 Hello, everyone. Thank you and welcome to another Mara guest lecture. We are very fortunate today to have Stephen Abrams. Stephen has had a digital preservation for the Harvard Library with responsibility for policy, strategic planning, innovation, and oversight of the Library's preservation initiative systems and services. He has been active in the digital library community for over 20 years and has served in an advisory or governing capacity for many projects and organizations, including ca.gov.web.archive, collaboration to clarify the cost of curation or C, data in the lab, data one, end of term web archive, facade, federal agencies, digitization guidelines initiative, the International Internet Preservation Consortium, the Jewish Women's Archive, make that account, the National Digital Stewardship Alliance, Planets and Pronome, and program committees for the DLF, IDCC, IPRES, ISNT archiving, JCDL, and Open Repositories Conferences. He was project director and editor for the ISO 19005 PDF-A Standard and the creator of the JHOV and JHOV2 Format Characterization Tools and the Unified Digital Format Registry, Samantha Quickie. His doctoral research with us here at SDSU in the Gateway Program focuses on frameworks for evaluating the efficacy of digital preservation activities when viewed as technically mediated human communications across time, in concubitant technical and cultural distance. We are incredibly lucky to have him here with us today. So welcome and we're thrilled to hear your presentation. Thank you, Carol. Your preservation as well. Well, thank you for the invitation to come and speak to you and your students. If I know you were gonna read out all that mouthful, I would have shortened it up tremendously. It's a little bit embarrassing to hear it out loud. But anyway, good morning or good afternoon, everyone, depending on where you are, afternoon for me here in Massachusetts. And I'm gonna be talking to you about a number of what I consider to be a very important digital preservation issues. And as a structuring idea for this, because of the nature of your educational experience, I wanna talk to this in the context of building a digital first archive. How hard could that be? So to begin with, the University Archives here at Harvard about three or four years ago announced a very ambitious goal to translate transition rather to what I would call a digital first records management program by 2026. Internally and externally, they often talk about this as a born digital program or a digital only program. But I think digital first actually captures what's actually gonna happen because paper or other types of tangible assets are certainly not gonna go away. Although we would expect those to become increasingly more and more rare in the overall set of collections. So obviously when we're dealing with these digital materials, we have to be concerned with the preservation, the long-term preservation of that material, both in terms of our policy obligations, perhaps our statutory or legal obligations, but beyond that as a university, it has to do with our role, essentially as a memory organization and preserving the cultural record. So I'm gonna talk very, very briefly today about ideas I have about what constitutes the appropriate technical and social infrastructure for the effective and sustainable persistence of the university records. Although this is actually applicable to all other forms of digital information. Beyond that, I wanna end up with a little bit of a, give you a little bit of a flavor of my doctoral research, which is looking into the question of how can we know whether or not our preservation goals have actually been achieved. So to begin with, I just wanna talk a little bit about who I am. So you will sort of understand my perspective. And first of all, it's very important to note that I am not an archivist. It's very possible, if not likely, that all of you may know a little bit more about the archival world than I do. I would consider myself to be a librarian, both by experience and training. And while there is, of course, a tremendous amount of overlap in goals and so forth, there are things that we do differently. I suspect there are things that we perceive or conceive of a little bit differently. So again, just to hint at that experience and training, I've been sort of in this business for over 20 years now, started off back at the Harvard Library in 1999, when we were just first building our digital capacity. And this predates GitHub by a fair amount. So if you wanted something, you had to build it. And that's in fact what we did. And part of my tenure there was in building the first generation of our infrastructure, including the infrastructure for preservation purposes. After about 10 years, I went out to sunny California at the University of California's California Digital Library, where I was Associate Director of the UC Curation Center. I'm not gonna try to disentangle curation from preservation. That's a whole nother lecture, but it's safe to say that I was very intimately involved with a lot of these same issues, including preservation activities there. For the last five years, I have been in a doctoral candidate at the San Jose and Queensland PhD program, which is how I became familiar with DARA, who's a wonderful new addition to the San Jose faculty. And my research is, as I alluded to earlier in the area of the evaluation of the digital preservation enterprise, and in particular, how we can quantify success of that enterprise. And