 I don't know. I live in Amish country and I find it so refreshing sometimes to go Amish as I call it when we have a storm and power goes out. Well, my kids are in their 20s now, but my youngest lives at home and he comes rushing out of his room, right? What do I do? Yeah, well, again, thank you very much for that presentation. That was extremely informative and made me think about a lot of things. In addition to my work with the diary of 1817, and that was one thing I really wanted to do is try to discover the emotive aspect in that writing. In a structure, you know, diary entries are very structured, very uniform, but every now and again, the diarist would go on a bent and create some language that was out of the ordinary. That was structurally very interesting. But your study cast a completely new light and a whole new arena of possibility for interpreting text and having such a large corpora is wonderful. I think we all owe you a great deal of gratitude for getting these. It's not just me. It's a team. I couldn't do this by myself. And, you know, my student, Michael, is the person who built the transcription desk. He's done a lot of the programming to link the database metadata to the actual transcriptions. And, you know, we have the part of the platform that I showed you of Moravian Lives is actually the older part. What we're working on now is not yet public, but it's the entity creation so that we can create, we have created a whole new flexible database that links, you can search for people, you can see where people went. We have events starting and finishing. So in the Leibnizlav, as we know, we have people saying, I went there, I went there, I went there. And the way in which all of this has been encoded is to show us then the journeys of people and then who they intersected with, when they were at certain places, etc. So it's very exciting. And so that's all coming up soon. We haven't gone public with that yet. When do you think you will make it public, Heidi? Well, the database, I'll find out in a couple of weeks. The database, the structure has been created, the data structure. Now, we're feeding in the encoded memoirs. The corpus that we've started with is actually the corpus from full neck in Yorkshire, because that is a very complete corpus. And it's something that we can, when you talk about Leibnizlav, there's so many of them. And that's the problem, where do you start, where do you finish, etc. So we're actually, we've encoded all of the women, single women's memoirs in full neck, which have never even been cataloged, let alone digitized. And so that's why I've brought up Anna, in my talk, because she's sitting there in 1760, and she doesn't know that she's going to die in full neck and be buried in the same graveyard that she's hearing about. So really cool connections. And in some of the other work with the encoding, what we've been able to do is, for example, with the catalogue of baptized Indians, that was Hecoberdas, actually, at Wines, and Hecoberdas. We've digitized that, and we've actually created an interactive network visualization, so you can see how the different nations, how evangelization happened through the different nations. But anyway, so those are some of the examples of things that we're working on right now. Fantastic. I was struck that part of our discussion for this week needs to be the changing vocabulary and becoming American. When you put up those word things, it struck me that the Bethlehem folks used the word sinner was popping up. And that wasn't in the European section. I can hear the ghost of Jonathan Edwards, sinners in the hands of an angry God, perhaps the neighborhood environment, but other folks and other flavors of Christianity speaking more about sinners in the American context. Well, Eric, if you had a corpus of digitized texts, you could very easily look for that kind of thing. Well, let me speak to that because it's something that I, one of our goals is to have more of our materials digitized. One of the reasons that Reverend Dr. Curtin Roberts could talk about her experiences of the Caribbean context was because those records are held and have been digitized at Bethlehem, thanks to a couple of grants that they've come in. And we are trying to implement a process by which we can review and sort of triage our items to digitize. And hopefully we'll be joining that conversation in the years ahead. Unfortunately, our staffing is very tight. Our budget is tight. We're dependent on volunteer labor. The tradition of our place has been more heavily into translation of our materials rather than a digitization of those. And we have translated a good bit of excerpts, but as we were talking about this week, the purposes for which you translate or to publicize certain parts of your record changes depending on your research purposes. And I think part of our reason we want to have in part for this conversation is to broaden and let people think about the opportunities locked within these records as we let those be more accessible. And you have other questions, ask of the material like you're doing. And I appreciate that. And I will tell you as a tease that we have actually 14,485 memoirs that have been scanned. 3,000 of them were for persons who died before 1865. And the reason I point that out is because that would mean they died before the end of slavery. 