 This is a question I've actually been asking for quite a long time. It strikes me that library is a very, very carelessly used word, which doesn't make discussion terribly easy. As some of you will know, in the 1990s there was a great move to recreate the great library of Alexandria and they got terribly lots of money and expensive architects and ended up with this structure. It's quite odd if you go inside it. I don't know if any of you've been to the library at Alexandria, but when you go inside it is actually very empty. It is a structure. My life takes me around the Asia Minor looking at ancient sites. That's what I work on. Lots of you, anyone who's been on holiday to Ephesus will have seen this. It's an extraordinary structure, but you wouldn't know it was a library if someone hadn't told you. It's not perhaps the first image that comes into your mind when you think library, and I find it quite challenging to think about. Back in the 80s and 90s when people were constructing the first big digital resources, big collections, nobody could really work out the collective noun and library is quite often used as a collective noun. So the lower classical library, which is important in my field, has all of these, produces lots of books. It now exists in digital form, very much modeled on the structure of the books. In what sense is it a library? When I was publishing some materials we decided to call them a dynamic library because it's a whole series of texts in the same way as the library is a series of texts with various interactions. And of course, many of you will have come across the open library of the humanities and other such concepts that keep on appearing. So it's difficult to know what a library is. I also don't know what a research library is, or at least I don't know what a non-research library is. What is a library in which no research happens? So perhaps we should define a library therefore by what it does, by its purpose. You can't define it by its building or its appearance. What are the functions of libraries? Now pondering this, this is what I came up with, conservation, study, access and interconnection. I work in Asia Minor here. At the site I work at, Africa is there. I showed you the library Ephesus. I'll take you up the valley here. And I work on stones. That is as a way of conserving documents. This is arguably the most enduring and the least convenient. At least you don't have to shelve these. But working on inscribed texts on stone, and you start with something that has to be cleaned and looked after. I came across very early the fact that publishers don't particularly want to publish this sort of thing. And when they do so, they publish stuff like this at a price, which means that libraries don't want to purchase it. And I'm sure you've all had those experiences. And it's for those very practical reasons that I became increasingly involved in online publication as a practical necessity. So stones are a very good means of conservation, but they're not what libraries do. If you look at the Library of Celsus, which I showed you before inside, it's quite pokey. But there were, as far as we, there are these niches in the walls, and you can see it better at a nearby library, built at about the same time. Not very big spaces for storing shelves, which will have had papyrus rolls on them. And there's another one. It's interesting that you get a burst of benefactors building these structures in the early second century. Now, when I started publishing online because of the reasons that I explained, I thought, right at the beginning back in 2000, we were all saying, how are these things going to survive? And in those days, the answer was, once you've published your online publication, libraries will download it. Because it's the only way that people can consult it rapidly. Now, all my stuff has been published online with Creative Commons licenses. You're all absolutely and utterly welcome to download my publications, but it wouldn't occur to you to do so. Because in the meantime, communications become so fast that people simply go straight online. Which has the distribution model would have been fine. Because each library would have cherished the copy that it had downloaded. But now, where does that responsibility for electronic book binding live? Whose job is it? What is research data management? Is that not perhaps what libraries have always done? Just with a posh new name and a bit of technology? But that actually, that's a fundamental library function. And perhaps that's where the question comes. The answer comes to my question. What's the difference between a research library and any other library? A research library conserves research data. Now, in the humanities, I have plenty of colleagues who are happy to say that, oh, we don't publish data. That's a rather narrow concept of what data are, I would say. And I'm at this very moment about to launch into my first depositing through FigShare undertaking. And I haven't started it yet, so I can't tell you how challenging I am finding it, because I don't yet know. So when the library at Alexandria was built, one of the consequences that it had was that a university, a centre for study, which eventually became something pretty recognisable as a university, grew up around it. That's one of the things libraries do, is libraries do universities create libraries or do libraries create universities? Those people I told you about in the early second century, giving libraries to their cities, were copying the Trajan. And as you can see, he decided that libraries were important resources. He provided one in Greek and one in Latin. And there, the understanding is there being these big tables on which, imagine it, your reader comes in and asks