then for the past three years, I have come back to Harvard as head of its newly established preservation program, which is not to say that we have not had been doing preservation for these 20 years, 20 odd years, but it was never sort of publicly identified as an administrative unit, which it now is. So that's about me. I would like to know a little bit about you. And in particular, if you could raise your hand either actually, or if you can do it sort of iconically through Zoom, just wanna get a sense whether you've taken either of these two courses, which seem to be the ones in the MARA curriculum that are most germane to the kind of preservation issues I've been talking about. So I don't know if you're allowed, are you allowed to sort of raise your hand if your video is off? You can't quite see. Well- You could be able to press the reactions button, although I know at least a couple of folks here are brand new for semester first year MARA students. And so they would not have taken those courses yet. Okay, well, that's useful to know because I don't wanna bewilder anyone. There's gonna be some acronyms and some concepts that you may have heard, you may not have heard. Please drop something in the chat or I'm trying to leave lots of time at the end for questions in discussion. So what is the problem we are trying to solve? Let me kind of illustrate it this way. Harvard University is a very old institution and this is one of the oldest records in the archival collection. This is a piece of correspondence that was written by the first Harvard president, what is that, about 380 years ago. And we still have it, the piece of paper and it's sitting on a folder in a box in a controlled environment. And as you can see, we actually have a digital facsimile of this thing. Let's fast forward 380 years. Here is a piece of correspondence from the current Harvard president, Larry Backow, that he sent out to us, all of the staff just about two weeks ago. The problem, 380 years later, we still have that original piece of paper. I suspect 380 years in the future, we're still gonna have that piece of paper. Can we say the same thing with any degree of confidence about this piece of email? Now you may think, well, how hard could that be? It's just a bit of text. Well, let's look at it a little bit closer. This looks like an embedded graphic, but in fact, this is a link to an external image sitting out on a server someplace. Is that server gonna be there in the future? Here is a link to a calendar app, again, external to the email itself. Is that gonna be there in 380 years? Similarly, we have a link to a website down there. Same question. Overall, this looks like text, but it is actually encoded in HTML form. Is anyone gonna know what HTML is or how to decode it? Are they gonna know what all those little angle brackets mean and so forth? More significantly, Harvard uses Outlook as its enterprise mail system. So even though this message is represented in HTML, it's actually being stored on a server and on my laptop as a PST file, which is a proprietary database format. It's binary format, it's not actually text. So is anyone gonna know how to read that all those years in the future? So it's actually even this seemingly simple example presents some really significant preservation challenges. So you might ask, well, in the email case, how good are we? Well, what have we been doing with email? It's been around for a while. Well, let's take a look at some things and try to put this into context. 180 years ago, we actually know what the first telegraph message ever sent was. And in fact, we have the message right there. It was the nature of telegraphy back then because it got printed out on a little piece of paper and that's in the Smithsonian. Fast forward to another couple of years, was about 150 years ago, we know what the first telephone message was. We don't have a tangible representation of it because it's ephemeral nature, but in fact, we know what was said by whom and to whom. What's the first email message? That's only 50 years old. Well, we really have no idea what that first email message was. Ray Tomlinson, who was the person who sent it and actually invented our modern notion of email. As far as he can remember, he sort of danced across the keyboard and sent a sort of a nonsense message just to test that it actually arrived at the other end. What we can draw from this is that significance, archival or cultural in the broader sense, often only emerges in hindsight. So the first telegraph, the first telephone message, those were seen at the time as historically and culturally significant events and were very well documented. Email in 1971 was a very, very fringe thing. It was done in the context of the ARPANET, which is a research project. And the idea was, well, maybe at some point, there'll be a couple of dozen obscure computer scientists who are going to send mail back and forth. No one contemplated it was going to become one of the preeminent channels of communication. And therefore no one thought to actually save it so that it can in fact be preserved. So what do we actually mean when we're talking about digital preservation? Well, if you've had any coursework in this or you've had any interest or looked it up, there's a variety of definitions you can find very, very easily in a number of places. They're all gonna probably sound something quite like this. I picked this one in particular because this is about the newest definition you could possibly find. Dan Nunin and his