 Okay, welcome back everybody. I make it to be about 2.15, so it's time I think for us to get started on our concluding session for the fall 2022 member meeting. I trust you have all had a wonderful meeting. You've had a good opportunity to catch up with colleagues old and new and compare notes. I hope you've heard at least a few interesting talks, learned some new things, became aware of a few new projects or ideas, and I'm very grateful that you all have been able to join us for this. I just have a few housekeeping announcements before I turn to the main agenda for this closing plenary. First off, I would just like to take a moment to thank all of the presenters. The sessions have just been wonderful. Thank you all. I also want to take a moment as part of that, and I don't know that we need to do a separate round of applause, to also recognize the presenters for the quarterly prerecords, which I view as an essential compliment to the in-person project briefing sessions at our spring and fall meetings. Those of you who have submitted material for those or are thinking about doing that in the future, my thanks on behalf of all of us here to you as well. I'd like to say a thank you to the wonderful CNI staff. They make this look so easy, and they work so hard at it. Thank you for pulling off another really smooth meeting. I hope that I will see many of you in Denver at our spring meeting. We will be making video of all of the sessions available. I expect that that will probably happen by early January. Some will be available before then. We will make announcements through CNI Announce as that material becomes available, but please look for that, share it with your colleagues, use it as an opportunity to check out some of the project briefings that you couldn't attend because you were in a different project briefing. I hope that you'll find all of those recordings helpful for you and your colleagues at your organizations as well. Let me turn to why we're here. Paul Evan Peters was the founding director of the Coalition for Networked Information and a very good friend. When he died suddenly and unexpectedly, CNI and its two sponsor organizations, EDUCAUSE and the Association for Research Libraries, established an award to honor his memory. And really, I guess the best criteria for explaining that award and the people it was aimed to recognize is people who had made a sustained impact to the use and advancement of networked information and research in teaching, in public life more broadly. People who had vision, people who took risks, all of those were things that were very near and dear to Paul's view of what was important in the world. We pulled together a committee every couple of years to select an awardee for the Paul Evan Peters Award. They get a very nice bowl like this and they give a Paul Evan Peters lecture as part of either our fall or our spring meeting. This year, the committee consisted of Lorraine Herakam from the University of Texas at Austin, Kevin Guthrie from Ithaca, Francine Berman from now at the University of Massachusetts, and you will remember her as a previous Paul Evan Peters Award winner. We have been trying to bring at least one previous winner into the selection process. Those three have been joined by Diane Goldenberg Hart, CNI's assistant director. And they very, very rapidly converged on the selection of Paul Courant. There's a very nice little brochure that you might have seen on the registration table and there'll probably be some on the way out if you didn't get one that describes Paul's accomplishments. I'm not going to try and do justice to them here other than to say, I've known Paul for a long time. He is one of these sort of multi-talented people. I mean, he's been a provost. He's been a library leader, a university librarian. He's an economist, but he's an economist who also thinks a lot about how economics and information interact in complicated ways. For those of you who don't remember all the story, because not all of the story was all that public, particularly when it was happening, he was perhaps the pivotal visionary that got the Google book project off the ground and then guided it through to the establishment of Hottie Trust and all the things that have happened there. There is, I think, some truth in the notion that a little bit of hindsight and distance in time really helps you to understand the impact of brave choices that people make. And all I can say is that when we had this pandemic and so many of our institutions shut down hard, there were an awful lot of people who were very thankful for the Hottie Trust emergency access program. That made a huge difference to a huge number of people. And we can trace the fact that that was available and ready to go all the way back to decisions that Paul was in the vanguard of making a number of years ago. So that's all I want to say about Paul other than to welcome him. All of the Paul Evan Peters award talks are sort of sui generis. They don't fit neatly into any of the usual genres of academic talks. I think, as Paul and I were saying earlier, this one is probably sui generis as well. But I am very, very proud and very delighted to welcome Paul Courant, the Paul Evan Peters award winner for 2022 to the stage. Congratulations and thanks for being here. And I know somewhere Paul Evan Peters is smiling. Thank you, sir. Cliff, thank you for those kind words. And thank you for the honor of this award, which really is a fabulous honor. The description of Paul Evan Peters makes me think, boy, getting an award named after this guy, that's a good award to get. I have to give you all credit that in the course of an academic life like mine, one gives dozens, hundreds of talks. So how do you make me feel really honored? Why don't you ask me to give a talk? But this is actually unlike any other talk that I've given. When Cliff called me and read the citation, I found myself floating on air. In much the same way as I did when we won the Hottie Trust case, and the court told the world that mass digitization in the service of preservation, research, and accessibility to library collections is a fair use under copyright law. I commend the sense of floating on air to all of you. I never met Paul Evan Peters, but from what I know of him, as I said, I feel pretty good speaking in his honor about open science, public goods, and the missions of academic libraries, all in the context of network knowledge, its strength, and its power to change the world. We're going to talk about that quite a bit. And it's a thrill to be speaking in a room full of people who have changed much for the better the way in which the academy does its work. You all are achievers of helping make the world be what it can be. And I think that's something to be fabulously proud of. This afternoon, Cliff and all of you tell me, your words, not mine, it's in that little brochure, so I get to brag a little, that I have made notable lasting achievements in the creation and innovative work use of network-based information resources and services that advance scholarship and intellectual productivity. Advancing