for a papyrus scroll, you've got to find a space where they can unroll it in order to read it. And you've also got to make sure that they don't steal it or spoil it or whatever else. So the challenge of a library as a centre for study creates a whole new set of challenges. It also involves a sort of control. And this is another central question about libraries, or has been in the past, who chooses what's in the library, should be withdrawn, like all these discussions of what should be in the school library in an American school, what is appropriate, what is necessary or suitable. This is the place where I work, Aphrodisias, which had late, in late antiquity, had walls built around it, which look like that. And the nice thing about those walls is that they're absolutely full of inscribed stones. And a very, very relevant for one for us is the honours sent for a citizen of Aphrodisias, sent there by an honorific decree from the people of Halicarnassus. Julius Longianus was a writer, and he had produced poems of every kind by which he both delighted the older and improved the younger citizens. And they honoured him in various ways, but above all, they voted that there should be public presentation of his books in the libraries. Notice there's more than one in our city, so that the young men may be educated also in these, as well as reading the writings of the ancients. Library as a medium for thought control. This is very improving stuff, and we want our young people to be reading it. Librarians are constantly confronted with that challenge. Is this a good book? What does that statement mean? And linked in with this question of what should be in the library is the sort of other side of the same coin, which is who should have access to the library. How should people get in? And if one thinks about the great chained libraries, one of your concerns with access is simply to protect the books. We know hideous stories of theft for books, people stealing books, people cutting pages out of books. There is every reason for a library whose concern is to conserve to control access. There's just a slight sort of overlap with the sort of thought police concept where are we controlling on what grounds do we give somebody access? And on what grounds, always the important way to put this, on what grounds do we restrict an access? Who do we not allow in? And why not? Much easier in a way for libraries where people pay. But it's just something to think about. Now the access situation has now changed. It is rapidly changing. Anybody in the world can go and look, this is my first online publication and anyone in the world can go and read it if they want to. It would show good taste on their behalf if they did go and read it. I don't think it was very interesting. When we published this with this and others, we did give it an ISBN. And I remember a colleague saying, it has been much easier for me to persuade my library to catalogue this because it had an ISBN. But library catalogs have been very inconsistent in the way and theoretically this publication could be in every library in the world or that is in every library catalog in the world. And a lot of early cataloging separated something called electronic resources from paper resources. I don't know how many libraries still do that, but it meant that in some catalogs this, which is the digital version, the digital second edition of a paper book, they're not always together on the library page. And that's a reminder that access is useless to you if you can't find what you're looking for. That the catalogue is door into the library. And thinking hard about how to catalogue free open access resources is an interesting question, I think, for research libraries. Again, the question surely has to be on what grounds would you exclude this item? And it becomes more and more of a question because a single day more things are appearing. There's just an endless stream of stuff coming out online. I just took this one randomly. The ancient world online is a very useful mailing list sent out by a librarian. And it has new materials to refer to every single day. How do we deal with that? How do we cope? As I say, how do we decide what to exclude? And that's related to the question of interconnection. Now, the classic interconnection for a library is putting a group of books on the same shelf. And there's no one who hasn't benefited from standing in front of a shelf, which they went to because there was a book they knew about and other books next to it. And being drawn into new ideas and different discussions because of shelving decisions which the librarian had made. And that is a crucial function. And it's one which in the digital world we're still, I think, struggling to replicate because it's difficult to replicate the random and the unexpected as opposed to the organised. Now, the important development at the moment is linking data. In my field, linked ancient data. This plan is already out of date. There's more stuff, but it gives me great satisfaction. I like to reuse it. It makes the point. The ancient world is a very good example of an area of study where people have worked hard to pull together data of very, very different kinds. As it were, poems and coins and all kinds of other things in between to see how those interconnections can be expressed. And that I think is because in the particular case of the ancient world, where our data are in fact limited, we to access the ancient world, we have to use kind of every tool we possibly can. And that I think has been a motor, a driver behind the very energetic uptake of linked open data approaches in ancient studies, but utterly replicable in other areas. And of course, behind it lies the founding concepts of Berners-Lee. I think