 Hi, so my name is Matt Zimmerman. I'm an online platform manager Whenever give a talk, I always have to do this one part in the beginning where I say I work for the spring or publishing company And some people know and some people don't but there are two Publishers called spring or there's a very big German publisher called springer nature, which probably most people know of they used to be called springer Verlag There's a very small Publisher in New York that I work at called the spring or publishing company and we're totally separate from them We're founded by a member of this springer family, but There's no connection. So We have about 16 police. We're primarily a textbook publisher for Advanced practice nursing which are people who were going to school to become a nurse practitioner or to get their doctor of nursing practice and we also publish in the graduate level for Counseling and psychology we do some test prep books. We do some journals, but we don't make most of our money off of journals and Originally, I was going to say I'm talking today about using annotation In a way that doesn't really create knowledge that it's more of a tool, but the more I've been hearing other talks I think actually I accidentally discovered some Things and created some knowledge even though it's just trying to use it as a workflow tool So let me tell you what we were doing at springer when I started there three years ago We didn't have our own online platform We had our journals on a third-party in Jenta and we sold some ebooks through Kindle Apple iBook and you know some licensing through EBSCO and things like that if anyone else here works in college textbook World print is still where we make most of our money Students like like print books, but more and more there's at least an expectation that there's going to be a digital component That comes with the book and also we want it to Start trying to maybe even sell some very book collections into libraries. So we started our own I mean we build a new platform where we're putting both our books and journals it was it's on a platform called Scolaris which is Made by company called Semantica, which is actually two months after we signed with them. They were bought by high wire So we're on a high-wire platform, but not the typical high-wire platform that Jennifer was talking about earlier And so the issue we're facing is we had never had our books on line before we were doing them in HTML and PDF And we were actually Doing them at the chapter level. I am kind of proud of the fact that we have DOIs for all of our chapters We have keywords for all our chapters. We have abstracts Which is pretty neat The problem was that you know, we had we had these converted into bits XML We knew the XML was fine. It was well-formed and valid But we really had no idea what these books would look like on a platform Especially since we were doing back issues back based to work in journals back titles Which if you again if you work in the book world books are just weird sometimes some Times there's chapters sometimes there's parts and chapters sometimes there's units parts and chapters They don't have that kind of consistent format that journals do especially if that book was made not with the thought of ever going online, so we We really had no idea what they were gonna look like. So I was at the high-wire meeting two years ago and Heather was giving a presentation on Hypothesis and I thought oh, I said this would be a great way for us to go through and proof and QA our books instead of having Typically what we do is proofreader or copy editor would would look at the book and they would just record stuff in a spreadsheet Now I didn't need these books were printed. They'd already been sold. They didn't need the detailed copy editing a book would do I really just need it people to look at them and say Or all the chapter showing you know or the image is showing. Okay, are there any weird tables? You know basically do they look okay? So Yeah, so the idea was we would We had some freelance proofreaders and I set them up with we had a major private group and I let them go through and mark up the books now I first I thought wow I'm really clever for having this idea but apparently people have been doing this for a long time, you know, I Don't know if you knew this but before computers proofreaders I go through and mark things up and even they used to before the web, but they would do it too. So But really there wasn't a great way. I think to do it in HTML before I think Hypothesis really great for this so So the results the Editors the proofreaders they loved it because they could just go through and bump them mark tons of stuff up They didn't have to be writing things down They didn't have to write you know in the third paragraph of you know chapter three They loved it a little bit too much 150 books after we did 150 books there was 10,000 annotations So is Matthias still here from Norton? He had said when they proof their ebooks they get a thousand per book, which I think is pretty crazy a thousand Urrs or something so this was still a lot so this is I mean this is where I made the mistake and I have to say I haven't really we haven't been using it for about a year because One we had this big influx of books 400 books since then we just do like 10 books a month So We've been going back to doing it the old way with the spreadsheet But I want to keep doing it and I think there's just a few things that I need to do to make it work And also This is also a call if there's other people here who are doing this or who would like to do it I'd be happy to form a little Group or something to talk about this because I think it can be used in all sorts of you know, not just QA But proofing part of the production process, you know a lot of different ways So the first mistake I made was Each chapter in our books is its own web page and the book title isn't in the The web page title so when they were going through and marking these things up and I went back into hypothesis I really didn't have it. I didn't have a good way to know I Couldn't really quickly say show me all the ones in this one book We could probably fix that easily next time by just having them tag it with the author name or something like that But you know the horse that already left the bar left the barn on that one Also, I probably should have had them have some sort of controlled list of was this a typo Was this a dead link? Was this a missing image so I could easily? categorize them, but I Really didn't have a I Didn't tell them and do that either so really what we ended up with was these 10,000 annotations What I did was our third-party type setting vendor I just had them look at each page and if it was