 In April, April. Oh, can you hear me? Oh, okay. I was like, oh my goodness. The net first. Sorry. Yeah. No, but welcome to another ADHC talk. It is April 14th, which means that we are wrapping up the semester, spring semester, and we are all feeling it pretty hardcore, but I am really excited today to present Jerry Rohingya. And we are going to talk about the future of DH, future of DH in higher education, the future age of DH scholarship, just sort of all things, future of DH. So yeah, Jerry, I'm going to start off with a question. How did you get here in your DH journey? Oh, I forgot to read your email. That's okay. So I'll introduce myself then just to keep that in there. So I'm Jerry Rohingya. I'm an assistant professor in religious studies. I'm the DH higher in the department. And I am the incoming assistant director for the Center of Digital Humanities at Princeton University and the Princeton University Library. So this is actually my last hurrah last few weeks here at at UA. So I'm glad to get to do this before before departing. My research looks at the intersection of kind of critical data studies, American religious history, gender and technology. I studied Seventh Day Adventists in my dissertation using computational text analysis, but I've kind of moved into these bigger questions of of what is DH? How do we do computational research in history? How do we do it in the humanities? And what does that mean for knowledge production? So how did I get here? By a very wandering path. I can kind of happenstance. I am not somebody who grew up programming. I'm not somebody who kind of was early adopter on these things beyond Oregon Trail and like Microsoft Paint. So good Oregon Trail generation person here. I got here because at the end of my BA in philosophy, I realized I wanted to continue in the humanities, but I wanted to focus on works that engaged outside of academia. So I went to Yale Divinity for my master's degree and really realized that I wanted to engage outside of academia and discovered that the world of kind of public history and digital public history. So I took a public humanities course at the graduate school there in around 2010. Got started to think about digital projects started experimenting with some computational text analysis. In my Divinity School courses they weren't entirely sure what to do with that, but it was a good it was a good effort from on all of our parts. So that led me to George Mason University which has one of the leading digital history centers, digital public history centers. And so I'm one of the early generations of scholars who are actually trained in digital methods because the history graduate program there has required digital history courses. And I was also able to do a minor field and what we called the history of new media thinking about what is technology what does that mean for the production of history. How do we engage with media studies how do we gauge and kind of critical code studies, which was just just starting. So I got a lot of experience there on the public side of things working with digital humanities now, which I think no longer exists journal digital humanities which also no longer exists. But also the software omega, which does still exist and is excellent. And then 2015 I started working as full time administrative faculty at the George Mason University Libraries. I was the digital publishing production lead, which was a mouthful to say and I'm personful to do. I spent split into multiple roles but during that time I worked with open journal systems for the library, our institutional repository on d space and kind of thinking about what a digital scholarship center would look like for the library. So my training up to this point I finally finished my PhD in 2019 and then started here trying to do that within the role of kind of a disciplinary scholar and religious studies. I'm just wondering path to hear lots of different experiences lots of different types of DH experiences are so rich and give you such an expansive understanding of what a DH practitioner actually needs. And how to how to get there. Right, which I think yes, is, is not always. Not always all of our paths. You know, so, so very like your very curiosity driven, I think, and also opportunities that came your way. And you have a very, very broad understanding of sort of conceptually what DH. Yeah, and right and how it looks when you're in different subject positions right so what does it look like as the graduate student what does it look like when you're within a administrative faculty position in the library. It's more of the service support role what does it look like from a disciplinary faculty member. And so yeah, it's been, and it's kind of a nice thing about being in that, it's probably say third generation of DH scholars, and that it's still undefined we're all kind of learning as we're going but it means that you get to try, and you get to sit in different places. And I think we're still there but there are a few more kind of direct routes and so it's easier for people to specialize in some ways. But I think all of that that experience helps inform how what how I understand DH and how I understand different. So I understand DH to be the critical application of computational methods with tools within humanities research, as well as the critical development of computational systems informed by the research priorities of the humanities. I understand it to be both and both at the same time, but the different aspects are emphasized in different areas. So what I've learned in in being in the disciplinary space and both in my grad school work and in the faculty position is that the emphasis is much more on the critical application of the methods to the content work. So what I found about myself is my interest is in the critical development of those systems informed by the humanities work. And right now because DH isn't formally a field in the US context it is in Europe, and I'm kind of jealous. Because it isn't we end up doing that work and research centers like what I'm going to or in libraries. Because it doesn't fit into the disciplinary structures that that we currently have. Right. Right. So, we've had some conversations about sort of the future of DH, and you are super interested in that sort of development of computational