so if you're talking to a group of people, you know, you're in a Zoom call, you're in the same Google Doc.
You're actually aware of the changes kind of in real time as they're happening for the most part.
On the flip side, though, I realized this when I was working this week, is that,
I'm not a very good AI-driven developer, and so I often make small changes myself., if I see a bit of wording that isn't quite right, or
like, I'll just make the change myself in in the file. And then when I go to do the next prompt in the conversation.
I realized a couple times that it was unaware that I was making those changes. It was blind to those changes.
And of course, would then sometimes trip on them, because it was, oh, this line in this file changed, I gotta check that,
it was completely unaware until I interacted again, which is a really interesting, it makes sense, of course, why would it be aware? But
Ned Zimmerman: Yeah. Okay.
Yeah, I mean, it's sort of like the new age version of you're in a code editing environment and you go into your system, file system, delete something, and then the window that you had opening editing that file is suddenly, oh, this isn't here anymore. What? You know, but I mean, that's interesting because in some ways that points to.
I hesitate to frame it this way, but
the accessibility of the authoring tool for the AI part of it., how do the AI how does an AI integration get signals that it needs to update what it's working on, or be aware of changes to it?
And that, I mean, I didn't know that that was, I'd never run into that. So that's very interesting.
Jutta Treviranus: Actually, I just finished reviewing a thesis where that was the whole premise of it, that the justification for accessible authoring was so that it would be accessible to the AI.
And they propose that that should be the next justification.
Ned Zimmerman: Oh, no.
Jutta Treviranus: Yeah, I know.
Wendy Reid: I mean, it does it's weird, it's awful to think, I had this thought a couple of months ago, I've been trying to put it into a into a blog post of some kind, that
I am there's especially with recent announcements from some of the large AI companies around oh accessibility is good for us, too, is that most people will be driven to make their websites and their products accessible, not for humans, but for
for AI.
And it kind of makes sense because I realized in doing work recently with — actually, weirdly, a lot of Google Docs work with someone who is a screen reader user, that knowing that things are changing thing is basically an exclusively visual affordance.
in Google Docs, just as for example.
Unless your screen reader is literally on the thing, your screen reader focuses on the thing that's being changed, you do not know things are changing around you until you go back to them.
And so it is kind of the same problem where it's, even if you're in the same file, you don't get
You don't get that experience.
Jutta Treviranus: Yep.
Well, back then, the argument was regarding search engines, and digitization, and metadata, and specifically related to ARIA, etc.
Jutta Treviranus: So very similar. It was machine readability, was, was enhanced with accessibility.
Wendy Reid: I was just thinking about the layers concept.
The authoring layer.
I don't know what to call
Ned Zimmerman: I almost wonder if
Wendy Reid: Oh, go ahead, Nick.
Ned Zimmerman: I was going to say something like.
the
The review layer.
so something like that would capture both things
automatic spelling grammar check indicators, but also comments highlighting.
annotations of content.
Suggestions in editing interfaces that let you suggest a change without committing it.
Jutta Treviranus: Yeah, and that brings up whether it's before authoring or post-authoring, right? I mean, it's
Jutta Treviranus: Pre-production, post-production, or whatever.
Ned Zimmerman: Yep.
Jutta Treviranus: And one of the commitments we made early on, in the last round was that we would lean towards.
Not repair, but more
the preparation of, or like a proactive.
But of course, I think there's many more tools and resources.
strategies available now for repair, about review and repair.
Israel Cefrin: Hey everyone, can you hear me?
Wendy Reid: Yes.
Israel Cefrin: Okay, so I'm not driving anymore. I'm in the office here right now, so I can enjoy the conversation freely. Just some question on the layers for review layer. Does it include something like a whizzers when you have something that's correcting but generated at the same time?
So because a lot of tools has a wizard of templating. So is that covered as well?
Or it should be something different than a review layer.
Jutta Treviranus: Yeah, I think templates are a really critical thing. So, especially within government, you know, Canada has the common look and feel where the template.
Jutta Treviranus: predetermines whether accessibility is there or not, and I think that continues, so it's something that
So that is good to preserve from the last round.
Israel Cefrin:
Okay.
Wendy Reid: Yeah, I'm trying to think where it fits in. 'Cause it's almost, I mean, I guess it depends on when you, when you.
