 So good morning all. Thanks for popping in. I'm Wayne Morse. I'm the co-director of the Emory Center for Digital Scholarship. And my talk today is going to be really a comparison in contrast of three different projects. One starting one kind of our first iteration as a center and then one we just launched or relaunched three weeks ago. So there are some similarities and there are some differences and I think we can learn from both. At least I know that we have as a center about how to do future work and how to handle the problems that I've heard at almost every session of sustainability and scalability to make sure that these things live on past just when we're working with faculty. Website I'm gonna go through these slides very quickly just to give you all a quick basis because I really like to spend some time talking about the projects. That's a little shot of our site. We are purposely have a giant glass wall to make Digital Scholarship transparent to the outside. We are we actually are part of the library and IT services division but we sit between the library and IT and we were put there purposefully when a group of librarians and faculty and IT folks brought us all together from some centers that were originated back in the 90s. A text digital text center, digital pedagogy center, data center were all blended about six years ago now June because this committee saw the writing on the wall where faculty were trying to put their research as part of their teaching and vice versa and the digital worlds were combining. So we've come together in this unique kind of organizational structure which I would argue and I will talk a little bit more as we continue allows us to really position to draw on resources from both sides of the organization and still report up to the same the same person so we can have alignment of mission with GIS digital pedagogy, statistical analysis, mapping. So that should be something that is familiar to folks. We've added virtual and augmented reality in the past three years. And we have about 13 FTEs comprised of librarians, IT professionals and what I want to call newly minted scholars, recent PhDs. And it's been unique to blend that group together. We also have about 30 graduate students on the payroll and they fit in roles from helping patrons when they come in the door to learn that new technologies all the way up to project coordinators because we can't call them project managers because you need some special letters behind your name when you're project manager. But they have they they run our projects for us. They are active team members as part of our teams. And quite honestly, we couldn't do what what we do without them. The graduate students really allow us to scale as projects kind of ebb and flow without having to hire people or set up new types of relationships and let them all go when those projects go away. So it's really been a wonderful relationship. We have a very wonderful program to train our students to when they come on board that you can see online. So here's our principles. Focusing on Emory strengths, the collection, the faculty, the students, the region really make trying to maintain sustainable practices. That includes open source and empowerment. I can't stress the open the empowerment term enough. We just talked about it in the session I was just in earlier today. We're not the kinkos of the world, right? If you know faculty want to come in and do a project, they are active project partners. And they have to attend project meetings. And they have to understand because they're really the subject matter that makes this thing work. We're focusing and including accessibility in our initial mo use when we start our projects. So the faculty and or library patrons, or we have some graduate students with projects as well, all know that that's key. Everything that comes in our door has to be a form of public scholarship. So that's one of the things that helps us so we don't have to drink from the fire hose so much. But if you not if you just doing your own research piece, and it's going to be part of something that you just do and you don't want to share with the world or it's not a final piece of whatever you're going to do as part of your, your tenure and promotion that you will share, then we don't do the special one off. And that for better or worse helps us maintain and kind of manage the the need for our services. And we have to collaborate and coordinate, right? When I said earlier that we sit between the library and it, if we don't have agreements and know that we're all coming in the same direction, we're never going to get their help and we need their help. So we range just like anybody else from consultations to workshops to publish a digital scholarly peer peer review journal to all types of scholarly projects and we call everything a scholarly project that comes in the door. That's not a consultation because it helps us manage resources. So the faculty member wants a once a map that they want to have created in our GIS library and helps them. That's a project because at our team meetings, we have to know that she's out of pocket for that project. So today, thanks for bearing with me for the, for the, for the overview. I'd like to focus now on three projects, as I said, their aspects, their similarities or differences, some of the observations that we've had after we had a little time to reflect on how they went and how those, as importantly, how those things impact how we do business and