scholarship and intellectual productivity is what my career turned out to be about rather than my surprise. A series of lucky professional accidents turned me from an economist specializing in public policy generally in tax and expenditure policy, it's really the good stuff, in particular, into an academic administrator applying economics and public policy to the management of universities and their libraries, trying to create productive environments for their faculty and students. I can think of no calling that I value more than the advancement of scholarship and its uses. Both elements matter. One looks toward more and better scholarship. Scholarship itself being one of the principal uses of scholarship. And scholarship in service to expanding our understanding of the world and how it works to advance quality of life and to enhance the joy that we take in learning and teaching, joy for scholars, students, researchers, everyone. And that gets me to the utopia that is in the title of this talk. Actually, we are talking about many utopias. Hannah Gray, former president of the University of Chicago, wrote a book about universities called Searching for Utopia. Her book starts by talking about Clark Kerr's book, The Uses of the University. I claim that the uses of the university are many and varied and involve in no small part the construction and use of utopias. Higher education has remarkable reach. We invent and produce everything. We are humanity at its best, learning and doing, aspiring to advance understanding and the quality of life. We make and use knowledge. The library and the academy are coalescent institutions. They are many libraries and many universities enhancing each other's work and play. Imagine, just take a minute to do so, a world without libraries, bookplaces and schools, learning places. Learning places and bookplaces, information places and teaching places, have to sit next to each other in order to operate effectively. Networking increases our ability to sit next to each other by expanding the practical meanings of sit and next to and places. There is also a utopian aspect to networks. They enable us to do more and better by looking over each other's shoulders. Universities work in part because of complementarities. That is, we add domains of understanding and we use them to describe and calibrate our work in novel areas. There is high payoff to this. Scholars want scholars with related expertise nearby. Economists have a lovely word for this, agglomeration economies. What could be a more lovely word than agglomeration? You know that economists value pretty things. Cities benefit from doing many things in the same way that colleges do. Why do we have great music schools attached to universities? Because you can't get good mathematicians to live in places where high quality music is not easily available. Why do we like having many disciplines? Because they give us the possibility of interdisciplinarity and related innovations. Universities require libraries which provide a continuing powerful instance of how we can find and employ knowledge, which is to say, among other things, to share knowledge relatively easily and find uses for things we don't know. Professions and disciplines both fit here. We use them to inform each other. In the networked world, we can do more by combining many domains of knowledge and understanding. Universities are especially well suited to nurture the scholarly and cultural record. It's an administration's job, the institution's job, to make it easier to find and share knowledge, data, and more. Perforce, these activities require attention to preservation. But sadly, many scholars and academic administrators seem to believe that their work somehow automatically gets preserved in a library or some other safe place. They really do believe that, by the way, it's quite something. There's nothing automatic about it. And the risk of permanent loss is especially great when there are many formats and many producers and where much work that is not current is without market value. If we are unable to preserve our ability to read and keep track of past scholarship and other work of interest, we will lose it. The scholarly and cultural records are always at risk of disappearing. Abby Smith-Rumsey's book is called When We Are No More and it provides several scary stories of the fragility of digitized scholarship. We use the library as an instrument of embodied scholarship, helping us to find things in ways that employ scholarly method, attending to sources and their uses and to preservations. How do I put this? Academic libraries are extraordinarily well-designed to advance scholarship, which is to say, to be academic libraries. The library is very good at finding things and as needed of finding them again. I'll mostly talk today about academic libraries because that's where I've spent most of my time and because CNI and Edgard Carlson ARL, the sources of this award, are all centrally concerned with the academy and its quality, utopias and their uses. But I would be seriously remiss if I didn't note that the career that led me to the Google Book Project in the Hardy Trust actually got its start in a public library, the Bayport Blue Point Public Library, of which my mother was director in the late 1950s. I like to say, completely inaccurately, that I got my start sniffing glue in the basement of the local library. I didn't sniff glue, but I did lick those date-do stickers that we used to put on the back when of a circulating library book was due, back in the era before stickies. This was the first of my lucky professional accidents. My mother went on to bigger and better libraries, including the Patchog Medford Library, which was the largest library in Suffolk County, New York, which she also directed, and then the New York City Library of Performing Arts at Lincoln Center where she ran the information desk for a decade after she retired. When I told her about this talk and its sponsor, she told me that she started the Patchog Library on its road to electronic resources in about 1980. She didn't know what was going on with electronic cataloging, although she preferred index cards, she could see the handwriting on the keyboard, and she knew that if her library were to be effective, it would have to develop this expertise. It was my good fortune to follow in my mother's footsteps. Scholarship and its methods, including teaching, learning, grading, certifying, have this remarkable, powerful, unique arrangement, such that the things we do as academics and scholars, learning and teaching and sharing, are also flexible tools that we can employ to figure out how to configure and operate the institutions of learning. Our work is utopian in its methods and its purposes. We define good work and follow our definition. We review and are reviewed by our peers. Our research, teaching and collections depend on continual detailed criticism and involvement. Here's the utopia that I've been aiming at and continue to aim at. Suppose that pretty much everything that had ever been published were available online