methods and structures. What do you think are some of the big future impactors that we're looking sort of, we're looking down the road. I mean, it's been five years and years, even next year. I mean, some really big technology things have happened in the past year. Right. So like, what, what do we have going on there in your, your vision. Yeah, so my sense of kind of top top priorities are top things that are going to be productive research areas in DH but also primary concerns are around data science and machine learning. They more around education around infrastructure first and then also education, how we train people to do DH work. So thinking about data science and machine learning. The launch of chat GPT turned everybody's world upside down but this is actually something that's been in development here for 10 plus years. So, chat GPT is an example of large language models this idea of training neural networks to be able to do predictive work with text but kind of getting to this idea of structurally understand understanding is loaded but recognizing patterns in the structures of text. So long conversation about generalized intelligence and whether or not that is actually what we're doing here as a community scholar I would say no, but this is going on the internet so I should probably not not put myself out to too hard here. So this is an area that's active research has been a lot of pushback. For years from scholars such as Emily bender to Nick Peru, Sophia noble drawing attention to the, the limitations the biases the costs of these systems the potential use of them to further disadvantage populations like these huge problem set. DH has been kind of relatively late to those conversation. And which is unfortunate because I think we need to be right in the middle of it not just in the limitations of the technology but in the conversations about how to shape the development of data science as data science methods and machine learning methods are becoming kind of the standard epistemological framework so science is increasingly data science data driven humanities are increasingly data driven but what does that mean, and how do we do data work in the humanities that brings humanities ways of knowing that emphasize complexity, the contingency and the contextualness of data and put that at the center as opposed to thinking about trying to find like big patterns thinking of data as objective or solid or something that can be easily used to get your results and go right find your way to make money and go but how do we create epistemological practices that put that complexity that contingency and that context at the front and make people grapple with that because data is not less is not is not a more objective it's in fact even more more complicated because it's undifferentiated. And so those those kind of structural issues become even bigger once you are at data at scale. They don't go away. You can't you can't build your way out of them in that way. So unless the humanities figures out how to join that conversation at kind of that deep technical level and that deep epistemological level the types of knowledge that the humanities contribute are just going to be further devalued devalued culturally and so I think, figuring out how to engage that conversation and figuring out how to make the humanities contributions front and center is kind of the big existential crisis right now in in DH. Yeah, I think that it's just making me think of just examples of DH work that I'm seeing happen on our campus, and I'm seeing scholars who do not have formal computational training, but have very very formal humanities have a vision of doing a digital project that tells them something that they wouldn't know. But as you're talking, I'm thinking about all of the barriers that they encounter, right. There's, there's such a chasm between these big picture, big deep, big age, like world of DH, conundrums and challenges and opportunities, and then the practice of DH, which is often being done by folks who who aren't a part of those conversations, who are a part of their, their disciplinary conversations and they have a dream, right. And I think that is the real opportunity that I see for DH centers is to try to bring those two things together. But it's, it's such a challenge, right, as far as on, you know, on our campus. I know when I talk to people who are engaged in DH projects, they, they have a real need for some sort of larger structural support for what they're doing. And beyond that you're spending hours learning how to do the skills that they need in order to execute these projects. And when it comes time to put their tenure dossier together, or some sort of application that work. But the value doesn't translate. Right. So, I guess my next question for you Jerry is what infrastructure is necessary to support sustained DH scholarship. Right. What, what infrastructure is necessary on an institutional level like, how do we get there. Yeah. And I think you're exactly right to draw attention to the role of the center in this conversation because individual scholars they're going to be creating data that is contingent complex contextualized, but very related to their specific research. And it's the center that's in the position to look across multiple projects and to start to see those larger patterns in how data can be constructed because it can see across the multiple the multiple data sets. So in that sense, that sort of central place where people can come to to learn skills, can learn how to structure their data so that it can be put in conversation with other people's data there's there's so much that happens when people are just learning on their own that is to bespoke to their individual projects because they don't yet know about metadata standards, or how to structure their data so that it's reusable or what is tidy data and how to make their data work work with computational system. All that needs to be learned either directly or in collaboration. The center at Princeton does a lot of one of their programs is to do grant programs to faculty and students where you can apply to go and create data. And then you can apply to do a project with them, but they have that data step because they found that a lot of people they don't yet have the data that they need in order to