 Alright, I think we can get started. Alright, so this session is a panel on accessibility metadata. And we're joined by several panelists, I am going to see if I can pin them so that you can, you can see them all group together and I think while I do that we can ask them to all introduce themselves so if somebody would like to get started. Let's see. Alright, I have, I have our panelists names here so I'm just gonna say Amanda Lee would you like to start us off. Sure. Hi, I'm Amanda Lee. I'm a publishing metadata consultant I work almost exclusively in onyx. So if you have onyx questions I'm your gal if you want to talk about the metadata within e pubs. Also so my pronouns are she her. And I am also working on the e bound project which Deb introduced yesterday which is the accessible publishing learning network so if you have any questions about that I want to be involved in that I'm also your gal. I work with publishers small through to large on their metadata as well as publishing organizations so while I am myself not a publishing employee I have a lot of publishing perspective sitting in the back of my mind. So that's what I will be bringing to today's now. Thank you so much. Who would like to go next. So I'm Aaron at from red shelf. And if you don't know what red shelf is we are reading system, mostly based in the higher end space, although we do have some inroads into like, you know, business training that sort of thing. So accessibility metadata is been a huge thing for me for the last few years so I'm so happy to be able to share what we've done on the panel today but I'm excited to hear what everyone else is doing and learn a little bit more, just kind of how this works internationally to kind of would you like to go next. I just came in. So can you catch me sorry to put you on spot we're just introducing ourselves here. My name is Kai, I'm an accessibility analyst for Nels. And so do a lot of testing but when I'm not doing that I'm helping out with cataloging. Wonderful. Megan would you like to go. Hi, I'm Megan. I work at the Center for equitable library access. I am the metadata coordinator. I started a new role that they just created. I started in it in October. And I'm probably going to bring a little bit of perspective on Daisy audio, some Braille, I work mostly in Mark, but we also work with onyx. But I'm mostly here to learn and contribute as much as I can but I'll just say that. I work kind of what we're doing now and I'm always open to hearing, improve what I can do to improve and fix and make things better so. Wonderful. And Gregorio. Hi, everybody, I'm Gregorio. I work for our foundation Leah, we are based in Italy in Milan in the north of Italy. We have a certification process and our certification is based on metadata. So for us is really important to create, distribute and consume metadata so this is my, let's say point of view, and I'll try to contribute to the panel. Great. Thank you so much everybody. My name is Marisa, and I'm from the Daisy consortium. I'm happy to be moderating this panel today. I really look forward to hearing what everybody has to say. So should we get started with our with our first topic. I think we were going to get a high level overview of the metadata supply chain. Yeah, so maybe I can start then maybe Amanda can help me. I have a flow chart to share with you. Let me check. Okay. Can you see it. So this is our idea is not really our but this is the idea of an accessible publishing ecosystem. As you can see it starts from content creators. In this slide we have a chart with the different actors of the value chain of digital publishing accessible digital publishing, and we start from content creators, then the distributors aggregators, and so on. The platforms for selling or for lending content. Then we have the devices and the reading apps. One important thing is that and we can see in these other slide. The data is the only information that flows from one actor to the other. Sometimes happen that for, for example, online digital bookstores or digital libraries, don't even get the file itself because the file is stored on the aggregator servers and so on, but they get the metadata and are the metadata that are shown to the end user. So, since the metadata is the only thing that flows from one actor to another is super important to embed accessibility information in those metadata to help the end user to understand what they are buying or lending or so on. So, if Amanda wants to add something and they. Yeah, that looks great. I agree that it's getting the metadata is of course really important because it does flow everywhere. But it's interesting to me that frequently we forget that within the creation of the metadata step there are actually several steps from the publisher that you know that very first step on Gloria's chart here where it's just ebook plus metadata. Before that the publisher needs to determine what are the accessibility features of any given ebook and where best do I store that information so I consider those sort of step point five of the supply chain, figure out what the accessibility features are figure out where to put them put them there, then step two which is distribution. Thank you so much for that overview. So, next, think we want to talk about this, you know this very beginning metadata creation. The question is how do you incorporate accessibility metadata. Well, I'll jump in and start first just because the publishers with whom I work. You know that's they're creating the metadata that's where they start. For many of them there's a big challenge around both the on ex metadata and the metadata in the pub file themselves just as I