colleagues out at Ohio State released this as a public document just about a month and a half ago. Now, there is nothing particularly wrong with it. Everything that is said here is absolutely correct. It's absolutely necessary. But I would argue that while not wrong, it is also not complete. So let me sort of take this apart a little bit. First of all, there is this reference to managed activities. Essentially what this means is that the Preservation Act enterprise has been conceptualized as a managerial endeavor. It has to do with the mechanics of you got something and you're gonna at some point hand it off and it's what's in the middle there. Again, absolutely necessary, but not really the whole story. Again, if you've done any study in this area, you've probably come across a diagram. It looks like this. This is the open archival information system reference model. It is pretty much the standard instrument that we use to explain what we do to design systems to do it and so forth. And by calling Preservation and Managerial Activity, we're basically saying it's really only concerned with this little middle bit. And very significantly, the concerns, the needs, the experiences of content producers and content consumers is held to be sort of outside the purview of Preservation proper. I think that's a little short-sighted. Come back to this in a moment. Another thing we see here is that the end goal of Preservation is to provide access. Again, that's necessary, but perhaps not quite enough. Access has a number of meanings, both sort of the dictionary meaning. It is actually a fairly codified archival term. You could look like in the SAA dictionary and you'll find that it has to do with being able to be findable and retrievable, both from sort of a technical and administrative and a legal point of view. You can get the thing into your hand, but it doesn't say anything at all about what you can do with that thing that's now in your hand. It does not address usability. And after all, the whole reason we are archiving and preserving this material is so that it will be used by some future user. Here is President Bacow's email message. If you didn't know anything else about it and you retrieve it, you have access to it, this is what you're going to get. And it's very possible that in the absence of other information, that's not going to be particularly useful. Going along with that is the sort of central focus of the Preservation activity is the digital object, the actual encoding of abstract information into digital form. Digital objects are the primary unit of preservation management. But again, the scope of preservation, I would argue, needs to go beyond just dealing with objects. This perspective of looking at things in terms of the preservable objects is what I'm calling an artifactual perspective. We're dealing with that physical, well, we're virtual, that digital artifact. We're not necessarily there in the concerns we should be on the human experience engaging with that artifact. In essence, we want to turn this binary data into something a little bit more familiar. We want to be able to read and manipulate this email message in the context of some sort of an email client. Lastly, the risks to preservation are couched here in terms of technical risks. And there are lots of those, and that certainly occupies a lot of our time in ameliorating those risks. But basically, there's an assumption here that because we're dealing with sort of archival time spans, there's obvious technical distance that arises across those time spans between the point of acquisition or creation acquisition and the point of eventual use. And that's what this definition is trying to get at. But also accompanying that increasing time dimension is cultural distance. The future user is in a very, very different cultural position than he or she was, they were at the point at which the information creator actually put something down. So to illustrate this, you're probably having a, where I didn't give you a chance to read this message, you will see that in it, President Bacow is very generously giving all of us some additional time off during the upcoming winter recess. Someone in the future just reading this in isolation might think, well, that was kind of a nice thing for the president to do. They would be losing the whole point here. COVID is not mentioned anywhere in this message, but it is the imperative context for understanding what was actually being done here. Over the past two years, Harvard, like many institutions, hopefully yours, has been making a variety of accommodations. To its staff, to its faculty, to its students. And in fact, this extra vacation time is nearly the latest instance of that. So although there is no explicit mention of COVID, you really aren't gonna understand correctly the purpose, the intent and the impact of this message without knowing that it was done after two years of pandemic conditions. So what is digital preservation really? If I was of a particularly poetic mood, I would give sort of a metaphorical statement like this, that digital preservation is about the continuity of memory, whether individual, institutional or cultural memory in the digital age. I actually think that's a very nice encapsulation. That's getting to the point, the goal, what it is we're trying to do, why it is in fact so important. If you wanted a little bit more nuts and bolts to this, I could say something like this, that it is about human communication. There is a managerial, a