build the thing that to make the vision to make their dream, which I now have her full stuck in my head so thank you for that to another mother of children. Disney all the time. So they need a way and they need to learn how to build the data in a way that it can become important, but I think it's thinking even broader how to make it count in a way that works, we have infrastructure issues in terms of publishing for digital and publishing for the different aspects of digital work so digital projects are data, which is itself an active interpretation and scholarship. They're composed of some sort of interface or analysis, which is again an active scholarship, and some sort of presentation layer some sort of way of visualizing it which is again an active scholarship so any digital project should be at least three components. But so often there aren't ways to publish it with all of them connected there aren't clear ways of making that count. There are melon is funding a number of digital publishing initiatives but the big one at Stanford that lets you kind of do your thing is ending at the end of this year they're not taking on new projects and so there's ebooks. There's no clear line through our academic publishers or even our institutional repositories for these larger digital projects to exist as kind of first order scholars scholarly goods. You can, you can hack something together, and I have and I'm happy to talk about how to do that. But it doesn't function well as as the thing that has been created and it's not clear how to reward all of those layers of work that they are not just one thing it's a living thing that's composed of multiple things. So that makes it a challenge and and and that that institutional larger disciplinary structure for it needs needs help and attention. Because if we don't know how to fit it into the scholarly ecosystem, then the work isn't going to be done because as you say how do you get it into a 10 year dossier how do you get it into a promotion dossier how do you communicate about what it's doing in a way that makes sense to those larger review committees. And until we have some way of doing that it's always this like, well you did that on the way that that's just part of your research, instead of, you know this is actually a scholarly contribution of itself. Right pounding. And we have models for that over, you know, in the sciences. Yes, building a data set is an act of scholarship in itself, and worthy of grant funding and that is the contribution, and you deposit that data set somewhere and any number of people can use it. Right, but, but it's that is, that is an end goal in itself, and I've been thinking a lot about DH in that. Through that lens of, I've been looking at a lot of the projects that we have been a part of over the past years I've only been here since October so it's been. I think I just hit my five months. I think it was as the head of the ADHD, which, you know, in some ways feels like it's been longer and some ways, it feels like I just started yesterday. And, and I think about all of these projects and in my conversations with people, like, when somebody decides to do a DH project, they are often thinking of that third level, right that presentation level. That presentation of the digital project that they are considering as, as the deliverable. And in my mind, I definitely have like, we have all of these other deliverables because you have your data set. Well, like, you can build another project out of that just as easily in, you know, years to come. And, you know, there's there's just the methodology of a process. There's the methodology of processing that data and analyzing it, you know, running it through these frameworks. New frameworks are being developed. Wouldn't it be so cool to see like a legacy of several different iterations of a data set, right. And then to these, these final presentations, but, but I think the humanities, historically have been so focused on that deliverable of a monograph and in people's brain, the, the website or the omega collection, building it out. It's kind of a Maca, but like the digital exhibit part of it is like what they think of as the equivalent of the monograph or, you know, and it's, but it, it doesn't have the same impact. It doesn't have the same weight, and it doesn't have the same lifespan. Right. So, like, how do we, how do we fix all of these things because a DH project that was built 15 years ago may very well only run on a computer that is not on a network anymore. Sitting in somebody's office because, you know, technology has has radically changed so very much. I think, I think those are things that are so present on my mind, especially as I'm looking at the way that our center is resourced and the way that we are organized and the fact that, you know, we are continually coming in, and how do you, how do you commit to maintaining them as the technology ages out, and do you keep all of those projects on a server when a, when a center considers itself more of an incubator, right, then an archive and where is it going for that. And how do you honor those projects that were built, you know, under your shepherding, while deciding what stays and what needs to migrate or archive, right, and then what happens to them after that. Absolutely. Complicated. It's such a complicated, like, conundrum to face as, as a person sitting in a center, trying to decide how best to use your resources. Yeah, yeah, you know, absolutely. And, and kind of talking across that so yes the preservation piece as part of that infrastructure that still needs to be developed we still need to figure out kind of standard practices for how to archive what is archived. In a dissertation I focused on creating web archive files of the final presentation layer. But I think what you were speaking about earlier, that people start when they're thinking about the project they're thinking about that display version. And that's the most ephemeral of the whole thing. That's the thing that's going to die. It's going to disappear it's going to become obsolete the fastest because it's the most dependent on the technologies that are moving, moving so fast the data. The part that's the least rewarded in our current system is the most stable part of it and it's the most useful for other for combining with other