how we're going to start doing business in the future. First one I was, I'm going to highlight is the Belfast Poetry Group. This is from our wonderful collection of Irish poets that lives in our Rose Library, our manuscript and rare, rare books library. The, the information of, of these poetry groups had several types of analysis that was, that was, that was brought to bear. As you can see the different types of visualizations, the different types of analysis. This Belfast Group met in the late 60s and early 70s. So along with the site, you get input about, or you get bios of all of the different authors and poets. You can do a network graph to see how, through correspondence and how they wrote, wrote and called each other out in some of the, some of their poems. You can analyze that. You can also do a chord diagram to see the strength of these relationships. And then you can see how the two different groups, because they met at two different times in history, how those two groups interact. Those were all excellent. We also did the, the map of where we think and where the correspondences came from. The key point that I would like to highlight today is this piece, the essay. This was a scholarly essay, is a scholarly essay, published by two of our team members, Dr. Brian Croxall and Dr. Rebecca Sutton-Coser. Brian is now at BYU and Rebecca is now at Princeton. But they both had English PhDs and the purpose of this site was to argue their essay using that data. And when we did this, this was one of the first projects we did as a formal center. When we did this, we had lots to learn, right? We learned about the visualizations, we learned about copyright, of course, and how things we could show and what we couldn't. And we learned how to support scholars in their work. And we thought this was going to be one of the traditional modes or one of the most popular modes we would operate under going, going forward. Unique to this project was that idea of it was presenting a specific hypothesis. There was a scholarly argument to be made from, from their work. It test, it tested a common scholarly belief by Brian and Rebecca. The scholars were part of the center. So it was they already had an understanding of what all went into the work, how our center worked. They were part and parcel with those to those discussions. They were the two primary contributors to the site. They were the ones who are we call them our faculty champions. They were they were the ones who were driving the project. And although the original site was earlier, the 2008 version was a new analysis was their unique analysis. So as I said, Brian worked in ECDS, the center, and Rebecca worked in the library. So they were our champions. And last year, about 3,500 unique visitors to the site. There are a couple of places right now speaking of care and feeding that need to be looked at again. But overall, a very successful endeavor, the things that we gleaned from this to apply to other ones or hopefully knock on wood was this idea of a single scholarly article, the net the network analysis I thought was fascinating, and we can probably reuse some of those things. Copyright considerations are everywhere all the time. So we learned a lot to that process. We did provide for the download of the data via GitHub. So the idea was that folks could download could test these these hypotheses and run data on them against run tests against that data and see what they got. Python Django is our primary development stack, which seemed to be pretty stable. And right now we're talking about moving this from an on premises installation to cloud hosting for many, many reasons. And I'll talk about that in just a second. Second project. Variations on the same corpus, different pieces of the corpus, the new American poetry project. This is part of the Danowski collection, again in our Rose Library. The initial argument for this was to talk about a scholar named Donald Allen's piece about how this how many of these authors who published and and the publications connected. So again, we're going to use some of our analysis techniques. We published the covers of the journals because we didn't have a copyright to publish every article. You could go and search some of the contributors and see how they interconnected just by their names. Again, we showed the argument of Donald Allen and what he thought were the connections between those authors. And then we gave users the possibility to go and dive deeper and see what the full collection of data showed, whether or not that proved or disproved or brought brought new things to light. One main change that I recognize was the change and how the article or the scholarly argument was brought forward. Kevin Young wrote this and he wrote this as an invitation to other scholars to come in and look at the data. It wasn't a definitive argument against Donald Allen. In fact, it kind of pivoted on that initial plan. And I'll talk about that here in a second as well. But I thought it was very interesting. It's just two or three short years to recognize the change in one scholarly argument and the data that went against that to an invitation for scholars to come up to the Rose Library and see the collection and also analyze the data that was brought forth on the site. So that was one of the unique pieces of this. There was another unique is that this had to be strongly connected to the Rose Library and to the library in general. So it's less of an independent scholar or a non-library