technically managerial aspect in the middle of it, because we are dealing with digitally encoded information. There's always gonna have to be a technological component to it, but we should never lose track of the site that at either end, there's a person and that what we are doing is to enable communication of past informative expression to some future consumer of that expression. And a little bit more specificity about the actual preservation activity, I think it all comes down to two complementary activities which has to do with assuring the persistence of authentic digital information objects, as well as the persistence of opportunities for legitimate digital information experiences. So here I'm very carefully drawing a distinction between objects and experiences. We need both, the one is fact, you can't happen without the other. I'm also drawing a distinction between my adjectives, authenticity versus legitimacy. When we're dealing at the artificial level of the object, it makes sense to talk about authenticity, which is a somewhat objective measure. Essentially, something's gonna be authentic for pretty much universally independent of who is evaluating that. Experience, of course, is particular to time, place, person and purpose. So we can't really talk about authentic, which implies that there's a single canonical form of it, but we can't talk about legitimacy, legitimate for the purpose of the particular individual. So what do we sort of need to put into place in order to make good on those two points? Well, we can think about things in terms of a variety of programmatic imperatives, things that we're doing at the institutional program level. And what, how many are here, do I have? I sort of collected this into eight categories that more or less move from necessary things to hopefully fully sufficient things as we sort of move down the list. They're all in the same category, as we sort of move down the list. They're all important, but if you had to prioritize them, this is sort of the way in which I would go about trying to develop capacity to make good on these things. First of all is proactive analysis and decision-making. Unlike the archiving of tangible materials, which is often, you know, can be referred to as sort of, you know, well, a regime of benign neglect meant in the positive way is not necessarily a bad thing. We take a piece of paper, if we deacidify it, or it's acid-free, we put it in a box, we keep it cold, we keep it dark, we keep it at low humidity. We ensure that there's no fire, there's no flood, there's no bug. Beyond that, it sort of takes care of itself. And if something does go wrong, we have opportunities for reactive amelioration. In general, in the pressure of digital world, that all gets flipped on its head. If something goes significantly wrong, you may not be able to react. So it's all the more important that we have a very proactive set of programs and initiatives in place. Beyond that, the single most important thing that anyone can do in terms of preservation is to acquire it and bring something, piece of material into a, you know, proactively managed and sustainable preservation program. So we want to make sure that we have the widest sort of most opportune acquisition of materials and the least prescriptive bars in terms of eligibility to be able to bring things in. Once we have something, we need to know what it is. We want to be able to provide a variety of description. We want metadata of all different kinds about it. Sometimes the metadata comes along with the thing from someone who presumably knows. Other times we are able to sort of extract descriptive information from the thing itself. There's often a lot of interesting technical, if not descriptive metadata that's baked into our various digital artifacts. Beyond that, there is the opportunity to have enriched description. Say we have a digital text. We can actually, you know, using natural language processing techniques. We can identify named entities, people, places, dates. We can do sort of a punitive subject analysis by looking for concept clusters. We can make a stab at what this thing is actually about. We can do keywords. There's all sorts of things we can do there. There are similar kinds of things we can do with other types such as audio streams and video images and so forth. Now we have the thing. We know what the thing is. We need to protect it. We want to assure its integrity, its authenticity, its usability. We want to make sure that it is under appropriate curatorial control. And ideally we should be giving our curators the ability to exercise that control sort of on their own without having to go through necessarily an intermediary. And finally, because we know that we're doing this for a purpose, there's a whole body of potential stakeholders out in the world. We need to make sure that we have the possible discovery and delivery of the material. And then finally, this is not a one-time thing. We need to continually adjust to changing conditions as well as continually evolving stakeholder needs, aspirations, not to mention the ever-ongoing evolution of technology and the inevitability of technological disruption. So we're going to have to have a variety of infrastructural components to help us deal with this. So what are sort of the operational imperatives, both of actual technical systems, as well as social systems? Preservation makes use of a lot of machinery, but that machinery is always directed by people. It's a human activity as well as a technical