pretty much everywhere, and the legal and technical environments were such that the works would continue to be available for the indefinite future pretty much forever on reasonable terms. In such a world, it would of course be easy to create networks of knowledge using source material consisting of almost everything that was available almost everywhere, which leads to the animating question behind this discussion. Why have we been unable to organize libraries, universities, the publishing industry, and related institutions to create such rich networks, and what will it take to get there, or at least to get much closer? Note that organizing the institutions involves law and economics, including suing and getting sued, and attending to costs and benefits in and out of markets. Our purpose, per Clark Kerr, is the realization of what the academy can do. There's no excuse for doing less using the extraordinary resources of the library and the academy. That's sort of my personal mission statement of the problem of the day, produce utopia, get on it. I want to repeat the importance of the special claim that scholars make on our own expertise. Society lets us do research and teaching and gives us the legitimacy to evaluate the quality of research and teaching. This is a terrific deal for us. In return, we give them new knowledge, educated kids, smart graduate students. This is the heart of the bargain between the academy and society. There's a powerful social interest in preserving the arrangement and keeping it going. We trust our judgment to know what is good work. And we use that trust and judgment to advance knowledge, scholarship, the institutions of what we call liberal learning and liberal democracy, and much more. Much as we might prefer otherwise, our claim to be socially valuable depends in part in our ability to describe what we are up to and why. This is not the same thing as explaining how our work is useful. That's a different, complicated discussion that we need to have, but we should be able in any case to describe what we are up to and why. It should be describable. Networks are crucial to this story. Scholarship to be produced effectively and efficiently involves widespread sharing. And I would argue that the most important business model of the library is sharing, which is also the business model of the university, although the latter is more complicated in part because there are products that can be bought and sold and there are economic interests that go with that. Networks make it easy to share information. And it's nice to be living in a world, I keep noticing this, that produces share buttons all over the place. You can't get on the computer for three minutes. Let being given an opportunity to share. That's actually both a fact and a metaphor that we should make use of. We learn how to build these institutions that constitute the academy and we learn how to use them experimentally and operationally. We use the academy and its methods to understand and improve the academy and its methods, which leads me to talking about the Hottie Trust, Google Books, and related entities and institutions and activities. Here's a point that I want to stress. You invited me here today and recognized my work because I was able with the help of many friends and colleagues and Newton's shoulders to build things that worked, build it and see if it's a good idea, see how it works, is a good mantra. So is something, another good mantra is, something is better than nothing. To learn about network knowledge in libraries, one needs to build at considerable scale. So now let's turn to some things, some of what I was involved in building, along with lots of other people, including John Wilkin, Dan Greenstein, Brad Wheeler, Wendy Luget, Mike Keller, Pat Steele, John King, many other friends and relations and fellow travelers, and actually always, in all of these matters, everywhere in the country, Cliff. So the contribution of these hundreds, and then there's the contribution of the hundreds of people who actually know how the technical side works and who can actually build the elements of our utopia. Without that, nothing happens. So let's talk about Google Books, The Hottie Trust, and the Orphan Works Project, The Long Forgotten Orphan. Our study starts with my being excited about the possibilities that come with development of the electronic and the network. Pursuing the affordances of information technology seemed like it should be a good idea for a number of reasons. And as provost at Michigan, I wanted to figure out how to do our business more efficiently. I remember telling someone that I wanted to understand what it was that we got for spending $40 million a year in the library. Well, actually, for one thing, we get access to local collections, which is a big part of the story of why it makes sense for universities to have pricey libraries. The university gets to keep its demanding and pricey faculty and researchers doing what matters to them. Libraries can be used for many things at once. Locally available collections are the lynchpins of scholarly environments that draw scholarship to the university, but then you can do so much more, which we'll talk about in a moment. I was, by the way, as provost, annoyed by how costly it was to rent access to academic journals, and doubly annoyed that we, our university, and many others had to pay for the work, including faculty salaries, and then had to pay to read the work that we had already paid for. That seemed to have, but that seemed to be hard to do anything about. From a faculty perspective, the arrangement was lots of benefit and not much awareness of cost. And from the publishers, and the publishers understood that keeping the faculty happy was the route to market success. Indeed, the journal publishers were quite good at persuading the faculty that their access to electronic library resources were somehow gifts from the publishers, rather than the result of direct difficult negotiations among publishers in the library using resources that could potentially be used, you know, for other things, like faculty salaries. We did lots of cool things with public domain works, spending to get high quality editions and create readable digital collections, but we could barely articulate, much less do anything in the direction of that digital resource utopia that I've been blathering about. Wendy Lujet and John Wilkin and Mark Sandler were doing valuable library work, milking uses out of the collections and creating joint enterprises, such as the Making of America with other libraries. This was excellent and valuable work, but limited in scope and it barely hinted at the utopia that we aspired to. Then one day, Larry Page showed up. Larry was visiting with the engineering college, giving a couple of talks, interacting with students and faculty, imbibing that good academic elixir that places like Michigan and Stanford produce. At some point during the visit, Larry asked the dean of the engineering college, Steve Director, if he could spend some time with Bill Gosling, who was then director of the library. The