Do you start with a template, or do you
add the template later, because I could see it being as part of the default authoring layer, in the sense of okay, I need to go create a document, I know I have a template for this, I use the template, and I work from the template.
Jutta Treviranus: Well, the the discussion we had earlier was who's who is.
responsible for the template is the organization, or is it? And what what responsibility can the authoring tool developer take? So it we had long discussions about how our templates created and what support is there for templates.
and enterprise-wide authoring environments, what types of constraints, and so it got into a discussion of.
the, yeah, the mechanisms available to restrict or to constrain or to, yeah, so you're right.
It gets complicated.
Jutta Treviranus: Yeah, and I think adding the word wizard to that is also good.
Wendy Reid: Oh, this is what I was thinking about.
was,
We're talking about authoring
Authoring roles.
And
I think this led got us to this
Where was it? We were talking about this last week. The state of the content.
There we go.
Because
Content can have different states, and then who is responsible for it.
In its different states.
I also hit me,
in the digital age.
Is content ever really finished? Is there ever a final product?
Or a final version, because you can always update and edit and
And then I know with accessibility, at least in the publishing world, it's not uncommon to have, like.
You know, say a book is published by a publisher and it's released.
And then other parties can come in after the fact, like alternate format providers, take that finished, air quotes.
product, and further transform it.
into other formats, or even the original format, but enhanced.
Jutta Treviranus: Yeah, and this is more of a WCAG discussion, but permission to actually
Fix things, right? And.
Wendy Reid:
I guess something, actually, we'll go back to UX design patterns, because I guess
the thing that makes authoring the authoring tool part of things so complex. I think I think this is where the the struggles in
WCAG world are coming from is because what is defined as an authoring tool is often integrated within the context, very often integrated within the context of a WCAG.
of a WCAG context, right? Because, like.
What is an authoring tool? An authoring tool is almost any text entry.
interface.
For example.
And so it's any text I guess it's any text entry interface.
That offers more than.
Plain text.
Like, if you can, if you can bold or highlight or underline,
It's an authoring interface.
Jutta Treviranus: Right.
But you were saying that it's entangled in WCAG?
So
The delineation between what is accessible?
accessible content versus the ability to create accessible content? What are some of the fuzzinesses in that?
Wendy Reid:
I mean, I think the fuzziness is when the content is created.
What to do with it?
So a lot of the recent discussion has been especially around user-generated content and who's responsible for user-generated content.
Jutta Treviranus: Right.
Wendy Reid: Should a
Should a website be responsible for the content generated by its users?
Wendy Reid: That it let them generate.
like Instagram.
Jutta Treviranus: Right.
Wendy Reid: And I do think it is it's an interesting thing.
That
Technology has changed pretty dramatically.
Because it is you know, even even a few years ago, it would have been crazy to suggest that maybe
Like, user-generated content should actually be
taken into account in the accessibility of a website.
But now we have all these tools, and we have all these methods, and we can automate so much more of this should there be a higher level of responsibility for user-generated content?
Jutta Treviranus: Yeah, and
Right. And are you the user an AI as a user as well, modifying the content? Yeah.
Wendy Reid:
I think that's where the fuzziness comes from. There's also they were they were starting to explore some requirements in WCAG 3 that, like.
sounded awfully ATAG-y.
So it was like content, I forget, there was one where I was, this is not yours.
But it's in text production, I think.
Jutta Treviranus: Right. Yeah, so the fuzziness or the confusion is coming that WCAG is moving further into.
Authoring tool.
It's that, yeah.
Jutta Treviranus: Which, to some extent, was the justification for doing WCAG 3, that it would integrate the three areas, because I think it is quite fuzzy, but unfortunately, there
much of I mean, there is very little attention to the authoring environment, so yeah.
Ideally, it would have been good to have had a much more robust look at the entire thing, the whole system as a whole.
And UAG as well, I mean the user agent.
Wendy Reid: Yeah.
That would have been good.
But this is the approach we're taking now.
Jutta Treviranus: Yeah.
Wendy Reid: All right, so we have design patterns, citations.
Jutta Treviranus: So we decided intervention was the term we would use for what was the problem?
