 Welcome back to another episode of Domains 21. I have the distinct pleasure in this segment to be talking with both Kathleen Fitzpatrick and Scott Chopri, who are both part of the MESH Research Center. And we'll find out more what that means shortly. But Kathleen is the Director of Digital Humanities and a faculty of English at Michigan State University. And Scott is the Assistant Dean for Academic and Research Technology. So without further ado, I'd like to invite them both to the stream and say hello. Hi, Scott. Hi, Kathleen. Hi. Hi. So I wanted to start this chat talking a bit about MESH. A, what does MESH stand for? What does it do? And tell me a little bit about your groovy mission statement, if you will. We'll do. So MESH is not an acronym. It is meant to be a parallel in the MSU universe to matrix, which has been a digital humanities center here at MSU for a very long time. So MESH is a research and development unit at MSU that's focused on the future of digital scholarly communication. And our value statement that we have on our website at MESHresearch.net really focuses on the principles of openness, of interoperability, of transparency, in governance, and of really thinking about the ways that the platforms that we use for scholarly communication today might best be academy-owned and governed, might best operate as part of the scholarly networks of communication that we are relying on today. So MESH has been around for about a year and a half now. I'm the director of the unit. Scott is one of our associate directors. And we're working on a whole series of projects at MESH, including supporting the future of Humanities Commons, including thinking about new models and technologies for open peer review, and a whole range of other things besides. So there's a lot coming out of MESH that I think is potentially exciting. I was particularly taken by the discussion of MESH privileges, open source, academy-owned, collaboratively developed, transparently governed, I wrote it all down, and highly interoperable systems. And it's as if the mission statement is as much about equity as it is about thinking about the technologies in the infrastructure that maybe can afford or provide some of that equity at the level of the campus and the academic institution. Does that make sense? And can you guys talk a little bit more about what that looks like and how you're doing it? Yeah, I think it does make sense. And I think it's a really core component of not just the projects that we're working on, but also the ways that we're trying to work on them. Part of what we're doing is setting in the infrastructure for our work and the work with the faculty in our college and around MSU and beyond that allows for us to do all of those values that Jim just spit out back at us there that we're talking about, but setting up those infrastructures to allow that to happen and to then work with them to get people into those structures and working in that way. So it's both a technical, but also kind of a really social and contextual piece and community piece that we're working on developing. Absolutely. And I think that bit about transparently governed and academy owned is really key. What we're looking at is a lot of the systems that are being used on college campuses, college and university campuses right now that are basically scraping enormous quantities of university produced data in order to sell that data back to the campus. So what we're trying to do is find ways to create the platforms and the infrastructures that will allow campuses to be more self-determining, to maintain ownership over their own data, to have some rights in the ways that the platforms that they're reliant upon develop into the future, to have a kind of autonomy and self-reliance in that regard I think is really crucial for the future of the academy. And so developing those platforms to support that kind of work is a core part of the mesh mission. You know, in a lot of ways, we focus on open and openly available platforms. So open source projects and that. But we don't necessarily exclusively focus in that area because there are a number of things that we bring on to our campus that are more corporate or for-profit products, but with institutions and companies that are working to make those data more open and make the systems more open and interactive. So some of our faculty reporting systems in order to work with annual review in that are systems that are allowing us to be a little more open with trading data into and out of those systems and really owning the data that faculty are putting into those systems. Yeah, I think it's important to note that there are better and worse players among corporate partners in these kinds of projects. So we do privilege open source platforms and products in the work that we're doing, but where there are good actors who are willing to share data with us, we are more than happy to connect with them. Another thing that I was really compelled about by the MESH statement, and I really, I spent a lot of time just googling at the MESH site is the way in which you talk about providing kind of like cutting edge digital scholarship tools. And you deal with a wide variety of them from no one provider. And you will talk a little bit about some of those projects and how you're doing that, what they provide for your community and how the MESH research center, which is linked with the library, which I find a lot of these things are, kind of lifts up everybody at the campus, so to speak. Yeah, absolutely. I mean, I can start by talking a little bit about Humanities Commons, which is the largest of the projects that MESH is working on right now. Humanities Commons was originally established by the Modern Language Association and it was meant to be a platform on which scholars, regardless of employment status or physical location or credentials or anything else, could start an account and create a public profile and communicate with other folks in their field that they wanted to be in communication with. And it was intended as well, Humanities Commons writ large, to be a platform that would support scholarly, like small scholarly societies that need to foster communication and collaboration amongst their members, but weren't able to support their own platforms in order to do so. So this was gonna be a collaborative space in which those societies and the individual scholars would