right answer to this sort of question from a distinguished alumnus who was a potential gift prospect is, yes, you bet. Although the dean would actually prefer to be the main point of contact with the potential donor. Larry asked Bill, if you am, Michigan would be interested in digitizing everything in the library. Ding, ding, ding, utopia calling. Bill called John Wilkin and the two of them came to me. I was provost at the time and we agreed on the shape of an arrangement that would work for us. There were quality issues, technical specs that would have to be worked out. This is another case of build it and see if it'll work. So we, with Google, set up a prototype digitization operation. Fairly early in the collaboration, we staked out an important piece of territory. We would keep our own copy of the scans and the right to use our copy consortially. Note that these were very limited uses. We were limited in our uses of the scans. We could not use copies of in-copyright work for reading or reproduction, except when reproduction was for preservation. Google allowed and would display snippets of their copies. Michigan was more cautious. The limitations and cautions, by the way, continue for the most part to today. Lots of people think that because of the Google project, everything that's ever been published is online. It is to laugh, ha. But we were on our way and the rate of production accelerated from hundreds of volumes per week to several thousand per month. We built out a better facility, got the president and the regents on board, talked with the council, and actually didn't expect any legal problems as long as we didn't make expressive use of materials that were in copyright. Our lawyers were reasonably confident that as long as we didn't take utopian advantage of the scans, we were okay. Google was then doing the copying, not U of M. Both Google and U of M thought that they were okay legally, which in the end we both were, but that was years away. The program was public. We weren't hiding anything. We weren't selling or copying. We could do research, natural language processing, computational linguistics, anything we wanted, and anything we wanted with materials not in copyright. We could make preservation copies and insurance copies. We could search the corpus for lines of text and use the results for research purposes. We had a valuable new set of tools that were especially valuable for searching our collections, but we were still far away from our utopian ambitions to be able to read and share the contents of the library. Even so, the project at this time was extremely ambitious. The plan was to copy pretty much everything in the library. People thought that there was some fancy technology for turning pages. Not really. You could and still can see the occasional copy of a thumb in the copied files. But there was a visible reach of millions of volumes. Across the libraries, people were wheeling custom-made book carts into the library, emptying everything in a section of shelving, about 30,000 books at any moment, bringing them to the scanning station, which was designed to fit a rental truck, scanning, and then repeating it. Think of each book cart as a faculty member at graduation ceremonies. Everyone stands up. Every book in the trunk is put on a cart. The faculty march in order to the assembly hall. The show happens. They turn around and march back in reverse order of how they came, shuttle-like until they have filled the assembly venue just as it was. They sit down and then they do it again tomorrow. The key word and key idea here is mass digitization. No part of the process was novel, although Larry Page was very pleased with his role in troubleshooting scanner design. But the process as a whole was at a scale heard or for or unheard of. Google is and was unafraid of scale. And at the crucial moment, so were the University of Michigan, the California Digital Library, and Stanford. Note that this technology of taking everything off the shelf, scanning and putting it back, is not a respecter of copyright status. Harvard, New York Public, and Oxford all scanned, but they worked from lists of public domain material, not from the whole collection. Mass digitization of bound books was a new thing. Suddenly a piece of the utopian vision looked feasible. We were careful not to make in-copyright or plausibly in-copyright material available for reading or for further copying in digital form. I couldn't do what I wanted, which was to use the scans as the principal use copy of the works that and reconfigure the library's operations accordingly, making the uses of the library both more effective and more efficient. But now all that stood in the way were legal and contractual obstacles, which is to say a lot stood in the way. The technology was in gear, producing about 30,000 books a month, and the quality of the data, the scans, and the metadata, though imperfect, was improving. We were on the edge of having a usable in the technical and physical sense copy of the University of Michigan Libraries with other libraries in a position to add to the collection. A piece of the utopian vision, almost unimaginable, a couple of years before, was coming into view. The scans were in reach. All we needed was the right to use them. There was one crucial negotiation that involved the relationship between U of M and Google. We insisted that we would have our own copy of the scans and that the University could use those copies for a variety of purposes. Our uses violated copyright law, if our uses violated copyright law, oops, which they didn't, that would be U of M's problem, not Google's. If Google did anything infringing with their copy, it would be their problem. But equally important, our agreement with Google gave us permission to use the scans and consortial arrangements with other libraries at academic institutions. During the negotiations prior to running the digitization project at full scale, Google took the position that we, University of Michigan, would be just as well off, maybe better, if we had the right, just merely had the right to use Google scans. After all, they were engineers with expertise in computer science and aware with all to improve the quality of the scans over time. Well, we were mere librarians. And in my case, not even that, really. And we insisted, John Wilkin deserves a special shout out here, that we were academic libraries serving the interest of scholarship. The scans would constitute a copy of the library's collections, which could be deployed in the interest of preservation, which is one of the core functions of the library. Collection development is also at the heart of the library's ability to function, as is the ability to use collections consortially across libraries. We wanted and obtained a backup copy of the library that we could share and use subject to copyright. The author's guild and the Association of American Public, the American Association of Publishers, thought that the project had already gone too far, and that Google was implicitly claiming too much in the way of rights. They sued Google, not