alluded to they have to determine what those accessibility features are before they can talk about them. So, in order to incorporate accessibility metadata they need to have a workflow that allows them to speak with the producers of the book whether that's in house or third party. And in Canada it's frequently third party. I can't speak to other nationalities but here we do a lot of third party book production, and then once they've determined what the features are. I know at least the on ex metadata is sometimes added directly to the database, and then it's sent out from there just as all the other metadata it's not different, you know I think often it's sort of we think of it as accessibility data in terms of on ex it isn't it's a product feature just like saying your print book has 182 pages and it's got flaps on it. So, from my perspective as someone who works on it, it's as simple as learning what it is adding it to your database. I would say in the US a definitely big third party piece there as well. So that's where a lot of that targeting of the certification of that workflow comes in making sure that everybody in that whole process is certified. And I think what our publishers really seem to struggle with, at least in the higher ed space is when we first started looking at including accessibility metadata something we could display to consumers. We started with Onyx and we just weren't finding a lot of it in the Onyx feed. And you know that's why we sort of made that radical shift to kind of pull both but rely more on the EPUB metadata so you know it's it's interesting to to see how that is kind of played out over the last few years. I have so many thoughts about that Erin that I would like to discuss with you but let's hold that one until you know a later question perhaps. In Italy, our process for certifying eBooks does the complex work for meta dating the eBooks so we receive the eBooks from publishers we have a checklist with more than 40 different tasks. And so we validate the eBooks and the output from the validation are actually the metadata so we then send back to the publishers those metadata only accessibility metadata both in EPUB metadata and in Onyx and they dispute them. This is quite important from for our model because in this way we are the authoritative party for metadata. All the metadata in Italy are consistent because we produce for the world, the old publishers, and in this way, all refers to our website where we have the certification and every user can check if that particular book is okay or not. So now getting into the different formats I've heard people mention Mark and Onyx. How do these different schemes compare in their support for accessibility metadata. I could start maybe a bit on that. I guess so with CLA we produce and we also distribute accessible resources. So we do work with Onyx, and we then create mark records to get the resources or to send out to public library partners. So for our mark records, I find I've been looking for some sort of standardization of how accessibility metadata is incorporated into mark and I've looked at a lot of other accessible libraries to see how they're incorporating these fields but it is a bit of a Wild West out there and everyone does it a bit differently. So it's, I think there's lots of room in mark to incorporate descriptions and features and and there's flexibility in that sense but that also creates a bit inconsistency with Onyx I'm finding that it's just more structured and we're able to the language is that pretty well defined in terms of how it's described so yeah that's what I noticed. So Megan are you doing almost a crosswalk from the Onyx data that you're getting into a mark record is that something you've figured out how to make work. Not yet. Have you talked to me please. I'm looking at it right now like we're not parsing accessibility fields or the product form feature fields from the Onyx that we get from suppliers and we also create our own Onyx. So that's kind of on our roadmap is let's parse it from our suppliers and then start adding it if we are creating the resource ourselves. And then from there, I think there should be a pretty quick, like crosswalk for the accessibility fields in particular because in mark there's only, there's about two fields there's 532 and 341 and I think there could be like just some transparency and language between what Onyx has and what mark has so like full audio. Yes, we could easily put that into the mark fields and, and the way that the Daisy specs are kind of stylized like just make that more interchangeable. I think that would be really useful but we're not quite at the crosswalk point yet but we're hoping to get there. Yeah. So should we move on and talk a little bit about the supply chain journey. So once you have authored and created this metadata, what happens to it, as we saw from the helpful diagram from Gregorio it has quite a journey ahead of it. And the goal it's I feel like it's like a some kind of game show where the goal is to like get it to the end of the chain in one piece. So how do you do that. Yeah, so it's like a broken telephone game. I don't know if you know that particular game. So, maybe Amanda can say something about the distribution itself of metadata, and I can add then something about issues that we have found so please Amanda start because the phone is just a perfect description of the metadata supply chain I'm sorry to say, just like I said publishers put it in their database and they press the big button that says distribute or they send it in a spreadsheet to someone else who presses the big button that says distribute. And then it goes into the