scholar's work. It had to be very integrated in that. The scholar framed the question. He was, again, English faculty and some of our litz developers were also partnering with them in fact several because it changed, like I said, originally from the Donald Allen argument to a broader argument that Kevin Young brought forward. So the five, again, primary contributors. So we're up in that number and a little more than 3,000 unique visitors in its two or three years since 2018 or in 2018. So pretty good visibility. Again, our shared items, which we were happy to reuse. The first three are ones I spoke of already. Download for the data we thought was very successful with our initial one. So we put that up for people to analyze the data and we saw people using their own tools, software tools as one of the primary ways for them to get on to this analysis. Again, moving it from on-premise to cloud hosting, Python Django continued to be kind of our base application stack. And we really thought that we could apply some of those lessons from our initial Belfast onto this, which we did. So project number three, the slave voyages trade. Slave voyages.org, which just two weeks ago, two and a half weeks ago, was relaunched with this brand new, can't really see it, but it's an awesome graphic, brand new look and feel integration of new data from not only it started out with the transatlantic data, 36,000 voyages of enslaved people to now the intra-American database. Much more looking for, hadn't been touched since 2008. So this is a newer look, much cleaner, since it's been around for quite a while. Several different resources for the methodology involved in the site. It has a little more longevity than my first two that I showed, so it has quite a history. There's a peek at the database. One of the primary changes for this is that you can manipulate the data at a very high level on the site itself. Now this went counter to what we had had historically done with allowing for the data to be downloaded, which we do for this as well. But the history of this site had shown that one of the primary user groups are K-12 organizations and K-12 teachers. They don't have a lab where they can download SPSS and SAS and R and run against the dataset. They need this capability within the site to generate and have their students go on the site and generate research. As also did traditional scholars. They used it quite, quite heavily. They were almost 75% of the user base. So this needed to be incorporated. Did it increase the complexity of the site? Yes. By a severe magnitude? Yes. But was it necessary for us to retain and support public scholarship with this data? Yeah, it was. It is. So the site also contains some estimates that scholars put together about how many enslaved people started on the voyage and how many actually got to their destination. As I said, you can go into all of the different voyages. All 36,000. And now we've just added another 11,000 or so from the Intra American. Go in there and get some highly detailed information. Export that as CSV. It's very, very powerful site. There are some visualizations about the pathways themselves that the ships took. There are several essays from scholars all over the world. Again, because this is the, I would argue the longevity of this. And also Dr. David Altus, who is now emeritus at Emory, was the one who started this. Many, many of his graduate students are now in positions all over the globe and using it for additional research and for their research and adding to the research. So you can see how many different essays are involved. So a little different from our first one where it was the essay and the argument and the data supporting it. And also different from the second one, which was an invitation for people to use the data. This is kind of a hybrid. It uses both. It uses both. It has definitive essays on board, but it also invites folks to get on to the site and do their own analysis and come up with their own arguments. It also includes an image gallery. And as of three weeks ago, the first and only one that we know of, 3D representation of a slave ship taken from 18th century plants that you can go down and see. So there's a video on the website if you all would like to take a look at that. Fascinating, powerful, and upsetting all at the same time. And you can imagine the project team. This is just from 2019 to the present and it scrolls way down beneath what my screenshot was. We both have teams back and there's two other iterations. 2008 actually was when it made the migration from CD to online. But we felt as important as I've heard in several presentations and have had heard since the last time I attended CNI three or four years ago. The importance of capturing the people who were on the project team and did work as part of that team. Be it graduate students or be it IT professionals or be it librarians or be it whomever. To have that as a record of the people who worked on this scholarly project is one of the keys. So a quick review about unique items here, the analysis of the database in multiple ways via the site, provided online, really to support that accessibility for what we're calling a non-traditional scholar. To have people actually interact with the material and the database without having to download their own their own systems. It's translated in three languages because of the importance of the data both in English, Spanish and now Portuguese especially with the addition of the intra database. The