us, that came later, for copyright infringement, claiming that the mass digitization project, in which Google was copying and essentially everything, whether in the public domain or not, in the chosen libraries was a violation of copyright. I will spare you the legal history and analysis, but I will spend a little time on the lawsuit, author's guild versus Google, and the proposed settlement agreement, because it gives us a clearer picture of utopia, illuminates recent works on the part of the Internet Archive, recent efforts on the part of the Internet Archive, and others to implement programs of control digital lending that would allow millions of in-copyright works to be made available to the academy and the public. Earlier, I argued that the library should have its own copy of the scans, wherever the collections came from. Having established that claim with Google, we then had the problem, as they implicitly predicted, that it was going to be costly to store all of those files. The provision that allowed consortial uses of the scans provided a way to keep copies for preservation that could be effective and efficient. Rather than have each library, there were now many others, preserve its own copy, we could share with each other when there were multiple copies among some group of libraries. We explored these possibilities, first with Indiana University, what was then the Big 10, and the California Digital Library, and discovered that we could afford to produce and keep higher quality scans by sharing access to those scans via a shared digital library, which came to be known as the Hottie Trust. We could produce a collective library at reasonable cost by using the oldest and most powerful of library technologies, sharing, where it's workable, which is very much the case with digital copies, sharing produces scale economies. Google had a similar set of digitization programs to U of M's with several other universities, and with several Big 10 institutions, and the Big 10 itself. Note, of course, that scale economies are at the heart of how we think about the economics of universities and libraries. A digital file is, in a technical sense, a pure public good. You can make and distribute many additional copies to many users at zero marginal cost. There's essentially no technological or economic limit on the extent to which a digitized work or clump of data can be shared. Of course, there may well be legal or contractual limitations. Note that the ability to share data in this way extends beyond books. Open science is just open access for scientific information. The sharing here can directly improve the quality and quantity of the work, because different entities can simultaneously access it and modify it and make use of it. Scholarship generally is enhanced by having many cooks sharing in the production of the soup. Note that everything we're talking about here derives from data. Find, produce, share, save, preserve. The basic idea of exploiting digital, exploiting scale economies across several libraries, we worked out with Indiana and the University of California. We sketched a deal and we invented Hathi Trust. It exploited scale and sharing and used technology. Within a few years, it had over 100 members. It currently holds over 17 million volumes. It's a tool of scholarship that is widely and well employed. A crucial element of the development of a Hathi Trust was access to the University of Michigan, its wealth, and other resources. As provost, I can make commitments on behalf of the university, including a commitment to fund preservation costs for the University of Michigan holdings in the Hathi Trust. Of course, I had to attend the administration and the board. I didn't have carte blanche. But once there was an understanding that the Google Book Project was a partnership with the university, I had a great deal of flexibility as provost and then as dean of the library. An interesting career path, by the way. To make non-trivial commitments and spend money and the attachment to institutions was critical. This is one of the things that mission-driven entities like libraries and universities can do. It helps to be well supported. The president of the University of Michigan, Mary Sue Coleman, was a steadfast champion for the project for which I was grateful and still am. And she was a steadfast champion of the part of it that also affected other institutions. That was very much her part of what this, her idea of what this ought to be about. Several of the big libraries, Michigan, Stanford and Indiana, were made part of the discussion and played essential roles in the settlement agreement. That would have, and would have been active in its implementation had the agreement been implemented. To my mind, the agreement was one more flavor of utopia with respect to academic libraries and their uses. The utopian features were spectacular, although there were some serious problems, not the least of which was a judicial finding that the settlement wasn't consistent with the relevant law. Oh, snap. The proposed settlement would have created a market and electronically available copies of out-of-print works that were plausibly in copyright. Google would also sell site licenses to colleges and universities, enabling students and employees of those institutions to have access to the collections in much the same way that they now have access to journals and other works purchased by their libraries. The obvious benefit of the settlement is that it would have provided electronic access to many millions of works in one fell swoop, saving the transaction costs that would be involved with Google, libraries and others seeking to provide access to the scanned works, had to negotiate work by work and rights holder by rights holder, assuming that the rights holders could be found. The ability to search simultaneously the collections of the world's great research libraries, to browse those collections and be able to obtain immediate electronic access provides huge benefit to the world of scholarship and beyond. It would save money and space for the institutions in the long run as the necessity of holding extensively duplicated print collections would be eliminated. The settlement also included the orphan works, that is works where rights owners are unknown. This would remove the risks that would otherwise attend to displaying works where rights are unknown and add to the value that would be available to students, professors and other users of Google's new giant electronic bookstore. We could build our access to the scholarly and sculptural record such that the principal use copy would be digital and searchable available to students and other library users with a secure electronic backup. This is an enormously powerful notion. Library patrons would have access to a digital copy that could be stored reliably and used by anyone with requisite authority. It's easy to find and use everything in the collection. The settlement also covered the American Association of Publishers and the Author's Guild. This looks a lot like the definition of the utopia that I