1474 persons were born before 1822 in our collections. And that's the date of the last year of our conference. So there is a certain research here and an even more tantalizing part. We have about 500 memoirs untranslated from German because we depend on translation based on who will come in as a family member and ask their memoir to be translated. Eric, I am going to jump in there and I am going to say this to you. There is no money in Yorkshire for digitization, right? They have no money whatsoever. So hold on a minute. I need to take my headphones off. I'm going to show you something. So this is what we do. We get one of these. Sorry, my there you go. Kind of in and out. Yes, yes. All right. So it's a scam tent and volunteers go into the archives with the scam tent and you can create a really good light environment and a regular regularized focal distance for the memoirs that go into the scam tent and you just put your phone on top and take the pictures. And so that's how you can take say 20 people with scam tents could digitize everything for free. And one of our challenges too is to see we're organized a little bit different from our friends in Bethlehem who are a separate 501.2.3. We are a church agency and our collection goals have been somewhat different. And what we're having part of this discussion is to try to see how we might expand and get colleagues to help us better make appreciated and accessible the records that we do have. So this talk is very informative as I'll share that with other members on our board. So thanks for showing the visual. No, that's fine. I'm sorry. It was so blurry, but that's what happens with virtual reality. Yeah. Yeah. But thank you, Katie. And I want to build on Katie and also make an offer to you, Eric, from our end. And we have a 3D printer that can print the stands for our phones to take pictures in a very stable fashion. So we would be happy to donate those 3D printed stands for every volunteer's cell phone to use. Part of our thinking is we have to figure out how and this is the question you think about it in the digital age. How many of you have a subscription to your local newspaper? And part of our funding model for support is that we are dependent upon the support of people to support our genealogical records. And we need to figure out a model by which we can add access to digital things and still have an ability to fund things. And if you talk to the folks at Bethlehem, they're still trying to figure out how the digitization process will affect the number of people who actually come to their place, who will be aware of their physical place, who might support their physical place. So it's all a negotiation that every institution that holds records has is that balance between access and support. Indeed. I just had a great email yesterday from a PhD student at Penn State who was asking me a specific question about my research and her research. And at the end of it, she said, thank you so much for Moravian lives. It saved my life during COVID. And so, and of course, we never thought of that right when we created the platform. But people have been able to use the platform even with the limited number of memoirs we have available on that. And the other thing is that, you know, again, I talked about citizen scientists, right, who are transcribing on the transcription desk. The platform, the project has actually moved into using machine learning and artificial intelligence to read German script through transcribers. And that's where I got the scam tent from was from that initiative through the European Union. And that but citizen scientists can still do that kind of work. And yes, it takes checking. Yes, you need to do that. But all of that genealogical interest also spurs people to go to find the memoir of someone, you know, an ancestor and to transcribe it not just for themselves or for the archives, but also for other users as well. We have some good questions that look back to the grant you want to get to those? Yeah, let's let's get some of these questions that are popped up. And for the audience that is watching, the way we do this is through the Q&A. And if you drag your cursor to the bottom of the zoom screen, you will see that icon Q&A go ahead and click on that and then start inserting your your question. So we have already five in there. And we have one from Paul Poeker. So we'll start with Paul's question first. And he asked, does the fact whether a memoir is autobiographical or not influence the emotion score? The emotion score? Okay, so I think there's two ways of thinking about this. When I did my first collection of memoirs from Bethlehem, I made it very clear that I wanted first person memoirs, because they might be more authentic than third person singular memoirs. And I think that for somebody who thinks about authenticity in an autobiographical document, as I argued in my talk, that would seem to be given, right, that you will go to the first person narrative. But I think we have to be careful when we look at the autobiographies, if we can find them, of marginalized peoples, of peoples who were historically disenfranchised and didn't know voice, because even if you can find the extremely rare document which is written in the first person, it doesn't necessarily mean to say that that is an authentic expression. And again, I talked about code switching, I talked about languages being being used and rhetoric being used if you'd like in order to gain entry into a