huge amount of enslaved people from Brazil up. The Portuguese language was needed. We have open educational resources, OERs as part of the site now. And as I said captured the multiple sites or the multiple sets of team members and last year was over 300,000 unique visitors. We do have additional data again from the users of the site how valuable it is and they are very vocal when something doesn't work when they'd like to add something new. So this is just one piece of the analysis and the evaluation for this site. The shared items again providing the data using the same digital ways or having digital data visualizations, scholarly articles, image collections, Python Django still is the primary stack. There are several applications that live on top of that. And this one is very interesting in the way that it's going to move from on-prem to cloud hosting. It's been lucky enough to receive several grants from the Andrew Mellon Foundation and from the NEH over its lifetime. The most recent one to do this latest update that you saw with the UIUX and the addition of the other data sets, all toll was close to half a million dollars between Mellon and NEH and Harvard and California and Emory. So we envision a time where that's not going to be as easy in the future. It's going to be more difficult. So the plan is right now is to create a consortium of multiple schools, put it into AWS, let it sit up there so people don't have to worry about, oh at our university we have Ruby and Rails not Python Django so we have to flip things over here if we're going to bring it on to ours now if you want to hand it off. So it becomes a collective resource. Everyone pays a little bit to be part of the consortium. We've spoken to at least a dozen faculty from different organizations who are interested in joining, saying is it cool for us to have it up there? Can you still write grants against it if you don't own it? It's not on-premise. And they said of course. In fact they actually prefer it that way because they don't have to fight their IT departments and worry about those things. If they can write a grant for a piece of it to add to add new data or to do something like that, that's much easier for them. They still get credit in their TNP. Their institution gets credit. They get to do their research but they don't have to worry about kind of all the other stuff that comes with moving a huge system to their local premise. So the idea right now and still it's informed the stage that each one will pay a little bit. It'll go to a kind of a collection of money to help support it and they can write individual things and they'll go in front of a group, a committee to say yeah this is good let's bring it in now. It's not. So we'll see how that new model scholarship works. I think the sciences have been doing it for quite a while. So here's all three. There were some things that we could transfer between the between the three and there were some things that probably didn't transfer so well that we learned about. So in the unique considerations role, and I kind of see it as a as a wave of complexity. Once you want to make one change how many ripples come from that and how do you handle those ripples? The bigger change you make the more ripples you have to deal with right? It's like a pebble in the in the water. More complex design when we did it for slave voyages, much more work from our developers. Actually we had to go out and partner with some external developers to get that done in the timeframe for the grant and that introduces its own rate of complexity as you can imagine. The coordination of that is something that we've learned a lot about specifically for the voyages project. Like I said the funding for that has come from multiple sources. So our project coordinator, she can't just focus on the project. She's got to focus on where all these grant funds coming from. Does every see them? Where do they all get to you know sent to? How do you how do you know disperse them in a way that makes sense? How do I make sure that the researchers over in England are getting paid for what they're doing before they send their work? So there's a whole different level of complexity and we are still we even partner with our finance folks and with our RAS unit to make sure things go through with the grant. But there's a whole another level of complexity that she has to deal with. Scholarly outcomes are unique. Some people want to make their argument. Some people just want to introduce and welcome other scholars to make their arguments. Accessibility is key, making sure that people can get access to these things. Band width of course considerations. Making sure we don't have to have any external plugins anymore. Sustainability like I said with the one model we have for voyages. I think it'll be an interesting experiment to see how we can sustain that in this in this current current times. And then the idea of bringing all these elements into the university's repository. How do we make that happen? How do we make these kind of non-traditional elements in there which is something that everybody is talking about the conference as well and we are wrestling with. So just to know that when you increase the wage or increase people in funds and time we can share our methodologies or we have shared methodologies on network analysis which is proven to be handy on not only these projects but multiple ones. Copyright guidelines have proven to be very very beneficial. And the use of having an