ether, and you can maybe go check on some websites to see if it's taken, but you can never really be sure that what you're sending is what's arriving, especially the big players like Amazon it can be very very very challenging. So from the publisher perspective of the supply chain we press the magic button, and maybe the magic spell works and maybe it doesn't. But we're working on that we're working on making that better. Yeah, what we've found implementing this ecosystem in Italy is that there are there may be some issues. So, for example, the first one from content creators is that sometimes content creators use the same ESPN for different formats, I know that it should not be done but it happened. And so you cannot attach accessibility metadata on that particular ESPN because if the same ESPN refers to an EPUB or a PDF. As you know, the accessibility features may be really really different. So this is the first issues we faced. The second is that sometimes the back office system don't have the fields in their database to store accessibility metadata. They are quite few. And so normally who implements those software are not focused on accessibility metadata and they have to do some work to receive those metadata. Then when they receive the metadata and are able to pack them and to distribute them, it may happen that bookstores or online libraries are not able to ingest the metadata because they are not ready to do so. And then the big problem, but for that we have, let's say solution, they don't understand how to display those metadata because some of those are really geeky, are really for nerds of digital publishing. And so it is quite difficult to understand how to interpret the metadata and to display to end users. So ESPNs, databases and display to end users. I can say that those two of those things were something that we certainly struggled with. We obviously we received the Onyx feed from our publishers, but we also know that the metadata lives inside the EPUB in a lot of cases. And it was like, okay, well, we know we can rely on the Onyx feed. And we use daisy ace to pull metadata out of EPUB, specifically accessibility metadata. But it was like, where do we put it? Okay, so that's what we kind of had to like, we have this grand plan of, oh, we're going to launch this thing and then I was like, oh wait, now we got to kind of go backwards and say we're, we need to set the way up to display it and then being involved in like creating those standards around how it should be displayed was really helpful for us. And we, you know, follow that as closely as we can based on the information that we get, but those are two definitely too big stumbling blocks for us because we had a completely different idea of how it might be displayed initially. And then we thought, well, why would we want to do this sort of in a vacuum and that's why you know we got daisy involved in the conversation to like shouldn't we have some kind of a standard here. And so now we do have a standard. So that was great. And then everyone can follow it, right? At least on our side, you know, and hopefully it'll start to move itself around to other avenues where people can say, oh, hey, here's this metadata that I'm getting, maybe I should display it like other folks are displaying it. Yeah, maybe I think we were going to also discuss a little bit later but but maybe now would be a time to drop a link into the chat to this standard for displaying metadata. Does anybody have it handy. Absolutely. Hopefully that, you know, this, the development of this document will help people understand what they should do with this accessibility metadata and that it is important to expose it to the user. So talking about metadata consumption, how, how the user interacts with this information once they see it displayed. And so what value does this metadata bring to users. I guess I'll jump in here, since I'm pretty much talking from the perspective of a user. And I think it is really helpful, especially in an academic setting where you're looking at accessible content and making sure that all the materials that the student is using is accessible. So, in that sense, there's a lot of value there, but also from a student's perspective, knowing that textbooks can be pretty expensive. I think it really helps to know whether the book has image descriptions and other features that you need before you buy can kind of avoid that headache of. You know, I just bought the book and then I paid like 100 bucks. And now I have to talk to the company the to kind of ask for refund because the book isn't accessible and find another way to kind of fix or to get my money back and to find another way to make the book accessible. So I think there's a lot of value there. Yeah, definitely this is similar on our end, you know, because one of the I know a lot of higher ed institutions, especially the in the accessibility space. They're a little bit afraid of EPUB still for some reason. And I, you know, we're constantly fighting that battle but it's a little bit easier for them to understand when they see that information attached to a book. Okay, all right, well, this has gone through some kind of a workflow certification and it's compliant to this. And so, you know, maybe it's not so scary, maybe I should have my students try EPUB. So that's helped move the needle a little bit for us. It also helps on the adoption side of things too so if you can demonstrate to professors, faculty bookstore managers who are making choices like that that you really should choose to EPUB and this is why and here it is in black and white over you know maybe a PDF that doesn't have that kind of capabilities