come together and collectively develop a platform for social and intellectual communication. And it's worked to some extent, right? We've got 26,000 users across the humanities and around the world who've created profiles and are in there doing this work, but we're at a point at which it became clear that to make this platform sustainable, we needed to bring it to campuses. We needed to really think with campuses about how the Commons as a platform could support their needs and could work with them to really get those campuses invested. Again, in that Academy-owned transparently governed system for thinking about scholarly communication. So we've brought Humanities Commons to MSU. It's now resident here and hosted here. And we have started the first institutional node on the network, MSU Commons, which is in beta, but is about to open up to the entirety of the campus and allow any member of the MSU community to do all that same kind of stuff, to create a profile, to join groups, to host WordPress websites and all of that that the Commons affords. And that we are hoping is gonna lead to some significant expansion. First of all, we've got to serve more communities than just the humanities, right? So we're gonna be opening disciplinary hubs that will reach out to the social sciences and to the STEM fields and think about how those disciplinary and interdisciplinary conversations might evolve. And then from that point, we're starting to think about other campuses. Who wants to come in and be part of this larger network where a scholar can show up, have a single profile that they're maintaining and connect it with their institutional life, their scholarly society life and their public scholarship life all from one place. And you've talked about that pretty openly about trying to reclaim some of the work that academia.edu or some of the other sites that actually are kind of doing this. And it goes to that principle at the heart of Mesh of trying to bring some of that back to the academy and own some of that data, but also kind of build it. I mean, I imagine the Humm Commons is built on top of Commons in a Box, which is a City University of New York project with Matt Gold. I'm gorgeous at many other people who'll be talking at this event as well. So it's interesting that there is a pretty solid community that's trying to bring some of this stuff and kind of own the means of production, so to speak, when it comes to digital infrastructure. Exactly, own the means of production and ensure that we're working in an environment in which we've got what a kind of data sovereignty in which the people who are contributing the work to these networks can feel confident in their ownership of the work and in the ways that work is being used. That if it's being used by other people, it's in accordance with the ways that the scholars would want it to be used. So that kind of control also personal data privacy. I mean that we are not mining any of the information that's being shared with us for any kind of nefarious or non-nefarious purpose. So yeah, I think there is a real community out there from WordPress, BuddyPress, C-Box to the Commons that's really interested in thinking about how we can create open source trusted spaces in which the Academy can do this kind of work and maintain ownership and control over it. What are some of the infrastructural kind of options you're providing your community? Because I'm obviously speaking as Jim Groome from Reclaim Hosting who's running Domains 21. One of the things obviously you all have been doing for this is you've been running your own instance of Domain of One's Own, the MSU domains and you've been exploring with what we might call the MSU Cloud or at least Reclaim Cloud's MSU's instance of exploring that. So I was hoping you all could share a little bit about not only how are you using Domain of One's Own but why Domain of One's Own and then why this separate thing for the Cloud and what do each provider maybe not if that makes sense. Well, so Domain, so we've now been, had Domains at MSU for I think over five years and for us it was really a, we were doing similar work to what Domains has provided us on campus. But very similar to some of the stories that, anyone who's talked with Jim for a little while has heard about the origin story of Reclaim Hosting. It was similar in our case of we were running these systems that were taking up a ton of effort and time among our group but providing somewhat similar features to Domains and so when we got connected to Domains we realized that what it did was allowed us to take ourselves out of providing that core piece of infrastructure and running the servers and things and instead to start to focus on ways that we're using it. So the Domains project for us has really been around digital presence in public scholarship. So getting faculty on our campus and graduate students on our campus to think about writing in public. We have regular groups going. We call them fellows programs. They run almost every semester. We have one starting in the next week on public writing and where we bring groups together and we talk about this and talk about using Domains. And so that's sometimes a very simple use of Domains in a WordPress site and a website but our larger experiments have started moving into connecting to places like AWS S3 storage. So we might run projects in a Mecca based project that's then connected out to AWS storage for the artifacts or other spaces where we're connecting into Google Cloud Platform or other spaces. So that kind of led us kind of let us dip our feet into the cloud a bit. And so when opportunities came up to think about and work with Reclaim Cloud, that's something that Kathleen is really more than the one who's been working with it. So I'll hand it off to you. I've been doing a bit of experimentation there and was super excited about the affordances that the cloud would provide. Precisely because of our interest. I mean, as we talked about those core values that mesh research has in the ways that we work, we're building on open source infrastructure. We're trying to develop academy owned platforms that will enable academic organizations to act with greater data sovereignty. And so we sort of, I mean, if you'll forgive the phrase, we kind of want to eat our own dog food. We had been relying on a whole series of tools like Slack, for instance, that are great and