social group or in this case a religious group. And I think that's very true with the memoir, that especially for groups, for whom otherwise there is not even the idea that Africans have souls in the 18th century, right? I mean, this is in question among the European Enlightenment. So that even if you have the idea that there is a memoir written in the first person, that person knows that there are expectations about their identity that they have to fulfill in order to be accepted into that group. And it might be gratitude, it might be, you know, I want to join the Moravians because they make me feel good, right? Is that really how the person feels, right? And that's why I think that going to use the LIWC is so helpful to, because you can use multiple tools, it's so helpful because you can look at different subconscious ways in which we use language to be able to perhaps augment what we read on the surface, right? I think as close readers, we're very good at reading irony. Computers are not good at reading irony. So what's the way in which you could computationally do that? And I think using LIWC is a great way of being able to detect whether someone is making themselves vulnerable, whether someone is making themselves honest, right? And the study, the actually the algorithm that was developed for authenticity was based on making people deceive as a listener. So that's how they were able to create the algorithm was to say, make me believe X, Y, and Z. What kind of language do you use? And then that's how you detect the mechanisms that we use. Does that help? So yeah, thank you, Katie. I want to build both on Paul's question and your answer was a follow-up question. And when I listened to your section about James Pennebaker's model, I was very curious how much cross-cultural competency went into establishing these categories, because for example, when he says that when the past is being evoked, it seems to evoke sadness, but we heard from Dr. Curtin Roberts last night how important ancestors are in creating identity and then creating a viable spirituality. And so invoking the ancestors is not a sad posture to take. I just agree with you actually. I think that when we think about what it means to look to the past, what it means to look to the ancestors, it may not be sad, but it may be kind of melancholy, that there is of course the idea of nostalgia, which is this very romantic concept of a desire to return to one's beginnings. And so the ancestors do guide us. The ancestors are there as voices that we use to guide us in our present and in our future. But it is also not necessarily something which is full of hope, I would say. That is not my understanding of the way ancestors function. So I would question what you just said, but I just want to put it out there. Well, I think African descent, people can speak to that better than I can for sure. And I would say it's also relates to the question of indigenous thinking about past and future as well. So it's just a question. Well, it's a question of time. And of course, time is a very different concept in different cultures, right? Exactly. And so you can have the idea of there being an eternal presence of all, so that you have the ancestors who speak to you in a present, and you have those who are yet to be born speaking to you as well. I mean, in the Ebo tradition, in Ebo culture, that Andrew belongs to, that is very much the case. There is those who are yet to be born who will speak to you as well as the ancestors. But I mean to get back to your question about Pennebaker. His group, his test group at the time when he did this was American students, and that's always the problem, right? It's using, it's creating algorithms from the students you have in class and what did those students look like, where did they come from? But I teach cross-cultural communication as well. So I'm very aware of the differences between different cultures and the uses of tense, the uses of pronouns, et cetera. We have another question from Dave Bloom. He asks, is there any plan to integrate the information from the Dina Bleto to fill in the other outlines of people's lives? I don't know of any plans to do that. But what I would like to say is that what we've come to realize is that Moravian lives could start to act as a platform that is not just about the Lebensläufer, but is also there for other projects because obviously the concept of Moravian lives is quite large. It's not just about, you know, autobiographies or biographies. There are projects in Germany that would like to join this. For example, Martin Prell and Gisela Meteler have just launched a new website on the Zegerminer, on the Sea Congregations, and they are linking to the memoirs of the people who were in the particular sea voyage that they have transcribed the diary of and also are having daily releases of where they were on the ocean as they're crossing the Atlantic. So I think that, you know, if you have a project like that and you'd like to digitize something and you'd like to link it, talk to me. We also have a colleague in Germany, Lubina Marling, who has created a collection of Sorbian memoirs and she is also linking those Sorbian memoirs to the Moravian lives project as well. Wonderful. Winnell Kirtan Roberts, who talks on Friday night has a comment, maybe a question. Curious to read some of those memoirs of persons of African descent that you mentioned. Yes, they