one. So I would identify what seven or so imperatives. They're probably more, but these are the ones that can only lead to mind or that I certainly keep in mind in the development of our own program here at Harvard. First of all, we want to ensure that there is maximal stakeholder clarity regarding planning activities and in fact stakeholder participation in those kind of planning activities to ensure that it is meeting their purposes. We want to ensure that there is maximum availability of our systems, of our content, at a time and place and in a manner of user choosing, ideally at user choosing, not what we think they want to do and when, but what went the way they want to do it and when. Those stakeholders should be able to rely on our systems, both human and machine, to behave the way they expect them to, to behave the way they've been documented to behave. All of this should provide maximal productivity, which is the users of these systems should derive maximum benefit for a minimal amount of effort. There should also be, we should be enabling maximum participation and adoption at minimal cost. And we want to make sure that we're minimizing any potential negative impacts on finite institutional resources and in fact finite environmental resources. Because this is all based on technology, everything is plugged in, there's an on switch that's using up a lot of resources. So compute cycles, storage corresponds to carbon. To make that carbon, we're creating heat, to reduce the heat, we're using more energy. More energy is more carbon. So it's certainly something that we're very much engaged with in minimizing our footprint in that regard. And then finally, we want to make sure that all these systems remain responsive to a changing environment. Lastly, I want to talk about what I initially labeled as functional imperatives. But because of the way in which I've written this, I would say these are actually perhaps better couched as functional aspirations. Ideally, we would want to have the capacity to deal with digital content in any genre or any format in any number or any size with any form of description or potentially no description. We should still be able to do something with it, not as much as we would want, but something. In any language or script, in Harvard's library collections, English is the minority. You go back, you know, both bibliographically and in other collections, we have about 60% of all of our materials that are under our stewardship are in non-English. We should be able to deal with any sort of duration. The appropriate lifespan and stewardship for a given piece of material is going to vary widely. And we want to be able to deal with things from the transitory through the persistent to the permanent. We should be able to deal with any eventuality, being proactive whenever possible and reactive whenever necessary. And we should be able to enable any sort of use or reuse, whether that is purposeful or serendipitous and whether it is a mode of use that we expect or that is unanticipated. So those are all kind of heady goals. How are we going to know whether we've made any progress towards that? In order to do that, it is very important that we have mechanisms in place to allow us to evaluate the efficacy of preservation and archival activity. As Nara said, it's part of their records management assessment program that they do internally. Assessments are very useful for purposes of evaluating existing practice as a way to think about new ways to provide better guidance, better policy, better training and better tools. It is also important. Here's a nice quote that I like from a couple of years ago. This is dealing with library assessment as opposed to archival, but I think they're both germane to both areas. That assessment and evaluation, it's very important that it goes beyond sort of the obvious markers of use and satisfaction to examine much larger and global questions of impact and outcomes. Another instance of where sort of the state of the art in preservation is not wrong, but again, I think is not complete. Because of the sort of perspectival bias within the field towards artificial management, this lends itself to an evaluative benchmark of trustworthiness. Has the process, is that worthy of being trusted? Do I have confidence that the preservation process has done what it says it's going to do? That's a very important thing, and we want to make sure that our processes are trustworthy. But when you think about it, trustworthiness is essentially a measure of preservation means that preservation ends. If we can expand our conceptual perspective beyond artificial management to that of communicative experience, what is that person, future person doing with that managed and preserved object? That suggests to me at least a complementary benchmark of preservation success. I would argue and have that success, this is really the critical quality in terms of what we're doing. Because if we did not know whether or not we'd been successful, how in the world can anyone expect us to plan rationally for preservation activity? How could anyone reasonably expect that we're going to get successful outcomes out at the end? How are we ever going to effectively measure those kinds of outcomes? And importantly, certainly at least for me, and it should be important for stakeholders, how can anyone be held meaningfully accountable for the outcome, whether it is successful or less than successful? So what does it mean to be talking about digital preservation success? It's