proposed earlier. Pretty much everything, pretty much everyone, pretty much any time. I'm now going to skip over one of my favorite anecdotes because of the interest of time, but the basic message of that anecdote was that when we were proposing that we would use the use copy, the digital copy out of the settlement as our use copy for doing the library's business, I asked one of the engineers at Google, so what happens when you go out of business? And he said, we're not going to go out of business, we're Google. We're just getting started, you old fossil you. He didn't say that, but he sort of said that. And I said, oh no, that's not right. You are going to go out of business. You're just a private sector corporate entity. We're a library and a university. We're not going to go out of business. Churches and libraries have specials, churches and universities are the things that don't go out of business. So I've actually reached an agreement on what would happen at the point where Google went out of business and we had to still use the use copy. See the digression version of that took even longer than the written version, so that's useful to know. Back to our stories. Instead of the Universal Library, which was where we used to aim, the settlement agreement would have produced a similar outcome in the form of a universal bookstore, which turned out to be too good to be true. The court didn't approve the settlement and made powerful statements to the effect that the legal basis stretched the boundaries of what class action could be called upon to do, restructuring and creating an enormous market, bringing all the players, including us together with the authors killed on the publishers, turned out to be too good to be true. But we can now see another way to Utopia, a universal bookstore. So the settlement didn't happen. The digitization project did, however, continue and we were moving along digitizing tens of thousands of books a month. In order to comply with copyright, we did not provide display reading access to any copyrighted text. We knew that there were thousands, perhaps millions, of works in the project in Google Books and in the Hathi Trust, works that might or might not be in copyright, but which we did not know to be. Orphan works continued to be orphaned. We spent quite a lot of energy and money identifying orphan works via the copyright review management system under the leadership of Melissa Levine. But the pace was slow. We decided to try something else, a good idea that failed, the Orphan Works Project. The basic idea was smart. There were millions of works out there and we wanted to make a determination of the copyright status of each one. If a work got through a gauntlet of local checks, such that we could not find a copyright holder, the work would be publicly listed as potentially orphaned and we would give the rest of the world time to claim ownership or parentage. Then we would make the work available via the Hathi Trust. Because we were crowdsourcing the parents of these potential orphans, basically we were crowdsourcing the parents of these potential orphans. As it happened, no works were ever released under the Orphan Works Project and hence none could have been misused. We had told the authors Guild about our plan to implement the Orphan Works Project and we had scheduled a meeting with them to see if we could agree on a set of protocols and procedures for the project that would meet their needs and interests. After all, we had, so we thought, a common interest in establishing copyright status and where possible the identity of the rights holder. For the many works that had little market value, we could ask the right holder for permission to make the work available. The Orphan Works Project would have provided a mechanism to add to the set of works that could be treated as if they were in the public domain, adding to the set of works that were available for use. Whereupon the authors Guild sued the Hathi Trust and several of its members including Michigan. They sued us in part for the Orphan Works Project which hadn't even been implemented. They sued for copyright infringement in connection with the mass digitization project which had been going on for years. The suit over the Orphan Works Project was new as was the Orphan Works Project itself. In the end, the court found for Hathi Trust, ruling that copying and sharing and non-consumptive use of its members' collections was a fair use under copyright law. On the Orphan Works Project, the court determined that the copyright issue was not right, that's the kind of thing lawyers say, for adjudication because we hadn't actually opened up any access to any works. One wonders what might have happened if the author's Guild hadn't been so impatient. We had indeed done a poor job of identifying orphans, plainly misclassifying several works that almost any literate person would know were in copyright, but the design system actually worked, at least in a formal sense. People shouted our errors from the rooftops. We immediately took those works off the list. There were then more errors in the second batch and we suspended the project. The design criterion that we would not make consumptive use of copyrighted works was met, although in this context it wasn't so useful since we weren't making any use. I wanted to try again, but faced with defending a big lawsuit, Michigan and its lawyers decided to drop the program which had never really gotten started. The point of the Orphan Works Project was to provide a non-infringing mechanism for getting works into the hands and in front of the eyes of people who can make use of them by identifying works that were presumptively protected by copyright but not currently available in electronic form from their rights holders because no rights holders could be found. The idea was to free the orphans which required finding them. Another mechanism that would yield a similar outcome would require a digital doctrine of first sale. The traditional doctrine of first sale is based on long-standing law and practice to the effect that a legally acquired work is owned by the acquirer. Whoops, where did I go? Yes, owned by the legal acquirer. If I buy a book, I can do all manner of things with the physical book. I can use it as a jackstand. I can give it to Cliff as a gift. I can but won't burn it. I can donate it to my local library which in turn can do all of the things that I could do with it. I can lend it to my friend. If I'm a library I can lend it to a patron. It's not an exaggeration to claim that the everyday transactions between libraries and patrons depend on the doctrine of first sale. The doctrine that the original copyright holder exhausts her right in the use of the book when she sells the book. This is the doctrine of exhaustion more generally. Ideally, we would replicate the effect of first sale for digitized works by treating the electronic form like the physical form in the marketplace. You, the library, have a copy of a work that you bought. You scan it. You allow a borrower to use the scan while keeping your physical copy out of