Wendy Reid: I don't think we did. I think and I I think where we landed last week was that, there's a bunch of different we have, I think right now we have guidance, or interventions, or requests, or suggestions, or directives, or holistic we have
and it might be that.
we
the word will not come to us until.
You're actually writing.
that recommendation.
They all sound like they could work, but until there's
the actual language. It's hard to say which one will feel right.
Definitely we're going to try and go away from prompt because we know what prompt. Prompt is too heavy.
The prompt has taken on a new life.
Jutta Treviranus: Yes, it's been appropriated.
Israel Cefrin: Wendy?
Wendy Reid: Yep.
Israel Cefrin: Yeah, so I was just reviewing the part with the common authoring applications with the email applications and the word processing applications.
Israel Cefrin: So another item that comes to my head is the CMS.
And the CMS behavior application. So, sometimes you have a CMS, a pure CMS, like a Drupal, WordPress.
or Joomla, but we also have applications that works partially SMS. They are more like a journal publishing systems. And I can say OJS, JNUA, SCARS one.
They are at the same time.
publishing system.
But partially with the entries for text with a bold.
Styling and everything.
So I don't know how to frame that because not purely CMS.
But it's CMS-like.
Wendy Reid: I have CMSs up here.
In web content creation and editing.
Israel Cefrin: Okay.
Let me grab here.
At the top, Rocky.
Wendy Reid: Yeah.
Israel Cefrin: Oh, yeah, yeah, yeah.
Wendy Reid: Yeah, the common authoring applications are things that can, I think where I landed were, they could be on the web, but they could also be software.
Israel Cefrin:
For locally. Yeah. Okay. Right. Okay. Sorry.
Wendy Reid: Yeah.
Yeah, there's a lot of authoring tools out there is what I've learned.
Israel Cefrin: I just would add something with what you see, what is okay, the WYSIWYG with no WYSIWYG interfaces too, because there are SMSs that are just marked down.
So.
And they provide you.
if you have a knowledge enough, you can use a Markdown and make bold links and everything.
Wendy Reid: Yeah, I think the Markdown has become the
Ubiquitous.
Israel Cefrin: Yeah, it's like a standard GitHub.
everyone. So a lot of, you know, known database MS, they use Markdown as the flat text.
in storage before potted.
Wendy Reid: It's fun.
The funniest thing about Markdown becoming ubiquitous is that there is not one version of Markdown.
Ned Zimmerman: Yeah, I was just gonna say there's CommonMark, but then the
Wendy Reid: Then there's GitHub flavored markdown.
Ned Zimmerman: John Gruber says, well, it's only my, yeah, and John Gruber maintains that nobody else has, yeah, there's a weird dispute between the co-author of the original Markdown file, like Markdown implementation and those other ones.
Wendy Reid: And the differences between the versions are so subtle.
Ned Zimmerman: Yeah.
Yeah. And then some of them have other conventions built on top of them that come from
other formats,
what was the one I was looking at recently?
Israel Cefrin: Typist?
Ned Zimmerman: Is it footnote footnotes? Well, types, but also there's there's, like.
Footnote notation in some markdown flavors that comes from other.
Other,
What was it?
Wendy Reid: Anybody talks
Ned Zimmerman: Maybe latex, but yeah, anyway, yes, it's a whole can of worms.
Wendy Reid: Yeah, we were discussing it for EPUB.
Because we're working on an annotation specification, and it was, well, could we say that Markdown could be an acceptable body text format for the body of an annotation?
And then our team contact was, what version of Markdown? Because as soon as we say the word Markdown, we need a reference because this is going into a specification. I was, oh, never mind. We're not touching that.
Ned Zimmerman: Yeah.
I mean, yeah, it gets tricky with that too because there's just so many things that didn't really exist or weren't in common use when Markdown was originally created.
block quotes with citations. I mean, there's no cite element, there's no way to put a cite element in a block quote in Markdown that I'm aware of, unless there's an extension, but it's certainly not standardized in any way, so yeah, it's
Jutta Treviranus: And then math and latex and and and that's going to be Wayne Dick will be very useful in the committee.
Wendy Reid: Oh, yeah. I was actually interested in — so he mentioned it.