important here I think to recognize that archival preservation in general and digital preservation in particular usually happens as a delegated responsibility and that is played out within an implicit service provider stakeholder relationship. And this can happen within a given organization or across organizations. In the Harvard Library, I had a central service unit that offers digital preservation services out to all 70-odd university archives, libraries and museums. They're not doing it for themselves. They are passing along that delegated obligation to me. There are also perhaps in smaller institutions, this gets delegated outside the boundaries and there are a variety of consortial and commercial opportunities, which again are placing people into this sort of implicit service provider stakeholder relationship. That kind of a relationship is something that is very well studied, at least in the business management world. And there's a huge body of research and scholarship that tells us, what does it mean to have a level of satisfaction or success in such a relationship? And it all comes down to the hopeful alignment of service provider intention, of the realized state of the service provider's intentions and the stakeholders' expectations. So in other words, things are probably going to look something like this, that at some time in the past, I as a service provider, preservation service provider, had an intention to do something with regard to a piece of digital content that leads at some later time period to an actual archive state. And then at some point in the future, a stakeholder is going to want to make use of that material and has a particular expectation about what they want to see from that archive state. Those things could all be equivalent, in which case I think we can declare success, or they could be a little less than equivalent, which means we don't have perhaps full success, although success should never be seen as a binary quality. It is a continuum and hopefully we'll achieve things towards the far-ended scale. So what are the norms for success? Well, I just said that what we're interested in is trying to find an alignment between a provider's intention and the stakeholders' expectation. Unfortunately, those things are not generally made explicit or made visible in the preservation context. However, they are recoverable, sort of from their tacit expression in the form of digital preservation policy statements. A policy statement essentially establishes a social, if not legal contract, that is going to underpin the service provider stakeholder relationship. In fact, what is a policy statement? It is the way of the provider to assert a particular obligation. We are going to do this for you. In the face of that articulation, it is perfectly reasonable for the stakeholder to make an assumption about what they can expect. In fact, if you take a look at preservation policy statements, and there's a lot of information about this in the reference that I had on that last page, and you'll see it again at the end, you'll find that there are four primary norms for success that are implicit to policy statements. There's a lot of other norms, but these are the ones that are essentially our uniform. You find these again and again and again, in almost every policy statement that you would choose to look at. These four are accessibility, integrity, authenticity, and usability of the preserved materials, by which is meant, or at least I interpreted it to mean that the preserved digital artifact is findable and retrievable, technically and legally. The artifact is whole and unaltered from its accepted, if not canonical, state. That artifact is, in fact, what it reports to be, and then finally, that artifact is germane to and exploitable for the stakeholder's purpose. If we take a look at these things, we see that they fall into two categories. Those top three, I would argue, are more or less objective. The determination of whether these things are there or not is going to be pretty much the same for anyone who goes through that kind of an exercise. Usability, on the other hand, I would claim is highly intersubjective. It is, again, particular to time, place, person, context, and purpose. That makes things a lot harder, much easier to come up with objective metrics than subjective ones. So what would it mean for there to be successful usability? We've ensured usability successfully. Well, that means that some sort of contextually situated engagement by a stakeholder with that preserved artifact is meaningful, by which I would say that is something propulsive, is newly known, is newly felt, or is newly acted upon by the stakeholder that would not otherwise have occurred. In other words, we could say the preserved information has, in fact, informed the stakeholder. Now, different stakeholders could look at the same piece of preserved information and want to interact with it in a variety of ways. So there's actually probably going to always need to be a number of metrics or norms that are used to quantify the success of usability. And here are sort of the ones that, again, I like to talk about, these seven. And I know some of the terminology is a little bit obscure, so let me give some examples here. A system administrator may only really care about the physical manifestation or ontic manifestation. Is there a file with a given name in a location where it's supposed to be? In that case, this presentation you're watching right now, it's a number of, it's all of