super, super functional, but that don't afford us the control over the work that we're doing in terms of the tools that we use as a team for project management or for communication or so forth. So one of the first things that I really wanted to experiment with and I had wanted to experiment with for some time was Mattermost and open source alternative to Slack. And what the reclaimed cloud allowed us to do was the cloud version of a one click installation for Mattermost to create our own instance in which we now have control over the teams that operate within that Mattermost instance we have unlimited data. And so we don't have the problem with the free Slack where after 10,000 messages, everything disappears. We also don't have to worry about like what kinds of uses potentially or leakages of that data could happen in the background with a corporate provider of a software as a service because our Mattermost instance is ours. So getting Mattermost set up and running on Reclaimed Cloud was super easy. Customization took a little bit of effort and as Jim who was endless help to me in the process of getting things migrated over in the ways that I wanted them can say. We were learning together, Kathleen. We were learning together and it was a delightful process but it was a little bit, it was a little bit of a challenge and I'm still facing some of those challenges. We have, for instance, one of our project management platforms that we use, we use Trello. And so there is theoretically a connector between Trello and Mattermost that will let our Mattermost instance know when Trello cards have been updated and all of that sort of thing. I have had some trouble getting that up and working. It's not up and working yet. So it's, the cloud has been a learning experience and one of the things that we most wanted to be able to do was to think about like what can the cloud do that domains can't? One thing is host non-PHP based containerized platforms like Mattermost that are super powerful and super useful to us. We've also had a little bit of an experiment with one of our colleagues who wasn't able to join us today who teaches a course. It's one of our introduction to digital humanities courses and she last semester taught that course using cloud-based installations of Mattermost, JITSY and Etherpad. So introducing our DH students to open source tools before doing the kinds of work that she wanted them to do in class. The thing that's so great about having those hosted through Reclaim Cloud is that the semester ended, she wants to maintain those installations but they don't need to run again until fall. So we just stopped them, right? And they're still there. They can be started up again when we're ready but we're not having to maintain the kinds of levels of charges that we would if they were running non-stop this entire time. Another cool thing I saw, I mean, and this is me peeking into the MSU Mesh Cloud is you got a discourse instance up and running which again is something that we could not run very easily if at all in a shared environment like Domain of One's Own or a shared hosting but discourse is a great form software that only runs in Ruby and bam, like there's a great use case. It was great and it was easier to get up and running than I expected it to be. And we've been able to migrate. So it is a mesh research instance of discourse. We're planning on hosting all of the user support forums for all of our projects in that space. Right now it's mostly just the humanities common stuff. We've migrated over all of our FAQs and a lot of our support documentation and it's really super powerful and flexible and we're just delighted to have been able to get that started up. And one of the fun parts about Reclaim Cloud and I'm not sure how much you've used it at all is the ability to basically take any container, any Docker container and spin up, assuming it works, right? Because there always is not all Docker containers are created equal and spin that up so that you could have a very niche application for a faculty or for a researcher or for a student and just that one person needs it. But arguably with this setup, that's enough to if it's been maintained and if it's a Docker instance, it would work. So have you played it all with the Docker in Reclaim Cloud? I've not played with the Docker in Reclaim Cloud. However, I have some thoughts on how it really ties well to the work that we do in supporting the faculty and our work because similar to how Domain of One's Own has kind of taken away the technical administrative piece of spinning up some of these things, it allows us to more quickly and easily, as you just said, Jim, spin up instances of things, discourse being a good example of that. If our discourse instance goes on for a longer time and starts to gain a broader usage and it's probably something that we would move off the cloud, off Reclaim Cloud into a different space for broad usage. And part of that is because keeping Reclaim Cloud as this space for tinkering, for experimentation, for spinning up projects really quickly, it gives us this kind of core piece of infrastructure that we're paying for in the name of this experimentation and setting things up that we can then use if we move, in the case of this discourse instance, once we get this kind of set up and figured out, moving it to what I might term a more permanent space sitting in AWS or GCP or Google Cloud Platform or something like that might be long-term, that's where our infrastructure, our major infrastructure goes and then frees up space for another project to come in. So it's really important to us and we do this with domains, we do this with Reclaim Cloud now, we do it with some other pieces of cloud infrastructure where we're able to support our projects that are more in development or more emerging projects that don't have funding to attach them or don't have other means of paying for the services and infrastructure that they need. We can provide them with really powerful spaces for this to get them into proof of concept mode, to get them started to support some of these things and then move them and free up that space for the next project to start up. And a real next generation sandbox, right? Cause that's what for us, lamp environments that are fairly common now were back in 2004, 2005, right? Right. Yeah, and this is really connected to kind of what you've probably experienced at UMW