So MathML has obviously, I think, become
significantly better, and that it's still being actively worked on, and that we now have, what do they call it, MathML Core, I think is the latest, latest iteration on the MathML spec.
and is actually properly supported across browsers, whereas the old version was not.
but when he mentioned the authoring part, I I realized, I was, oh, I've never actually
I've never written MathML. I know what it looks like. I know how it appears, but I don't — I've never tried to write MathML.
And admittedly, I've also never written LaTeX. I miss that era of publishing where everything was
was typeset in LaTeX, and I know it's still common in some pockets, but I've never done it, and I'm, oh, that's an interesting problem, too.
Ned Zimmerman: Well, and I'm glad, I think, Israel, you just brought up typist, and I think that's an interesting one too, because.
Ned Zimmerman: Are others familiar with that? Jutta, have you run into this? Let me just get a link for it. I think it's
The reason it's relevant for our purposes is that.
It has,
HTML export now. So, I think it's meant as sort of a modern
a modern approach to something like LaTeX that's a bit more
user-friendly, at least in my experience.
But it includes, you know, PDF rendering, tagged PDF rendering, which is quite good. it includes, now export to HTML and all kinds of footnoting capabilities and typesetting capabilities, so
Israel Cefrin: Yep.
Ned Zimmerman: It definitely counts as an authoring tool now under the parameters of ATAG.
Ned Zimmerman: So this is another I don't know if this fits into any of the categories you'd identified, Wendy, but it would be a good one to put in there, too, as an example.
Wendy Reid: And
Jutta Treviranus: Yeah, and it brings up the export issue, which we grapple with as well. The export certainly has there's many more complications with that.
Israel Cefrin: Yep.
Yeah, type is more focused on the type setting, or at least on the my work here in the library. I have a colleague, a co-worker, he's a huge fan.
And he prefers over the pure markdown because you can generate PDFs and even math formulas from typists much easier. So it's a content authoring, content authoring tool. It's local but web at same time.
And the constant struggle with the mathematical formulas, TypeScript handles that really better than other tools and platforms.
Wendy Reid:
Am I describing this correctly, like latex authoring application?
Or should I even just more broadly and say advanced typesetting?
Ned Zimmerman: I would find a more type, maybe typesetting types, advanced typesetting. Yeah, I think that's what I would say. Yeah.
Israel Cefrin: Type setting, yeah. Yeah.
Advanced app setting, yep.
Wendy Reid: Yeah.
I also, it was, my, supervisor.
Peter for for my MRP. I know he's a huge fan of Overleaf.
Which is like an online latex editor.
Wendy Reid: Which I have never used. I've debugged LaTeX once in my life.
And I'm going to ask a dumb question because this is a small group of people. But it came up last week, I think, latex and latex.
And I don't know what the difference is between those.
Ned Zimmerman: I think it's one of those things where it is supposed to be pronounced latex, but everyone pronounces it latex.
Israel Cefrin: Yep.
Wendy Reid: Okay.
Israel Cefrin: Yeah, it's LaTeX because it
Ned Zimmerman: Like, give him
Wendy Reid: I swear I've seen it spelled with a C, too.
Israel Cefrin: tech from technology, something.
Wendy Reid: Mmm.
Ned Zimmerman: Yeah
Israel Cefrin: The root word from Greek, I think, is tack.
Ned Zimmerman: And it comes from the text typesetting system. So I don't know how that would've been pronounced, but I think, Wendy, it's just a GIF/GIF thing. Yeah.
Israel Cefrin: Yep.
Wendy Reid: Okay, okay. I've also seen people type it that way, where it's like TEC instead of TEX.
Israel Cefrin: And it's LATAC.
Ned Zimmerman: Oh, yeah. Yeah.
Wendy Reid: Which is why I'm
are there two? Is this a markdown thing, where, you know, one direction went CommonMark, and the other direction went GitHub-flavored markdown? And,
Israel Cefrin: Yep.
When you talk with the people that are purists for like a heavy LaTeX user, you say LaTeX.
when you talk with the people, average people like us, you can say LaTeX, that's fine. I was yeah, I was correct once by a guy, was he's a really huge fan of LaTeX, so it's LaTeX, not LaTeX. Okay, thank you.
Wendy Reid: Okay.
I'm gonna get rid of this.