circulation. This is the basic practice underlying controlled digital lending or digital first sale. One can borrow an electronic file in exactly the same way that one can borrow a physical book or an article. The situation is much less salutary if the work comes to us in electronic form. The holder of the library can choose to make a copy available to me via a license. I may have some rights to read the file but not to copy it, print from it, and give it to Cliff. All of those other nice things that would come if it were a book. And it can get worse. The rights holder might decide, for example, that it won't authorize the work to be accessed via a library. This happens frequently, by the way. And there's a recent chatter to the effect that some publishers, notably Pearson, are planning to use blockchain to track uses, borrowings of works for which they hold rights. I don't understand blockchain very well, but it seems well within the technology, as I do understand it, and would allow the publisher to be compensated every time the licensed work is accessed digitally. So much for exhaustion and first sale. And note that under the blockchain scheme, the publisher has monopoly power over all uses of the work. The only way to get a copy of the book is via the publisher's rules, which may have any number of restrictions applying to the uses of the text. Readers are charged for reading, even when there is no cost to publish. Which reminds me, as a card-carrying economist, of an important fact about the world. Capitalists like to talk about how much they like competition and competition has benefits. Capitalists actually don't like competition all that much, but they really like his monopoly. That's where the money is. And this scheme would support monopoly, leaving would-be readers in the cold. The Orphan Works Project was going nowhere, and librarians talked about digital first sale as a dream that they had no way to make come true. And then fate intervened again, this time in the form of the COVID pandemic, and led us to versions of what I, and many librarians, had had in mind as digital first sale in the form of the Emergency Temporary Access Service, or ETOS, a program that implemented a form of control digital lending. The emergency in ETOS was the closing of the buildings and the locking away of the physical collections in libraries all over the world due to the pandemic. This was a public health emergency, but it also produced a test bed for running a library based on control digital lending. It was also a test bed for consumptive reading use of Hathi Trust. The programs passed the tests. Hathi Trust members could use digital elements of the collection. They were owned by their own specific libraries. If a library held two physical copies of a work that it had purchased, it could keep both of them on the non-circulating shelf while permitting two simultaneous electronic uses. Any set of uses that did not exceed purchases would be permitted under control digital lending. Buy one, lend one is the mantra. I note that the Internet Archive, led by Brewster Cale, has also implemented a version of control digital lending. Hashed has sued the Internet Archive over these practices. To my mind, the Internet Archive and its body of work is one of the most successful and progressive nonprofit entities of our time. And Brewster's admonition that the library should buy, note buy, preserve and lend is a powerful one. ETOS was also used for course reserves, which are a natural use of control digital lending. Clearly, libraries had the ability to do more with ETOS than they did, especially if they had time and resources to optimize ETOS. Much material was available that would not have been without the program. From my perspective, the key isn't the emergency or the temporary, but the access and the service. You could do this at greater scale. Indeed, you could do it as part of the canonical utopian picture of this discussion. It's a practical way of doing library business. The clear lesson of ETOS is that it and control digital lending works. Once again, we notably including Jeff Mason at Berkeley and Mike Furlow of Howdy Trust built it in a preliminary way. We saw that it works, and now there is opportunity for refinement and expansion. So I ask you, and anyone else, is there not a pretty clear fair use justification for making a good deal of the collection available under these rules as the default, as the way we do business all the time? Right your congressman. These are an amazing example of a public policy dream that really happens where you can be efficient and effective simultaneously. I'm going to restate that slightly in a moment. Thanks to the pandemic, hardly a utopian idea, we got farther towards using control digital lending, which is definitely a technically feasible version of ETOS. More important, when it turned out that essentially all of the academic libraries in the country were forced to shut down because of a pandemic, the libraries proved to have a pretty good workaround in place. So I can't help but emphasize that if Google had lost Authors Guild versus Google and Howdy Trust had lost Howdy Trust versus Authors Guild versus Howdy Trust, this ability of the system to respond positively and keep libraries open and their students and faculty able to do their work would not have been available. During the pandemic, I got a lot of email thanking me for making it possible to have access to the library. Old news that, but good news. And it's interesting that the court's findings in favor of Howdy Trust and Authors Guild for the Howdy Trust did not contemplate the threat to libraries and the academic and the academy of a pandemic. The Howdy Trust was proving to be useful well beyond what was initially contemplated. Having told my stories, I want to turn to the future using some of what we've learned to look ahead. What's next? What should we be worrying about? What should we be doing to make things better? What are the properties of emergent utopia and counter-utopia? I start by considering I got about out of time or close to out of here. I'm OK. We've got 10 more. Oh, terrific. I think I'll have a glass of water. Yes, there we are. Having told my story, what's next? What should we be worrying about? What should we be doing to make things better? What are the properties of emergent utopia and counter-utopia? Recall that counter-utopia is the set of behaviors that make it more difficult for scholars, scientists, students, and programmers to make valuable use of electronic materials. I start by considering the features of pure public goods with special emphasis on libraries and networked information. These should be kept in mind as we evaluate our utopias and counter utopias. Always remember that we want it to be easy to use the technology in the service of scholarship and related activities. Pure public goods arise naturally from the ways in which markets work when goods and services can be delivered at no marginal cost. And they point to a way of being able to use resources broadly. If we're searching for utopia in libraries, we want to