are very interesting and I've written that length on the different memoirs. Some of them, some of them haven't yet been published except in various private collections or also on Moravian lives. I would encourage you to just go to Moravian lives, go to the transcription desk, type in Black Bethlehem and then you can find the memoirs. There are some other documents linked on there as well to talk about slaves in Bethlehem. But again, you have to know German script. I cheated and looked at the question and answer myself. I hope that's okay. There's also a question there from Suzanne Lovejoy about is there anything that transcribes print text and facture? Yes, definitely. You can use transcribes for that. One of my students has actually created, actually two of my students have created two different tools. One is to read transcribers through using a regular OCR and another one, I'm sorry, to read facture through a regular OCR. And the other one is transcribers, which you have to train the artificial intelligence to read the print, but then it comes up with a very like 99% accuracy rate. So yes, I would advise you to go to transcribers.eu to find that. So we have another question. This is a wake for a student, Fiona Burdett. She asks, is the custom of writing Lebensläufer still practiced today in Moravian communities around the world? When did they stop being written in the old German script? Some of that question I can answer, some I will defer to Paul. So the practice of dictating a memoir has been continued in some communities. I think it has also been revived in some communities. In Helnholt, for example, there is a project to revitalize the process of writing a memoir, not just the actual, so it's not just the product that's important, but the process of getting people to sit down together and talk about their lives. And the person who's leading that project is Jill Faucht, who's the pastor in Helnholt. She has got a fantastic project, which is especially working with people who grew up in the GDR, who were not necessarily trusting of other people to tell them any personal details of their lives, that she has created a group which is designed to encourage people to write their memoir to be read before at the funeral service. And I would have to defer to Craig and Paul in terms of dates about when people stopped writing the Lebensläufer in German script. I mean, there are Lebensläufer right now in the full net collection, which are written in English. And that's another reason why we wanted to focus on that was that we didn't have a large English corpus, a Meridian English corpus. And that's in just 18th century hand, that's not German script. But in terms of when they stop being written in script, I don't know how to answer that question. I'm sorry. I don't know if one of our archivists can... Maybe they'll chime in here. Yes, help me out though. So we don't have... I know that locally, the formal end of the German script use was in 1876 for church records, but pastors were writing in English for memoirs as early as the 40s and 50s. And people in their congregations were, one of our congregations was an English speaking congregation from its founding in the 18th century. But there was a by the third generation, the Meradians in this area, certainly English was starting to be used more and more in church records and documents. I had a question about the artificial intelligence and its ability to decipher the German script. I've encountered some German script in particular. Anything that is older than 1780 frequently has a lot of ink stains. Often there are pages missing and things like that. How good is this artificial intelligence in deciphering some of those older handwritten documents that are, for me, who can read German script are pretty still pretty difficult to get through. So the machine knows nothing, right? The machine only knows what you teach it. And so what you have to do, which is quite painstaking, is to create training models of diplomatic transcription of a set of documents or of a document. So that might be up to 70 pages of manuscript or 30 pages. And then, so what you do is you feed that into, I mean, this is very rough, but the machine, you feed that into the machine. You tell the machine where to look. So you, the brilliance of this machine, or the person who designed it is that you tell it which fields on the image it should read as corresponding to what you have transcribed. And so if you have a wine stain, and don't we all spill wine or coffee on our stuff I did today, actually. So it's not just an 18th century problem. You just tell it not to read that, right? You say ignore this field, right? Or what you do with marginalia, right? How do you deal with marginalia? How do you deal with footnotes? So there are ways in which you demarcate on the field of the image exactly which parts of that image correspond to what it is that has been transcribed. Another issue with memoirs, of course, is that you don't have all of one handwriting. You have many different handwritings. And so what we have had to create are multiple models to read different sorts of handwriting. And one of my students this year has created what I call a digital sorting hat, which is you, it looks at all the different styles of handwriting. And whereas I'm sure Eric and Paul and Tom say, oh, I know who's handwriting that is. I'm not that clever. But the machine can say this