I also had to correct myself, because I know better than this, but I had typed SciML earlier, and I was, wait, that's not correct. It's ChemML. And there might even be other marks. I know there's a,
I mean, if we get real technical. So there's obviously MathML, there's ChemML, which is not
an official web specification, but it's a spec there's a spec floating around for ChemML.
that I assume is implemented somewhere probably in a polyfill.
I know there's a musical notation language.
I think it's XML based.
Ned Zimmerman: Yeah, that's true.
Wendy Reid: I'm gonna guess that there's others, because chemistry notation, sure, but, like.
What about other scientific?
Ned Zimmerman: There's also
This is also
may be relevant, I just dropped it in the chat, but Mermaid, which is a charting
Wendy Reid: Yes!
Ned Zimmerman: Language? Yeah.
Wendy Reid: I know about Mermaid.
Wendy Reid: Actually, with AI, that's become incredibly common. I've noticed Mermaid is becoming
Ned Zimmerman: Oh, yeah.
Wendy Reid: very ubiquitous in terms of, like.
Ned Zimmerman: Oh, wow, they've actually moved it to mermaid.ai now.
Wendy Reid: Of course.
Ned Zimmerman: All right. Well, there you go.
Wendy Reid: Yeah, because a lot of I've noticed a lot of, like.
My skills and things have mermaid.
Wendy Reid: workflows embedded in them.
Ned Zimmerman: I don't think Mermaid produces very accessible output.
That's neither here nor there, but just something that I've heard about. Yeah. Yeah.
Wendy Reid: No, it does not. It's very plain text.
Ned Zimmerman: It's basically just ASCII art.
Wendy Reid: That's, if anything
It is it's ASCII art, really. it is actually interesting that some of these conventions that are coming coming out of the use of LLMs
And
So, like Mermaid,
I mean, arguably even, like.
markdown to a certain extent.
Raw markdown.
is
Like, I realize I love writing in Markdown, it feels very intuitive, but I also realize if I was a if I was a
screen reader user, and someone just sent me a markdown file, and I was reading it.
in its raw form, you would hear, basically,
Hash, hash.
You know, blah, blah, blah, blah.
Ned Zimmerman: Yeah.
Wendy Reid: You know, asterisk, asterisk.
But, like all those conventions, and obviously.
Ned Zimmerman: Yeah.
Wendy Reid: If you're a seasoned Markdown user, you know what that means, but it is
It's really it's noisy, right?
I don't think I've had to describe things
Mermaid.
Actually, I'm going to take, I'm going to take this out of authoring applications, authoring languages.
So we have lemonade.
JS.
Ned Zimmerman: Markdown and CommonMark.
What's the one there's what's the what's the one in that Wiki Wikimedia uses? Is it called Wikitext?
I think it's called Wikitext.
Wendy Reid: Yes.
Ned Zimmerman: Yeah.
We're just getting data sets now.
Wendy Reid: I don't want the dataset. Wikitext. Just wikitext.
Ned Zimmerman: Oh, yeah. Wiki markup or Wiki. Okay. Yeah.
Wendy Reid: Wiki markup or Wiki code.
Ned Zimmerman: And then there's the one that Atlassian uses in things like Jira and Confluence, which I forget the name of, but it has a name.
Wendy Reid: I know what you mean, because I've probably used it.
Atlassian, Martha's Language.
Markup syntax.
Ned Zimmerman: Yeah.
Wendy Reid: Or have they transitioned fully to
Ned Zimmerman: Well, it says they support
Wendy Reid: Markdown.
Ned Zimmerman: That's Bitbucket, though.
Wendy Reid: I think they support Common Mark.
Ned Zimmerman: Yeah, chair conference
Wendy Reid: I think they've
Ned Zimmerman: Oh, so their new editor has implemented Markdown support.
Wendy Reid: Yeah.
Ned Zimmerman: So I don't know what it was before.
But I know there was a different name.
Wendy Reid: Yeah, I remember using it years ago, but I think they've given in to CommonMark, basically.
Ned Zimmerman: Yeah, yeah.
Wendy Reid: I'm trying to think of, so we have obviously MathML.
And now I'm trying to remember what the name of the —.
Ugh, music rotation with
What are you called? Music XML. There we go.