produce science and scholarship and distribute what we learn broadly. The recent Office of Science and Technology Policy Memorandum calls upon federal agencies to provide publicly funded academic work to the public, open access and no charge. This is exactly how the production of pure public goods, according to economic theory of all things, is supposed to work. After all, open science is just another word or a pair of words for science. Do the science, the criticism, the production in whatever form, including publication, and make the public, make the work openly available. The particular budget and fiscal models will be up for discussion and innovation, but the basic logic is clear. Information is a pure public good and it really does want to be free because it wants to be used, generating consumer surplus when it is used, and it wants to be preserved for better uses, for further uses, ones that we don't know but are plausibly at risk. This is why I view the Hachette suit against the Internet Archive as an attack on libraries. And because it's an attack on libraries, it's poor public policy because we have many proofs and arguments demonstrating the social value of libraries and the underlying logic to the effective production and exploitation of pure public goods is economically efficient, where charging for their use is not. Both preservation and use are necessary for our utopia to succeed. Preservation of digital assets does not happen automatically, the way it sort of did with print. Preservation generally requires use as well. This is especially true in the digital world, which is subject to bit rot and link rot among other ailments. Note that the control of the future uses gets in the way, control of the future uses gets in the way of acquisition and preservation, and limiting acquisition and preservation directly compromises the ability of libraries to do their work effectively. That's why the doctrine of first sale is such a good idea. The doctrine of first sale protects future uses. Without it, we run the risk that a publisher could pull books off the shelf, rewrite things to change their meaning and such, leaving no record of the change, or leaving an inaccurate record, just as Amazon did with Orwell's 1984 several years ago. This is both counter-utopian and ironic. First sale allows the marketplace of ideas to function with the marketplace of chattels. The book and its meanings are open for discussion. We face a serious question in the era of big data. It too is a public good, so we don't want a business model that depends on charging for its use in storage. The classic preferred solution to such problems is have the good produced by the government and then distribute it in some rational scheme. The university and the library are the nice, are the natural places to store and manage big data. There's a payoff to coordination among the provost, the vice president for research, the chief information officer, the faculty, the grad students. There's a form of market here in that institutions that are good at managing big data and using it will draw faculty and students and even librarians, much as good libraries, the early form of big data, always have done. Many publishers have become skilled at producing combinations of textbooks, lecture notes, scholarly articles, sometimes even whole courses to be sold or rented or leased in bundles such that faculty and students find it easy to take to use these integrated sources. Publishers of textbooks often employ licensing arrangements that prevent libraries and others from distributing, even when fully paid for, digital versions of the text, shoring up the monopoly power that arises from the convenience of the digital version of the work. Publishers also seek rights to limit the distribution of data and increasingly troubling tactic in the era of big data. To my mind, there is some irony to the fact that the copyright statute explicitly grants monopoly power to holders of copyrighted material. Having learned from the government that monopoly is effective, it's not surprising that publishers seek to apply this insight more broadly. Big data allows a great deal of high-powered coordination and Canon does produce valuable collections of information that can be exploited by scholars and scientists. But it presents us with opportunities to create an exploit monopoly. Technology provides opportunities to let librarians and technologists be better able to work together but the professional cultures remain different. We want scholars in different disciplines and with different technical expertise to work together. And we need the Vice President for Research and Library Director and others who care about data to be able to figure out how to share the data and their uses and how to assure that academic leadership can articulate academic values. The provost should be working with the faculty here and both should attend to preservation. There are many other issues regarding utopian futures for the academy and libraries. They include advancing policies that would promote diversity, equity and inclusion. Employing the work for higher doctrine or other means to enable universities to produce and deliver a rational set of arrangements with regard to academic publishing. Assuring privacy, focusing on using information resources to make progress in solving the problems of the day. Whatever we do, we should work towards being able to use a digital copy of information as the principal use copy. Rare are the policy proposals that accomplish valuable goals better, faster and cheaper. This is one of them and this is where you should write your congressman. We have rich resources, data, text, print and more. Physically and logically organized to use and to share and to make a better world. We have the capacity to get to pretty much everywhere into the indefinite future. Let's take the utopia and run. Thank you. If Cliff tells me I'm on time, we have time. I'm happy to take some questions. A question at least if there is one. I'm glad you have settled all of that. Marna, I did. I named Hottie Trust. A bunch of us were sitting in an office someplace at Indiana University and decided that an elephant was a good symbol for a big effort at a universal library. Elephants are smart and they, don't forget, and they're big. And that's what we were, you know, that's exactly what we did. And I had been reading Kipling recently. So Hottie is the name of the elephant that wanders around in the Jungle Book and does various remarkable deeds. Thank you again. That was quite an amazing tour of the history of those utopias. And I'm particularly grateful for you reminding us of the fundamental importance of first-sale doctrine. You know, I think many of us are coming to believe that the loss of that may be the original sin of the digital age. Sounds about right. Yeah, great. Anyway, congratulations. Thank you for all of your work. And with that, I'm left only to wish you enjoyable holidays. A good New Year, safe travels back to your homes. And I hope to see many of you in Denver in the spring, if not before. Thank you so much.