that will go a long way in helping those accessibility offices so it's just kind of the full full circle thing. So I have a question. What would be a great example of displaying useful metadata at a point in the, you know, user experience so like the portal through which somebody is purchasing a book. What types of things do you as a user look for that would signal to you like hey, you know this book seems promising and it's accessibility. I think for me, I'm looking for image descriptions. Kind of good book structure for navigation, making sure that it's reflowable. If I do see a book that maybe does not indicate that it's reflowable then it's a pretty good sign that generally it's not as accessible for me as it could be. So, I think in that sense, that's usually how I look at it. For the most part I think because of the content that I want to consume. Even if I don't see all the accessibility features I want. I, I'm still probably going to try it, unless it costs a lot, and figure out a way to make it accessible, because in certain cases, you know I just have to get the information about whether it's accessible or not, and I don't really have much choice. But I think those are the things that I pretty much look for. And, you know, it's really cool to see that a lot of this information is now starting to be exposed. I want to just kind of talk a little bit about things from the NELS side. We're displaying accessibility metadata on our record pages for the different titles, and we're showing whether it's reflowable or not, whether it has image descriptions or not. Whether there's navigation in terms of headings which is one of the main features I think I talked about yesterday where a lot of users gravitate to. And I think that's a really good starting point. And I say that because there's still a little bit of inconsistency in terms of, you know, what users expect and what the metadata shows. So I think that's that's where there should be a little bit more focus as well. So what are some challenges when it comes to using a displaying metadata to its fullest extent? What are some of the uncertainties in this space? So I guess I'll start again. And for me what I found just helping out with cataloging is that the metadata in the EPUB itself that I guess we're using. There's inconsistencies in terms of identifying image descriptions or saying that this book has image descriptions because when I actually look at an image description, I'm expecting something that gives me enough detail to be able to use it properly along with the text, right? But sometimes I see a book that says it has an image description or it has image descriptions and the alt text is not quite there. Technically it does contain alt text, but all you're seeing are logos and, you know, for the book cover it's saying the title, which I guess it's helpful, but it's not, you know, giving me more as a user than what I'm expecting. And so, you know, I feel it's a little bit of a let down. And so I think, yeah, I think getting this accessibility metadata is really a starting point. Kyle, you're being very diplomatic in your description of this bad alt text. This I think is one of the challenges of accessibility checking is that you can automate. Is there alt text, but you can't automate? Is it good? So this, I think for me really reinforces the value of manual checking in addition to automated. And I know we hear this within accessibility, we say this over and over, but I think that maybe not everybody has caught up. Are there certain metadata fields that are must haves? I know we've talked a little bit about this already, but are there maybe even being more specific like using terminology from the specifications? Are there certain fields that you know you definitely must have? I think the most helpful one for us is the screen reader friendly. For sure. You know, sure the certification is important, but again that whole idea of the reading order and how you can navigate through the title to extremely important, but I think the one that we get asked about most often is screen reader friendly. So just to add something to screen reader friendly, it is not actually a single metadata, but it is a combination of different metadata telling that the user can access the whole content using a screen reader. So there are images description, the content instructor, everything is available as text and so on. It's a different combination and this is what we have tried to explain in the user experience guide to display metadata. So to combine different metadata to say screen reader friendly. I think it's a very good point about the book cover too. You know, we've talked about this in our in our Daisy meeting sometimes about how, you know, it shouldn't just say book cover or title like it really should be descriptive of it's not any different than any other image. You know, the student is saying ham using this purple book. And, you know, someone who doesn't have the affordance of site doesn't know that they're using a purple book. Like, it's, you know, it's a different experience so yeah that's something I've tried to encourage us to get more into describing our book covers but we again that's another thing we kind of have to rely on publishers to help us with, you know, and that's the other issue we have with metadata sometimes we just don't get it. I'll just add to that in terms of the library side of things with mark. For us we always add technical requirements and playback specifications. And we'll always add usually the structure and navigation like, like access and access by heading some pages information. So that's the Daisy specification as well that's for Daisy