Ned Zimmerman: Music. Music XML.
Wendy Reid: Music XML.
Jutta Treviranus: And there's many more music languages as well.
Wendy Reid: Just trying to think. Let's try specialist markup.
Document markup link. Oh, that's Wikipedia.
Israel Cefrin: There you go.
Wendy Reid: Okay, here we go. We got the OG, HTML, cybertext markup language, keyhole markup language, XML-based markup language used to exchange GPT.
Ned Zimmerman: Oh that's the that's the Google Maps one like the Google Earth files when you download over layers for Google Earth it's in that KML.
Wendy Reid: Oh, that's cool.
Ned Zimmerman: I hadn't seen that before.
Wendy Reid: Okay, this is like a delightful way to spend our last couple of minutes. We have markup. We have MathML. We have SVG, I guess. Yeah, SVG is a markup language. Text, LaTeX.
Ned Zimmerman: Yes.
Israel Cefrin: Good.
Ned Zimmerman: Yep.
Wendy Reid: Wiki markup, extensible 3D graphics file formats.
Jutta Treviranus: Oh, and then there's VRML and all that as well when we get into three-dimensional and four-dimensional.
Ned Zimmerman: Hmm. Yes.
Wendy Reid: Here's the fun ones, lesser known document markup languages. Okay, so we have ABC notation for music scores.
Ned Zimmerman: Oh, DocBook, that's an interesting one that I see quite often.
Wendy Reid: Yes.
Ned Zimmerman: for manuals and documentation. ASCII doc.
Wendy Reid: oh, the new Tex-Mex?
Ask Doc.
Ned Zimmerman: Maybe ASCII doc is another one I was thinking of.
Wendy Reid: There's — I've always heard this called chamomile. I wonder if it's different.
Israel Cefrin: The journal article tag suits JETS has a plugin for OJS, so can create like a journal submissions thing.
Ned Zimmerman: Oh, okay.
Wendy Reid: Oh, JATS. I forgot about JATS.
Israel Cefrin: Yep.
Ned Zimmerman: Mmm.
Israel Cefrin: Yep.
So really use only scientific scholar communication.
Wendy Reid: Yeah, wow, there's a lot of music ones.
Ned Zimmerman: Cool.
Wendy Reid: I'm gonna guess it.
Ned Zimmerman: Oh, restructured text. I was thinking of restructured text as well. That's another common one that is used as an alternative to Markdown.
Ned Zimmerman: Python, yes, it's used in Python a lot. That's right.
Jutta Treviranus: Yeah, this is a many technical piece, I think. I'm in the middle.
Wendy Reid: Okay.
I kind of love this.
Jutta Treviranus: Yeah, we're largely.
Ned Zimmerman: Yellow page directory listings? Oh my gosh.
Oh boy.
Israel Cefrin: Oh.
Jutta Treviranus: I've just been attending the XR meetings, and the number of XR markup languages that are there is just massive.
Wendy Reid: What's that, what's that famous XKCD comic, you know?
Ned Zimmerman: Oh yeah, there's types.
We'll invent one standard to fix all the other.
Wendy Reid: You know, this standard doesn't, there's too many standards. Let me create a new standard to unite the standards.
Ned Zimmerman: Do everything, yeah.
Yep.
Now we have another one.
Wendy Reid: Now there are 15 standards.
Israel Cefrin: Just a note.
Wendy Reid: Official ones.
Ned Zimmerman: Yep.
Israel Cefrin: just a note from my end here, I was just reviewing the authoring tool types and functions, and for the first category of a web content creation,
I think that you should add as well the, platforms call for journal management and publishing systems.
Because we have open source and have commercial, solutions, so I would say that Open Journal Systems is the most used in the world as open source, but we have platforms commercial, like Editorial Manager for, Xavier, ScholarOne.
Ned Zimmerman: Yep.
Israel Cefrin: So we have both open source as OJS. In January, we have commercial solutions as ScholarOne.
and they include the, WYSIWYG, non-WYSIWYG, JetXML, and everything, so but they're still all web-based platforms.
Ned Zimmerman: Yep.
Israel Cefrin: Yeah, correct, yeah, thank you.
Wendy Reid: Okay.
Yeah, it's a good one to know.