something they thought they could fix they would comment on it and say We're fixing this in the XML if they looked at it and said now. This is probably not an XML issue. It's probably Style sheet or it's probably actually something wrong with the platform. I had them flag it for review And I would I would get emails saying, you know, these ones are flagged for review That's why I knew to go back in and look at them and decide. Yeah, maybe that was something I needed to talk to hi-wire about So what would make this easier I think When the proofing is done I probably ideally would like to just export all this out just into a spreadsheet so that I can just have a really quick High-level overview of you know as a particular book that needs a lot of work other particular issues, you know, I really don't want to interacting with the Annotations on a page-by-page level really isn't useful for this. So I think that's step one And then I guess the next question is after that You know could the wrestler does the rest of the workflow does it have to happen in the annotation tool? I mean is annotations will possibly really good just for the proof readers and after that Maybe I need to use the spreadsheet and dump it into some issue management thing or Or even just use the spreadsheet as a way to To manage a workflow, but it's still like to look at ways I think that the workflow could some small changes that would make it easier So one would be if I was going through and triaging this annotations at least currently I can't go in and tag Someone else's annotations. So if this proof reader Made some comment if I could just go in and easily tag it and say either No, this isn't an issue or or this is an image issue like away from me to categorize it. That would be good and then also when the Type setter and XML people are done If there was a way that they could easily notify me that hey, this is done This is ready for you to look at and check again, it could be that they could just add a tag to the existing Annotation saying complete or ready for review And then this one's not a hundred percent needed, but it would be good if there was some final thing I could do to sort of close out the Annotation and say yeah, this one's fixed. This one doesn't need to be looked at anymore. So But yeah, I think that the most important thing is being able to do some export and then these other things might Solve some of the issues. So Let me just show you what our site looks like and some of the Annotations let's say I really like this site. So I like showing it off. So we have this is one of our most recent books we released and You can see we have all all the chapters I really like this because the chapter this isn't generated from a Table contents in the book. It's actually generated from all the actual chapters in the in the XML We don't have a lot of people who come here and buy these books online But we do have we know our typical our motto is that we put a scratch off code in our print books And then people come here and get access But we also sell them at the chapter level and we actually do get a lot of chapter sales So this is what a Chapter would look like We have this nice focus view. I really I know this has nothing to do with Hypothesis but or annotation, but I really like Like how the site like looks So so that's just what most of the site looks like this so an example of some annotations that had Happened so this was I say so this person Yeah, so typical things like these there were link there were HTML there were hyperlinks in the text, but they weren't actually hyperlinked and And some styling issues like some things were italicized that weren't supposed to be and things like that So I think again, I think the next steps would be I Would like to pick this up again So I'm wondering you know was maybe this just an issue of scale like 200 We're never gonna have to do 400 bucks at once again So maybe doing 10 a month will be a lot easier and I won't get overwhelmed by 10,000 annotations But I do I do I would like to keep working on it and try to You know tweaks on these things to make it more usable. So if I have all sorts of join me, please do Oh, and I said I didn't think this had been a I Didn't think this had been a knowledge that I wasn't developing new knowledge by doing this But it was really interesting going having people go through and look at all of our Look at all of our chapters So we discovered a lot of things like a lot of things that just really don't work Well online like a lot of big tables and books. So now we've you know, we've come up with a lot of best practices Even though we're still mainly print, you know, we're trying to convince people not to put But they really need to have big tables Just include them online just include them as a spreadsheet, you know, we can we can add a lot of supplementary material online One really interesting thing and I had an example and then I closed my browser is When you when you write a book you have a title that's like Radiology and then you have chapters like foot chest arm leg but when you have about 50 different books like that you end up with a 50 chapters called foot or a bunch of chapters called Chest pain a lot of similarly how I'm sure a lot of your journals have a thousand articles called book review or Editorial or something like that. So we've we've instituted a new best practice where when we name chapters We need to have the chapters be more descriptive because they're going to start to get Discovered on their own. They're going to get purchased on their own. So we need to start thinking of our chapters as Bits of information that can stand stand on their own. So Yeah, that's it. So Questions from out. Oh good. Um, this is great. Really interesting. Uh, quick question. You're talking about the The annotation workflow for the editors. Yes. Was that happening on like a like an open URL? Was that in some kind of closed environment? Like, oh That's a really good question. We had a staging site that the That highway had for us So we were doing it on there, which I know one of the the composition Professors the other day was talking about how when websites go away That that was one of the problems because they that staging site went away and so we really don't have any record of you know, that's a bunch of orphan But so it's a good question because right now we don't really have a staging site. So we have been You know, we load the book. It's in this embargoed State but lots of times we do just actually make it live and figure well no one's gonna Find it We can probably QA it in a short amount of time, but yeah, ideally we We're hoping we can work with Highwire to get a staging site because yeah, that was one of the issues. Yeah