these things, but at the physical manifestation level, there is a file sitting on my lap. Well, I'm a preservationist, there's lots of copies of this in various places, but it's just a set of bits sitting there somewhere. If I was concerned about the way in which, you know, this file was created or say I'm in a digitization laboratory, I might be concerned with the empiric encoding. You know, what is the symbolic representation of this thing? This particular set of slides are encoded in the Microsoft's Office Open XML format, for example. There's also a higher level syntactic expression that I as an author was concerned about. This is a slide deck, it's made up of individual slides. Each of these slides has some graphics, it has some text in a variety of ways. There's also a dimension of performing behavior. So I've been going through these slides one at a time. Some of them have had little internal animations. So that's an aspect of this presentation that has to get preserved along with the actual file itself. In this case, I've been using the PowerPoint application. There are other similar things that could do the same kind of thing. Obviously, of course, there is some underlying semantic meaning that's going to be at the heart of the preservation activity. In this case, the meaning here is what is Harvard doing to build a digital first archive. That meaning is initially expressed from my point of view. This is the meaning as I know it. You watching it are going to interpret what I've just said and seen in terms of a particular context. You are all seeing this in the context of the San Jose Mera program. It's also in the context of your lived educational and professional experiences. You're all going to have perhaps a slightly different take on what this is, which is going to lead to the final sort of pragmatic understanding, which is how is Harvard building a digital first archive as you've come to know it or understand it. With that, I am going to close my formal remarks with just a few high level summaries. As we are building our digital first archive, we're making use of a number of principles that really applies to the preservation of all of Harvard's materials. In doing it, we are attempting to be very complementary to what sort of constitutes the current state of the art, which as I've mentioned a number of times is not wrong, but not perhaps complete. So we feel it is important to approach digital archiving from the perspective of communication as well as management from the perspective of human experience as well as a technical artifact from the evaluating it in terms of overall experiential success as well as artifactual and procedural trustworthiness. And that this all has to be seen from a perspective that is intersubjective as well as objective in terms of the evaluative norms that we come up with. And with that, I will close my remarks and offer some information here. Please feel free to reach out to me later on. Here's some pointers to some recent publications of mine that goes into a lot more detail about many of these issues and questions that I've raised. And at this point, I would be delighted to hear any comments or reactions and answer any questions that you might have. And with that. Thank you so much for that, Stephen. That was fantastic. I know I've taken away a lot and have learned a lot and I'm really excited to turn over and engage again with some of your work students. If you haven't had the chance, please do read Stephen's article in archival science. It is fantastic and it's a really, it's a perspective that we don't have a lot in the field. So I will pause and give the chance for the students to ask some questions. I have many questions myself, but I would like to prioritize their learning and experience. We're all just so blown away that we're, okay, so I will start then. I guess one of the biggest questions I have is how do we deal as you, as you rightfully pointed out, there's a huge distance that grows culturally between where we are now in 380 years from now. How do we balance that distance in trying to address the expectational state of our future stakeholders when our future stakeholders are likely to be so distant from us? Well, that's, it's tricky. And there's sort of two ways you can look at it. There's the somewhat traditional way, I think, which is, well, isn't that what scholars do? This is the nature of the scholarly exercise. You dig up some source material that perhaps no one's looked at or you interpret it in a new way. And that interpretation, of course, has to be based on what you are able to recover or intuit about both the time in which the thing was created, what purpose it might have been intended for. But there's also the purpose that it may be used for in the future, which doesn't necessarily have to be the right thing. This is why I like to talk about legitimacy of use, as opposed to authenticity of the artifact. There's many different types of uses and some that were undoubtedly, well, certainly over-article type spans, completely unknown, unrelated to the original intention. Many maritime and naval archives hold very rich collections going back to the great age of exploration, the 16th to 17th century, including a lot of ship's logs. There was some boasts in sitting there every day, writing down what was going on. That included atmospheric and oceanographic information. What was the air temperature? What was the water temperature? That was for a very practical purpose. And as long as it was on paper, it was