handwriting is like this handwriting. And so we can actually identify groups of handwriting and then describe, which usually is going to be the choir helper of that particular choir, usually, right? So that has been extremely helpful in improving accuracy. We have a character error rate of around 99% now, which means we've been able to transcribe enormous amounts, half a million words in the last in the last 12 months. Yeah. Looks like we have another question come in. On lineback, does your analytical machine distinguish between the first part written by the person and the final part written by a relative or friend? That is a great question. And this is something that drives my computer science colleagues up the wall because they think, oh, why is it making so many mistakes, right? And then they realize that halfway through the memoir or the end of the memoir, you have a different hand. And so you actually have to separate out if you're training the machine, you have to separate that part out or and say, don't read that. All right, because it's going to mess up your results. But what you can do then is to say, okay, which handwriting is this like, and then put it into the machine with that particular training model. So the other thing to bear in mind is when the machine is reading, it doesn't understand a word, right? It just sees shapes. It sees, is this dark or is this light? It's a totally binary operation. And so then what happens is the the transcriptions, the machine transcriptions are exported to a web interface. And I have several students, wonderful students, who then can go through and check on the web interface, the machine transcription against the original. Now, obviously, they too have gone through their training, right? In the beginning, they didn't know Moravianisms like I talked about in my talk, you know, they didn't know what a sick waiter was, they didn't know what an overseer was, a room overseer, right, or something like that, or all of the wonderful abbreviations that Moravian texts have, you know. So they've worked their way through those two. And now they're used to reading those things. So and these are all undergraduate students. These are not graduate students, right? And I've been fortunate enough to get grants from my Humanities Center at Bucknell to pay for that kind of student research over the last few years. So but that's a great question. Yeah. But nobody's asking about the memoirs, you're asking about the machines. Yeah. Grant, can you bring Tom into the fold? He would like to answer one of the questions that came up earlier. Can you do that? I can't. Well, let's see, maybe I think I can. Let me click out of some things here and try to get to this window. Okay, so he's allowed to talk now. Hey, Tom. Can you hear me? Yeah. Hey, well, thank you so much for the presentation today. I've really enjoyed this and I'm excited about your ongoing work. I just wanted to say that locally at the Moravian Archives in Bethlehem, we've been also piloting or testing some work with transcribes and are really excited about its possibilities. I'm sorry I'm in a house with my parents and the dogs are barking, but I just wanted to say that I really appreciate your presentation. And okay, sorry. Bye bye. I'm not sure the machine can transcribe dog barks. And we really needed to hear those dogs, because you know, when you do zoom meetings, you always have a cat or a dog or a parrot show up sooner and later. And our furry and feathery friends are sorely missed at the conference, I have to say, for those of us who are loving their animals. Okay, so let's see there. Oh, there's a very, very long comment. I think we don't have time for that. There is a question about Magdalena. If you maybe if you make this, if this okay, if this okay, Grant and Eric, because I look at my watch. If you could make the question about Magdalena, the final question of our session, would that be okay with everybody? Yeah, we officially going over by the minute almost. So I think that sounds great. And of course, the last session ran a little late and we planned on going over a little bit on this one. But yeah, we should be able to go home. So let's answer this question. Okay, let's honor Magdalena here. So yeah. So actually, I had a quick look at the long question too. And I think I can fold in and answer as well. One of so the quick answer to the Magdalena question is how Labenslauf is included in my book, Moravian's Women's Memoir. So you can read that there. Also, if you go to the Moravian Lives Transcription Desk, you can find both the German and the English of her mother as well. To answer very briefly the other question, the whole point of Moravian Lives is to break down barriers between the inside and the outside of the academy. That it's very important to me to have this as a to be able to offer this kind of tool on the web, right? You don't have to log in with credentials from an educational institution. Everybody can use it. And I think that's really important because obviously I hope I didn't bore everyone to tears with my talk, but obviously the Moravian memoirs are a treasure trove. They are an amazing way to access the lives of the people who came before us. And although the memoirs may not be full of adventures and the kind of juicy details that we would want in a bestseller memoir of today, those documents constitute, as Sinsendorf said, the universal