audio but for Braille we do a few other things we contracted or un-contracted we have the UAB grade and system requirements if it's electronic Braille but again it's, I think, speaks to kind of through the supply chain that's displayed in the library's catalog or not is another question so we can give it to the libraries but some of them might just remove it or do what they would they will so. And before we move on to our next topic. Are there any other recommendations on regarding searching metadata, things that might be useful to have say if you are using a portal to, you know, scan the contents of a library or a bookstore. What types of search facets make a difference. Yeah, I think for me it's a lot of the same features that I look for when I'm looking at a record. I think one of the really cool things that Nelson's been working on with this accessibility metadata is also being the search. And so it's really cool. And I definitely encourage you to try it out where I can enter a title and then I can search for image descriptions or what we're calling customized display, which is reflowable or fixed. And, you know, I think that is really helpful but I think the other thing to point out to just kind of going back a little bit with the whole screen reader friendly. I guess, announcement on on a record page. You know, because it's a combination of all these different characteristics that make the book accessible to a screen reader. So, you know, having that label could be helpful for, I think, for people who don't really know exactly the terminology or the, all the specific characteristics that you're looking for, you know, because I look at a lot of people with some of this. But, yeah, I think just making sure that everything is displayed in, you know, an easily like friendly with friendly labels is really important going forward. All right, we're moving on into our final topic. This, we can look at news, what's new in this space and also looking forward what is coming up. And so what are some of the advances that you've noticed in the last year related to accessibility metadata. I can think of one for sure. But I think we already touched on but maybe somebody wants to talk a little bit about it. Well I see a lot more of the larger publishers, having their workflow certified. There was a lot more in the last year, because there were just a few big ones in the doing it at first. Now there's a lot of folks getting on board and asking us questions about it. Like we're not certified authority obviously, but we can point them in the right direction, but the fact that even small publishers are trying to ask us about it means that the words getting out there that people want to make accessible. Because there are certainly people who are making not accessible the pubs as well. So, yeah, I think that that's, that's a that's a good development. Yeah that's great and so much of certification relies on metadata, which I think Gregorio mentioned. And we also have the user experience guide, which we have a link in the chat from earlier. Which is, I think it was new this year, and should be extremely helpful to help people interpret metadata so everyone please check that out. And what are some things that you're looking forward to or what's coming up in the near future that you know you want to talk talk about. I think we are going to update the user experience metadata because we had feedback, and so we will try to improve it in this year. So, next year. So, yeah, stay tuned. I'm looking forward to try to figure out how to leverage this metadata in different ways, you know, internally for for our partners and campuses so that's my big project this year. I know that catalyst has been talking about displaying on accessibility metadata, which, you know, for them they don't actually even have the pubs so they can't work with the public data so this is a really good step for them and I hope to see more non retailers like the catalyst who don't have the pub use that on expended data in that way and I really hope and I know catalyst knows what they're doing so I hope that they use the display guide because it's really really functional and super useful. And this is just a dream, but I do think that now that this display guide exists and it's so easy to use the publishers could consider exposing that metadata on their own websites. It's just and then that would also train them to think about their accessibility metadata you know go audit it on your website, make sure it's correct and think about it a little more I really, I'm really looking forward to seeing that the growth in the display space. That's a really good point. And especially if you're, you know, you, let's say, sales person as approached you on a campus about using a particular title. If they're educated about metadata they can point you to that on the publisher site like it's it might be a brand new title that maybe we wouldn't have for them to look at. I wouldn't have that metadata to show them at the point of adoption either so yeah that's a really good one. And I'll just add that I'm hoping, and maybe this is also a dream but I don't think it is I think there's some talk about developing standards for incorporating accessibility metadata and mark and I know, think ABC and I felt like we're working on a project related to that and so I'm hoping to kind of tap into that but I really want to get the ball rolling and start thinking about how we can incorporate it in a more consistent way across the board and so if anyone's interested in working on some big project like that I'm totally all ears and I would love to be involved. Great. Well thank you so much everyone I think