Jutta Treviranus: Yeah, and one of the requirements we were considering before was whether to encourage the use of open standards.
Markup standards.
Wendy Reid:
I just realized I'm doing my own my own former industry quite.
dirty, because I am not talking about
Ugh, a little shame.
Israel Cefrin: Yeah, I don't know about a scholar one. I know OJS uses JATS.
as a plug in for article submission, and they're permitting a new version of a JATS in the future. So, I think it's open sender, hopefully.
Wendy Reid: Oh my gosh, why am I suddenly blanking? sigil
There's more. It doesn't need to be an exhaustive list. It's just so when someone sees it, they know what we're talking about.
Jutta Treviranus: Yeah, there's no way to be exhausted because there's new stuff coming out constantly.
Wendy Reid: Every day, especially with I mean, the thing I find kind of
it's weird it's a strange impulse, but it's I kind of understand why with AI and being able to literally just ask for things to happen instead of people looking for tools, now they just make their own.
They'd be, oh, I need a new thing that does this.
Jutta Treviranus: Yeah, so all the agents, yeah. And that sort of prompts us, I think, to go back to very, very simple definitions of what is an authoring tool and use this list to say.
Wendy Reid: I'll just build it.
Jutta Treviranus: These are included, but this is not.
full list type of thing.
Ned Zimmerman: Yep.
Jutta Treviranus: I think, you were saying anything that produces text, but of course it's anything that produces content. It needs to produce content.
Because it's all media forms.
Wendy Reid:
Yeah, it is,
I mean, it's interesting because the stuff with LLM is almost like we have to go back to basics because if everyone is going to
Theoretically create their own thing, possibly.
Wendy Reid: What do you need to know to do that? You need to know
Core functionality, you need to know what language to use, you need to know the conventions of that language.
So if I want to make my own latex
editor.
What do my tools need to know? They need to know how to authorate LaTeX. They need to know all the rules.
Ned Zimmerman: Yeah.
Wendy Reid: And they need to know how to do it accessibly.
to my needs, because I guess this is the thing let's say Wayne's use case of not being able to
to
comfortably use textbooks and, and, or even, software for MathML.
And it's, if Wayne wanted to create his own MathML authoring tool, what does it need to do for him? And what does MathML need to be able to do to help him?
He added a really, really comprehensive comment that I've not had a chance to read yet.
But it is really important. It's accessible STEM output. Yeah.
Ned Zimmerman: Oh, cool.
Jutta Treviranus: And I wonder if we could draw John Gardner into this conversation, because he did such an amazing amount of work on this.
Wendy Reid: Yeah. I mean, if anything, I want to try and articulate this in a better way.
Ah, how do we
Actively
I'm gonna say it.
We've got time.
a half-formed idea, but it is this idea because what I'm just reading his what he said correct quickly, and it's mathematical accessibility is a profound repair problem.
you know, papers are written and distributed as PDFs. Those PDFs need to be remediated, but the only person who can remediate it is an incredibly specialized accessibility expert, because you essentially need an accessibility expert that can read math.
at a, you know, graduate level.
Graduate level.
And it does it does raise the question of oh, surely there's gotta be something that can
Translate.
these things., if
If you can create the markdown, you can create a translation for that markdown, right?
Jutta Treviranus: Right, but the
Right, but you should we had a project with Mathematica, and the debates regarding.
the actual meaning, or how to write specific math things was just so crazy. Jason White might be a good person to pull in, because he's
He did his thesis on it, he's using a screen reader, and he, yeah.
I don't think, I think we do need math experts to tackle this.
Wendy Reid: Yeah. Okay.
For sure.
Jutta Treviranus: Or even have a subcommittee that tackles that.
Wendy Reid: Alright,
We'll probably have to have subgroups or something or task forces on some of these like really niche.
Jutta Treviranus: Yeah.
Yeah.
Wendy Reid: topics.
Okay, we're at time. I want to — Jutta, I know you've got —.
Deadlines. We will enjoy your long weekends, everyone, and we'll talk again next week.
Jutta Treviranus: Yeah, I do.
Ned Zimmerman: Thanks.
So yeah.
Israel Cefrin: Thank you all. Have a good day.
Jutta Treviranus: Thank you.