kept alive. And what if someone said, well, you know, this really doesn't matter. Let's just snip all that stuff out. It's just taking up a lot of room. Well, nowadays, it turns out, climate scientists are using that very same information, collected for a completely independent purpose to do very, very long-term longitudinal studies and explaining and exploring climate change. No one could have anticipated that. And it could very well have been discarded at some point. It's just, oh, this is, we don't need this. So it's hard. Hopefully we all, as curators, are always thinking about the things we know about and the things we expect our customers are going to want. But there's always the notion of, there's the things that we don't know. We know we don't know that they're going to want. And luckily in the digital world, digital space is relatively inexpensive. At least it has been historically. So the thought is, let's just kind of hoover up as much as we can and do as much as we can practically and try to leave things in as good enough state for that future scour. But again, capturing context, the fullest possible context at the point of creation, at the point of acquisition, at the point of prior use. You may be looking at something that's 100 years old, but there's probably 100 years of prior scholarship. People have written articles, they've cited this thing. We should be finding ways to pull together this vast network of these interconnections so that we're handing someone not just the thing, but the thing as the focus of a whole constellation of related information. So that sounds to me like there's going to be a lot of infrastructural, not just a change, but advancement that's necessary. Because like to my mind, you can't really start that without really building out some real semantic web and really that rich sort of, you know, link down at all of that. Is that an accurate assumption or? Yes. I mean, semantic web is our current, you know, technology du jour for dealing with it. And we always need to think about these things at two levels. You know, there's the information level, and then there's the implementation level. And we want to make sure the information persists indefinitely, knowing that the implementation, you know, if we can get 10 years out of it, well, we're doing pretty well. And then we're going to have to move to the next new, new, new thing. You know, preservation is a relay. You know, it's never over. You approach it, but you never actually get to it. Right. Always have to be handed that baton off. Yeah. So what would you recommend for those students who are here who or who will listen to this later on our webcast, who will be the ones receiving the baton? What do they need to know and to do to be able to take up such a big responsibility? Oh, gosh, well, there's just no ends. There's no end of knowing literally because again, tomorrow you're going to be handed something you've never seen before. And so it's, you know, being, being open, being able, being not too wedded to the past or the present. A lot of these ideas I'm presenting here today. I mean, I will admit, you know, I'm, I'm probably seen as a bit of a contrarian in the field. I don't like to think of myself as a contrarian. I think that's, I like to think of myself as a as a complementarian. Because again, it's not that I'm objecting to anything almost. Well, I do object to something, but I don't really object to most things. It's again, it's, I'm trying to say, well, yes, that's all true, but let's, let's look a little bit. Let's take a little bit of a bigger picture and let's not necessarily be wedded to, you know, receive wisdom. I mean, it's, it's, it is, it is wisdom. And it's useful, but sometimes it gets us a little calcified. And we always need to be opening, opening up ourselves to looking at things from, from a more expansive perspective, which of course makes it a fascinating field. You know, it's intellectually, it's philosophically, I've always found this just to be tremendously interesting because we're just, we're dealing with the raw stuff of thought and memory and scholarship and culture. And we're right at the center of it. Hopefully we're invisible if we do our job right. So, you know, don't go into it if you're looking for a lot of kudos, but if you're looking for a lot of satisfaction in doing a job that needs to be done and doing it well, then this is, this is a tremendous thing to be, to be associated with. Wonderful. So we have about two minutes left. I'm going to give our students one last opportunity to pose any burning questions they might have. Okay, they're just taking it all in, soaking it all in, I'm sure, because it was a lot and it was wonderful. Well, thank you so much for your time and your wisdom received it otherwise that you've shared with us today. Like I said, it's a really important and valuable perspective, you know, that we, we did, we need that both and as opposed to either or kind of expansion. And so we're really grateful to you for sharing your time with us with, and there's about one minute left. If you have any final thoughts you'd like to share with us. Oh gosh. Yeah, I don't know what to say. That's, that's the one thing I didn't prepare for. Okay. Well, I'll just close it up and by saying we are again very grateful for your time and expertise. And this was a really wonderful and thoughtful talk. And I, I know I learned a lot. I'm sure those even stood as well. So thank you so much. I'm very happy to do it. Take care. Bye everyone.