we've got one more question and I'm going to ask Gregorio to talk about the European Accessibility Act and then I want to be sure to leave a few minutes for questions because we are running into the end of our session. I will be super short. Yeah so for the European Accessibility Act, it is mandatory to display accessibility metadata to end user. So as I like it before to let the user to understand if that particular book fit is or have needs. So it is super important but but for displaying accessibility metadata publishers have to create the accessibility data to distribute them and online bookstores and digital libraries have to display them so the supply chain is super important, but for the end user it is mandatory to have it before buying or borrowing a book. That's all. It's really exciting and important to look forward to and I feel like the work done in this space thus far and the expertise that you all have can really inform the successful implementation going forward. So I think we have time for a couple questions I did see a hand earlier from Emma. Yeah, that was me. I just have a quick question I guess it'll probably be more for you Amanda. If the learning network, are you compiling a list of all of the vendors that are displaying accessibility metadata right now. I know that red shelf and demarc do but I haven't been able to get like a really concrete list of of who's displaying it and exactly how I think the publishers really want when they ask they want screenshots and yeah, something that they can follow. Yeah, I remember it's on our list of resources to build for the PLN, and I'm really excited to see it. It's, it has not yet been created but it is, it will exist, even if it only has three places on it if all it says is red shelf demarc and catalyst well and at least it says those four places. Awesome. Okay, thank you so much. Hopefully it says way more than that let's all just cross our fingers that I'm going to have a million things but that's a good point. Thanks Rachel in the chat points out vital source sorry vital source I slipped up. Any more questions feel free to raise your hand. Here we have. Hello. Hi, I'm, can you hear me. Yeah. Hi, I'm Monica hello I've had the accessible books consortium which has been mentioned several times. In the last couple of days, I just wanted to respond to Megan who had suggested that we were conducting a study on accessibility metadata where we're not at the moment. We're quite concerned about accessibility metadata given that we have a huge catalog of titles from over 100 libraries for the blind around the world. Owen, who sits on the ABC board is involved in a study for accessibility metadata with Carl, the Canadian Association of Research Libraries I think is the acronym, as well as oral which is the American research. American research libraries is conducting a metadata study. So if you Megan if you wanted to get in touch with her I can introduce you if you wish if you don't already know her. She's also she also sits on the if law libraries with print disabilities section. She's involved there. Yeah, perfect. Yeah, that's great. Thank you so much. Thank you. That's great to know about I love having everybody in one spot, you know, we can share information. So before we wrap up any other any final thoughts from our panelists or any more questions from the rest of our participants. Well I will take a moment just to say that I work with a lot of publishers and I would say every new publisher I've worked with whether it's someone new to staff or someone I'm training has expressed knowledge of and interest in starting to share the accessibility data so this is on publishers mind I think it can be easy to like publishers don't send this but we're working on it we're getting there we're going to send it. Sure because it also means for publisher to share their effort for accessibility, and since they are already doing it, only also being able to marketing it to end user is super important, not only for disabled users but for everyone agreed. And how do publishers receive feedback about their accessibility efforts, once they've put something out into the world like we heard from Kai. Sometimes it's not good. How, how are some ways that users can communicate back to publishers. Well, usually in the metadata. The publisher can put a link to a specific page for with contact information or a page to the certifier so through the certifier or directly to the publisher. Using the metadata, the user can reach the publisher, if that metadata is present. And for users with slightly less technical ability or understanding, you know, if you can get to their website and fill out the contact us form publishers will respond they do care they want they want to fix this problem. So just contact them. They'll, they'll, they might not fix it right away but you'll put a seat in their mind. We often get questions that we have to forward on to publishers or put campuses and in contact with publishers, you know, good feedback, bad feedback otherwise. So, from our end it's always a good idea to know who to point them to, you know, because if they want to give feedback we want to be able to allow them to do that. Now that's nice to hear that you know we can keep the communication open and that publishers really do want to do right. So, that's, that's encouraging to hear. Great, well I think I don't know if they're going to just whisk us back into the main room automatically or. Okay, great, they're closing the breakout rooms now and everybody can come back into the main room so thank you so much everybody this was really great. Thanks everybody. Thanks Marisa. Thanks everyone. Bye.