church. And that's why it's so important to find and translate and make available the lives of people whose lives were negated and in many ways still are being negated today, right? It's very important that we can read Magdalena's story. It's very important that we can read Andrew's and Peter's story. And it may change the way in which we think about the history of the United States and hopefully do something about the future, too. So to my mind, I'm very much somebody who wants to make the walls of the academy porous. Yes, I agree. If there were a way to end a session, that would be a beautiful way to end a session. But that was very well said. Katie, I don't know if you want to take a stab at that long question. Well, that was sort of my answer to the long question. Well, good. I think that's a great place to end things here. Ulrika, did you want to make any final statements? So thank you very briefly, Katie. I so appreciate your last comment because what you just said is really the foundation for the concept for this conference that we wanted to create a town and gown opportunity and Zoom allowed us to really be that open and to share the work in our guild with everybody else. And so thank you very much for that. And thank you so much for bringing us into the 21st century with your talk about artificial intelligence and these different ways in which we can make those archives more public, more open, more accessible, and undoing the silences of so many lives that were touched by the Moravians. And in many ways, the Moravian story is either the story of this country. And so my final comment is to my audience and my co-conveners and our wonderful, amazing colleagues. I thank you all so much for being with us on this journey. It was a hard journey. I think I can speak for Grant and Eric that we are all a little bit exhausted, but we are also very, very accelerated by the incredible fortitude of all of you to stay with us. And none of us had any idea whether this was work. We are the first virtual conference at Wake Forest University. We had an amazing tech team. I'm just so grateful for all of you. And so I'm speaking again from the heart with Katie legitimized by bringing out the issue of sentiments. And I am all for that, to have that holistic academic work as well, you know, that the head and the heart connects through an emphasis on ethics and holistic definition of what it means to be a person. And Cynthia Villa Gomez last night also stressed that as I think did so many of other, other colleagues. So a big, big thank you to all of us. The show will go on after this conference. We hope we will stay all in touch with each other. We have the Moravian studies collaborative coming up. The Renauder Museum of American Art will open shortly. Our next volume of the records of the Moravians with the story of the Cherokee removal will come out in just a few days actually. We have our next generation waiting to take over and continue our work. And so I think that is a word of hope and joy in terms of our sentiments. And I'm going on way too long. It's my it's a bad habit I have. So I will just give it over to Grandin McAllister. Thank you so much and Auf Wiedersehen. Thank you, Ulrika. Eric, did you want to have any final words? It's time for lunch. I appreciate that. You all have been munching. We probably have. And Katie, I appreciate you joining us from Germany or your Germany today is where you are. Isn't that right? England. England, okay. I know you've been shuttling back and forth, but I know you're several hours ahead of us and I appreciate you staying with us for this conference. Wonderful presentation. Thanks all of you. We are physically not open at the archives, but we are available online, moravianarchives.org. And by email moravianarchives.mcsp.org. We'd love to have you in our conversations going forward. Thank you. And I'd also like to thank everyone, all of the participation, the public participation, the audience has been wonderful. We've had great attendance at all of the events. And that is really the goal of this town and gown conference to bring academic work to the public to show what kind of resources the public has for for knowledge right here in Winston Salem. There are a lot of institutions working together, generating stores of knowledge. And our goal, like Katie said, is to make all of this porous. We want to give access and that is the main goal of this conference. So without your participation, it wouldn't have been complete, but it is now complete. Thank you very much. I'd also like to say that we will be posting videos of these Zoom webinars. So if you want to go back and listen to something, if you missed something, if you want to share it with someone, then please do so. We will post all of these videos, all of these webinars on the video page. And what we can do is continue to send out reminders through the registration list so that you can continue to receive links to the video page and continue to see what videos have been posted. And we'll try to leave those up for at least a week and potentially longer. And again, please keep in touch with the archives here. And because there are a lot of events going on that have something to do with the Moravian history and culture. So with that, we will close our conference. And thank you very much. And I'll feed us in. Feed us in. Well, cheerio. Yeah. Cheers. Cheers. Cheers. Cheers. Cheers. Cheers.