agile team. Not only us as a group within our division being agile and being able to bring projects in and then empowering faculty to take them on out the door through a dream host account or whatever we need to do to get them to live on so we don't have to maintain all these. But really being able to go down one direction with a faculty member and say wait a minute I want to add this or I want to change and the whole team back up and go to a little different direction. I would argue that you know the IT when they're keeping people soft running or the library when they're keeping the discovery tools and everything else going for them to stop and take a right turn real fast is very disruptive. Our team it's still disruptive but not quite as much for for us. Cloud hosting I think is going to be something that's very beneficial. Like I said we we use dream host for we have currently have 47 ongoing projects in house. And most all of them have a plan to move to dream host be them WordPress or Omega or any other tools that we support. And for a few hundred dollars faculty can pay and keep all the back end going. And then we can be there for the front end to put in new data and really try to do new new new experiences new ideas. Project coordination is key. Not only to make sure that things get paid things get done communication is critical. So it's much more than just someone to convene meetings this is really an integral this is more of an IT project manager ID. That someone who knows the project knows the content knows the connections around campus. And knows how to make this happen. And it's respected in that position as well. So it's more than just to you know someone to make sure that the meetings happen this is really a key team member. And then you add in the grants and the funding and it just to give me gets more and more more critical. Continuity also is another area. And by that I mean continuity within the team. Because if we swap out a developer halfway through because of some unforeseen instance. A new one has to come in and get all up to speed and ramped up and understand what's going on. So there's that part of the continuity equation. There's also part of the continuity equation with the faculty. If they get pulled off an area and we have to show this for a while they have to understand that they'll have to go back in our in our triage line and they'll have to be you know come when it when when we have an opening in the next six or eight months. That's there's a little problematic when you get grant deadlines with that. But I would hope that they wouldn't let the grant slides. We haven't had to knock on wood somewhere to deal with that yet. And then to transfer the care and freeing. So like I said dream host helps a lot in doing this. Qualified external vendors that we have a great relationship help do that. We do have a few that we host initially you know we host internally. Some of our tools that we build open access tools. We signed SLAs with our IT folks and make sure that those work. But I'm trying to keep those at a manageable level. Simply again because we need bandwidth to do new stuff. So we're taking care of all the old stuff. We don't I can't ask people to put in 150 hours a day for a week. So. And how it's changed how we do business. Really to record some of these things and make note of them is key internally. At our at our division meetings in our publications in our blog in our presentations to everyone. Notice how we learn from these things even if it is a mistake where we have to turn. We need to capture those. We also need to recognize the pace of scholarship. So. I've been with them now 22 years. I came from outside of higher ed. And the pace is very different as you all probably know I'm reaching the choir probably on that. But that's something that you just need to deal with. You're not we're not going to you know translate. I'm not going to give a faculty member an IT SLA and have them understand what's going on. So I have to understand how to not only translate that but also work with them to make sure they understand that. Again we have to map our processes from from from our partnerships. We have several partnerships not only with our departments in schools but also with external vendors. And also with external schools and regional organizations. And also the the idea of keeping things up to date but knowing when to when when is the difference between keeping up to date and creating a new project. So it's kind of bug fixes versus new new capabilities and that's a tricky line to walk. Because it's you know you get your customer client patron whatever you want to call them come in and say hey can you can we tweak this a little bit. And that tweak really takes you know 20 hours of programmer time plus some testing and quality control. So there are still some areas that you're saying hey that's great idea but propose a project we're going to make that as as small hurdle to jump as possible and then we'll get you on your way. And the community involvement I mean like I said we couldn't do it without our partners not only internal to our division but also partners in the state and partners in the region and really outside the region like I said voyages partnerships with Harvard and with Cal so there's all different types of partnerships and those aren't going to go away. I think those are going to be increase increasingly important and there's going to be more of it especially when you talk about public scholarship.