things for future projects and the analysis bits it's in the middle right. In both cases, it's that middle middle stage thing so how to shift how to increase support for doing data work, both financially because of the time it takes and it takes a ton of time to make usable data, which connects to education I'll come back to that in a moment. But also how to reward it as as the scholarly endeavor that it is, and that's what I focus a lot in in teaching and getting students to think about data is just keep coming back to how much the schema you set up with the structure that you're going to impose on the things in the world that you're trying to turn into computational data how much interpretation is going into that, because you do not capture everything, you never capture everything. It's all making choices about what you're going to capture and how you're going to capture it and how you structure it determines what you can get out on the analysis side which determines what you can display. So, I would love to shift the priority of the two to make the data that that gold piece, but you know 100% concur. But that's where that's where I am to. I had a graduate student in my office the other day, trying to figure out how to make a story map or a timeline. I have a lot of students who need to do those kinds of things and trying to describe, like, first collecting your assets and sort of documenting them in a spreadsheet. And they just looked at me like I had said something that, like, they were just like, and that happens almost every time it's like, they look at me, and they think that they've saved a couple of bookmarks in their web browser. And then they start writing, and they plug in some URLs, and it's like, no, you have to like document it in this, this way that collects the descriptive data that you need so that, so that when you're building out your digital project, you have something to put into it. Yep, because that URL is not going to be be the thing, right, you need right. You title author or creator publisher date. Right, right. And I know, as someone who does social media research, like, there's tons of metadata that I have to collect that has to do with, like, ethical thresholds that I've set for myself as a researcher. But these are not even things that that somebody may necessarily need to know on an item by item basis because they're not going to see it but if something reaches a threshold I treat it differently in the way that I refer to it in my scholarship then if something doesn't meet that threshold. Right, and right, it's like that data and capturing it and documenting it in a in an organized and clean fashion. It's such heavy ramifications for the execution of any project that happens with that data. Yeah, right. 100% 100% and then the humanities I think in the sciences it's different or social sciences but in the humanities we are not often. This is not our way of approaching the world right, we're very used to finding patterns and texts. We, and we know how to cite things in footnotes, but it's a very, it's a, like, three more steps of organization that you need to do. So that aspect of education kind of building in these, these kind of computational skills but the, I think the biggest challenge for education in DH in the humanities is is the model that we have that it's very single authored individual work. So the sciences have a lab model you, you are always a co author on papers in the humanities this is not yet. This is not a norm. There are places where you can find it. And I look to them as often as I can, but to ask students on undergraduate master's PhD projects to do all of that work of data creation, as well as the analysis as well as the presentation it is too much. As someone who has tried it, it is too much to do in the time frames of those projects. And you end up with not as great of a result because you're trying to do all of the things. Right. Education that shifts to much more of a collaborative research agenda where the data creation isn't a single person's single person's venture. It's a larger question that different aspects are explored so that there's always data creation but perhaps you're augmenting a data set because that's the aspect of the thing that needs to be captured to answer your question that wasn't originally, or you need to expand the data set in order to to reach a little further in time or to a parallel group. But instead of having this focus on DH projects. I used to want to think they could be done individually and at this point. I've had enough experience to realize they cannot, they really cannot the projects to do well need to have a joint effort in some way. Yeah. And for students to enter into DH as a field. They're able to join things that are in progress to learn how to do that data work to build on that data in order to get to an end result and a reasonable amount of time, so that they can set up their own project that does that work. And in this case, there are examples of this viral texts out of northeastern the color conventions project spun out dissertations, but they're very uncommon. I would love to see that become much more figure out how to do that within the humanities how to give that model, a chance to try, because I don't think digital humanities scholarship will be able to attain what it could be, unless we move into that that model if it stays the single author thing they're, they're going to stay small, because that's all you can do. Yeah. I feel like, I feel like we're making small incremental steps towards these ideas with linked open data and the experimentation of crowdsourcing data collection. So, a few things like that, but it's, it's taking that extra step of viewing it as a data set that you're building, and then viewing your own scholarship and presentation as a separate entity, right. The other thing that I keep on coming back to as I'm hearing you talk is that idea of contextualization, and just the, the understanding that often when you're building data sets, it's hard to capture that contextualization within the data set. You're building a data set that lots of people are using. How do you communicate the contextualization of the data set or how the data was gathered, what it represents, what types of biases it may include what is the, I always try to think of