when you were initially imagining what domains looks like. Our initial forays back when I was working with this C panel and other things like that back in that era of 2004, 2005 was literally to make it easier for me for when someone wanted to experiment with WordPress for them to just spin it up quickly and easily and play with it. And if they didn't like it after a day or two, then we just took it down, we could just take it down and get rid of it with C panel and with that infrastructure that domains is built on and cloud is really allowing us that similar sort of thing, all two, yes, right? So one of the parts of the some research that I do along with Kristen Mapes who we wish could have been here but wasn't here today, we do some work around distant visualization of graphical media and films. And so using tools like that to bring media down to our computer so we can start to deconstruct them and work with it is important. But what was great is I use command line YouTube DL for my work because I can script it and I can do it a lot faster. But when we do it with a classroom and we're just trying to get some students to understand how this process works, being able to spin that up quickly show it to some students and then take it down the next day because it's not something we needed up and running is hugely powerful for that. Kathleen, you talked about the start and stop and the start and stop web was something for me that was very hard being a lamp environment person for so long. I mean, we built UMW blogs on lamp like that's we'd started on Bluehost and then we moved it to Cast Iron Coding and then we moved it to Reclaim Hosting. But the idea is that those experiments always start small and then they have to be enterpised and figured out. But I couldn't wrap my head around the fact that you could pay cents for an application for an hour a week that you needed for a class or a lab and then turn it off. And in that regard, you wouldn't wanna run WordPress or a lot of the PHP apps in Reclaim Cloud. You'd wanna run them, the applications that you only need temporarily or like the Altube where you need it, you demonstrate or you're doing something and then you turn it off. And it just changes the model of thinking about computing in some ways, right? Or at least web computing. It really does. And that's one of the things that I think Scott was sort of gesturing at about how we might take some of the experiments that we've done like our discourse instance. And once we know we've got it stabilized, move it somewhere because it's likely gonna have to run 24-7 in order to be able to deal with customer needs and user support and all of that kind of thing. And so in fact, the cloud, depending on how the cost model begins to break down, it might not be the most effective place for us to run it over the long haul. But for things like where we're just getting started, we wanna evaluate something, we wanna figure out whether it's gonna work. We wanna experiment with connecting this to that and see what can be done. The cloud is perfect for us to be able to spin something up fast and then really evaluate what it's gonna do for us. And it's interesting too because one of the things, when we started this, so this was a pandemic project. Like in February, we had no idea what the next level of Reclaim Cloud was gonna be. And by April, Tim had found the software that allowed us to basically get this up and running over the summer. We're some of the first experimenters with that. But what it really struck us as is, as we've been trying to provide people support on PHP apps like Omega and WordPress and some of the things that kind of made us what we are, we figured the same thing is gonna be very prevalent for digital humanists and beyond with these next generation scholarship tools where they're gonna not only need the environments, because there is AWS, there is Google Cloud, there is Digital Ocean and we can't pretend ever to compete with any of them nor would we try. But there is also that piece of do you have someone you can talk to or a community you can reach out to to get some of these things running. And I think that for us was where Reclaim Cloud would make sense, not that it's a long-term solution because things will go to AWS and that's the way of the web, but like do we have places for faculty, students, researchers, et cetera to actually even play with the cloud or even imagine what it is and what it could do, right? Yeah, absolutely. I think that's a really good point too, Jim because I see our Reclaim Cloud instance as this kind of middle ground in the space that we work. And so for a big, very intentionally grown mesh project, we might not need Reclaim Cloud for it because we have our Associate Director for Technology in Mesh has the expertise and the knowledge to do some of the higher level things spinning up the spaces in the broader cloud platform providers. But I think it's those spaces around that we've been talking about around the experimentation around the startup where to me it's not worth expending a ton of technical expertise and time to spin up something very large when we're unsure how something's going to work for a certain situation. And that's where Reclaim Cloud is helpful for us in that. And I think could be helpful in spaces too where if Kathleen and Kristen and I could administrate and run Reclaim Cloud without Brian, our technical director, but we could not do that in something like Digital Ocean without Brian's expertise in setting things up. So for smaller teams in that too, it's really a great space to be able to do these things that, as you said, give the opportunity to imagine what the cloud-based web looks like. Well, this has been amazing. I wanna thank both of you for taking your time to join us for Domains21 and talking about your work, not only broadly with Mesh, which I really find inspiring, particularly your kind of mission and what you're doing and aligning the idea of infrastructure with the idea of equity, I think is a super important kind of mission going forward for many of us, but also just your specific work in domains and support of us over the years, we really appreciate all your great stuff. And we're really grateful for it and for the opportunity to do all of this kind of experimentation. So thanks a lot. Well, thank you both. Thanks for joining us and we'll see you on the flip side. Mm-hmm. Om nom nom.