my, my, talk about building data sets or starting projects, sort of the scoping, like, like a scoping scope when we're, when we're writing a computer program, we establish a scope, right. When we're developing a data set, we need to establish a scope, and that scope talks about the purpose of what we're trying to do right. How do we create data sets that articulate clear scope in context so that other scholars can come and use it, even if we're not around to, to communicate those things like person to person. Yeah, that's an excellent question. I have two examples that I keep looking to, but I'm kind of continually trying to figure out how best, how best to make it useful. So I'm going to pull something up real quick. That's why I make sure I have the authors correct. There it is. Yes. Okay. So the one example that I would point to and that I've had students try making one of these with mixed results but overall I think it was good. And that's a data sheet so referencing here the paper. Would it help if I throw it into the link into the chat. Yeah, and then if you can capture that. I'll put it yeah. But it's data sheets for data that's published by Oh, it's not letting me copy. Hang on. This is a good brew and a wall like in Kate Crawford and Jamie Morgan's during a whole collection of scholars, but it's focused for those these large data sets that are used to change to train large algorithms. And it provides a structure for documenting, like, what is in this data set. What are some of the potential ethical risks who are the the persons what are the limitations of the data set. So it's a document that's supposed to sit with the data set and gives that that context. Yeah. And I think that's really helpful. It's a kind of way to make that information visual and also kind of forced you to go through it to get to the data because, well, yes, and we need to do this documentation work it's still too easy to just like skip it and get the data and do your training stuff because that's where it's rewarded. You're rewarded if you can optimize the algorithm on whatever data. It's not if you can train. Yeah, you get that and you don't read the read me you just you just take the thing. And I'm as guilty of it as the next person right because we're all moving fast but I love this example I think it's really instructive to make students write it because it helps when they're making their own. And they, I've had them create data set so I make them write something like this to help them like, oh yeah I have to think about that oh yeah I need to think about that. On the humanity side of things example that I've been kind of thinking with is Catherine bode's world of fiction. And I'll throw another link in. I can't type and talk, hang on. So, this is a computational text analysis based book that looks at fiction in that got reprinted in Australian newspapers. And trying to find the digital part of it. Hang on one second. I should be online should be a link. But she proposes this structure of the scholarly edition of a literary system, where she is documenting have the titles that she was able to identify within the Australian newspaper collections. And then, there's a link to the appendix. Kind of instead of just saying I found a ton of literature. And here are the patterns in it that all of those literature pieces are really well documented in terms of their metadata and how they relate. So that you have a really clear sense of what are the contours what is the context of this data set. And as a historian context is really important I don't know how to interpret the patterns that I've found if I don't know the context of the documents that came in, if I don't know the, the extents of things. We deal with with spotty records all the time and this is not to say that I need to like have the whole world because you can't. But some sort of information to have a sense of where things start and stop of the quality of the data which is important for me and text analysis. How good the OCR is because we're working working with historical sources all of that sort of information and figuring out how to create an interface that communicates that. If you can't find the other days. I was like, if you if you can't find it you're welcome to send it to me afterwards so that I can just read into. That's the book part there's a second part that's the actual interface that I will find after and send to you. Yeah, just just to be an email afterwards. It might be linked in there I just can't find it quickly. But I really love this example and this is the one I'm trying to build off of with the materials I have collected from Seventh Day Adventist in the 19th century. The nomination digitized a whole bunch of their periodicals that's what I've used for computational text analysis. But it's important to know the history of the publication how many of them are still like how many issues that we have versus how many issues that were published, who were the editors in different years, and all of that informs how you interpret. There's a spike in this topic in a particular year. What's going on in that year I need all that contextual information to know so how do we build that sort of interface that does makes that apparent, not just the model. So as a historian, I love the model but I need to put that model in context. Yes, so China figure out how to do that. So that's where my, my interest in research is right now. That's amazing. Thank you for sharing that stuff. It's, it's just such a, it, I, I, I find myself just rabbit trailing through these ideas, you know, like, but, but this one and then that brings to another one and another one. So let's pull back to, we have, we have maybe five minutes left or so. Let's pull back to something super pragmatic that you've been dealing with for the last couple of years since you've been here at Alabama and that is training future scholars. So what are the top three things that we need to keep just sort of in, in our vision like what are the three things that that graduate students for example, need. And then if we think about future scholars as already tenured professors who wander into DH. Yeah, what, how does that look different for them, like how do we train them. What are the top three things that people need. So, this is going to be around about way to answer that but I will give some kind of patterns and things I focused on in teaching. I would say the first thing that comes to mind is digital natives are a myth. We all now live in a world that's mediated by computation, but we need to be started at like, I keep starting at the beginning orientation to machines knowing how file structures work knowing how to navigate your computer not just through the GUI, whether that's the mouse or the points and click and actually I think it's getting better because the interfaces are getting better and so we are further and further removed from the computational side of our computer systems, and we're more in this kind of visual consumption space. It sounds so old. But in terms of education, it's true. In terms of breaking it down, people are thinking of that visual display element in many ways because they don't have to interact with the underbelly, the insides of their machines. And so education for doing this sort of work needs to start at the data layer at thinking about data it needs to start at the kind of computational systems layer, how how your files work how computer programs work how to take your data and work with it in multiple files instead of thinking of your data as Excel data. It's not it's just structured data. It can be played with across all these different things. So I would say that's the first thing that I that I've noticed that that there is no assumed knowledge, regardless of age there is no assumed knowledge of how computer systems work. I would say the students who are coming up have a much keener sense of what potentially could be done, because they're they're willing to experiment with things or they've been hearing things. So they're aware of machine learning algorithms they're aware of kind of those big concepts, but the nuts and bolts of how to get there. We're all starting at the same spot. So focusing on those basic computational skills. Which should be part of a general education experience would be would be the first thing. In terms of people entering especially for humanities people. Starting with the computation I have found is not the way to go. The way to go is starting with the question that connects to what they're interested in whether that's disciplinary or just kind of how. Yeah, it's largely disciplinary, but it's the, I want to know something that's related to things I care about. So I usually start my computational courses, people in the computer science will probably have small heart attacks but like we start with a problem and how do we fix that how do we get here, and then we work backwards to the more fundamental concepts. Because I've found both when I was learning and in teaching students. I know computers do math really well. I know you need to know logic in order to get your computer program to work. You start a humanities scholar with math, you've lost them. They're not going to stick with you to figure out how to connect all the pieces to get to the thing that they want to do. So I start with a thing they want to do. And then we work back to try to fill in the gaps. And, but we're not trying to be computer programmers we're not, we're not writing papers for computer science journals and that's okay, it's going to be a gappy thing that it's going to get the job done, and then hopefully we can collaborate to do it better. The second thing is starting, starting with the end. The flip the switch or flip the script on the teaching part. And I think that's also for both students and people entering later. And then just more opportunities for training and it's been unfortunate coven kind of took a big whack out of a lot of the kind of extra curricular opportunities for DH training. There's still DHSI in Victoria and I just saw a golf version in Canada is is relaunching itself so that's fantastic. But there are very few places to go and learn kind of discrete skills, which are important. But then also kind of building into the disciplinary discussion is how do we use these within our methods how do we think critically about the method not just like and write a computer program or I can make a website. But how does this actually integrate into what we do as religious studies scholars as historians, as literary scholars. Right. I think kind of where I tackle that. Yeah. Yeah, everything that you're saying makes so much sense from just even my past five months of experience here, you know, right. Seeing the projects hearing what people are concerned about. Yeah. I love waking up every day and coming and talking to people about their really fun projects. It is like my favorite thing. So good. That is the advantage of that that sort of role. And I think DH gets to do a lot of that hearing what people are doing and start thinking about connections across those larger patterns and that that's the exciting part to me. Well, Jerry, I think I'm going to try to wrap this up unless you have any other questions. No, I think I, I would love to ask some questions but I'll have to go back to teaching so. Yeah, well we'll do it again. Yes, we'll do it again. Because this is an ongoing series so I can, I can invite you back. Thank you so much to talk about. Thank you so much for your time. You are so generous. And I personally want to thank you over the last five months you you've helped me think about things and get settled and like, helped me find context and you've made lots of connections for me and I just really appreciate you so much. And I'm sad to see you leave you a but I'm excited to have you as a library colleague. Yeah, it'll be fun. And I hope that we get to work together. As we move forward in our, in our respective positions so thank you. Absolutely thank you and we're very glad to have you, you here in this position so and thank you for all your work and organizing these this is fantastic. Yeah, absolutely. It's fun with all of it. It just me spend so much time with people listening to their, their work. So. But yeah, so this concludes our presentation today. I hope folks join us again next week for Kerry Hill. She's the digital scholarship librarian down at Auburn and she is going to talk about her own research. It's going to be a scholarship and fan fiction, and it's going to be so much fun. So I hope folks join us then. And once again, thank you so much Jerry. Welcome. Thank you.