it's quite interesting. If we're wondering, one of the questions people ask nowadays is they say, why didn't you do X about something that happened only about 10 years ago? And the reason you didn't do X was because 10 years ago it wasn't possible. And I think it's quite interesting that this very famous picture of the linked open data mug that he put this up in 2006, which is actually not a very long time ago. And so it's taking time for these kinds of concepts to roll through. But if you, those still are links, I think that there is a huge amount that's going to be done that is going to lead to different kinds of linking and connecting. I think we're right at the beginning of all this. But the classic way to proceed still is, or used to be, that you go into the library and you say, I'm working on the use of camel transport in the Roman Empire. And you used to go to the librarian and the librarian would say, well, the most recent study is this, that or the other. Or here's a place where you could look, which will tell you where the most recent study is. And the interconnection, there's going to be lots of automation of linking data. And there are going to be key hubs who are specialists. A nice example of a bibliography in a special subject is Xenon, which is again important in my field. Why is it so useful? Because it is a network of libraries. They're all the libraries of, that's the network over on the side there, that's all the libraries. And what they have done is they have entered articles as well as monographs. So it's extremely useful. And it's a job that's only really doable because lots and lots of different people contributed to it. And it's an excellent model of how to go, I think, and the kind of work that specialist librarians can do. I use Vyaf quite a lot. It's wonderful. I was looking at these pages just because it reminds you of how much every single thing in a catalogue can vary, how much all cataloging can vary, however consistent and virtuous everybody tries to be. And I've certainly used it in my own bibliographies to give people something to hang on to for disambiguation. So what is a library? I would argue that the core elements of a library are a collection of securely conserved resources and a librarian or more, more than one, a librarian, a team of librarians, but that actually a library in this sense without the library is not a system that is going to be fully automated in every sense has to be that specialist specialist hubs constantly contributing, constantly picking up connections, constantly saying, well, actually, this relates to that because the linking of open data is not something that just happens. It's seeing connections, observing relationships, seeing, well, actually, there's a book in the Middle Ages, which would be very useful to your study of camels in the Roman world. That kind of looking out of the window and seeing what's there and also librarian as a specialist and a scholar. And that librarian will have two responsibilities, I think. One is for the conservation of resources in their own library, making sure that they are conserved. And if those are digital resources, making sure that they stay available. And at the same time, guiding their readers in the vast ocean of all the other stuff on making it clear what their criteria are, what your, why you're guiding the reader in this particular direction. The library is potentially, at one end, the library is potentially infinite. It's everything that's out there. At the other end, it is the material for which you are conserving and your function as a librarian in guiding the people who turn to your library for guidance. I think that's how I would see it. I think this is the model. The librarians, the guiding Dante through the challenges of a remote and sometimes frightening world. If that's the case, I have a question. It doesn't seem to me that the digital shift in dangerous libraries. It does seem to me that the digital shift has terrifying implications for academic publishers. Now at the moment, we know that lots of academic publishers are making a very great deal of money. But that's out of digital publication. But I suspect that's partly because a lot of the services they provide were really quite difficult to provide as recently as 10 years ago. But all of this online publication and provision of data is becoming easier. For you and your library are responsible for the conservation of the resources published from your university. Where does the publisher fit into that picture? What's the difference between the university library and the university press? I suppose it's a difference of access in a way, isn't it? The press limit, but open access is blowing all of that apart. So one of the questions I'd like to ask you is what is the difference between a library and a publisher? What is a publisher? What does a publisher do or in 10 years time, what will a publisher be doing that is unique to them? I think that's a really interesting question, which it seems to me has been looming for a long time. But we don't know yet how it's going to evolve. But I would be very interested to know at the moment you sometimes feel that what publishers are for is ruining libraries by producing more and more things at greater and greater expense. And it doesn't seem to me that that particular model is sustainable. It became unsustainable in my field, which is an obscure, this obscure field of Greek and Roman inscriptions quite early on. There are other fields where bigger markets, which may keep it all going for longer. But it just seems to me that there's a real core question for people to think about. What is the process, where on the process of creating knowledge and delivering it to people, where in that journey do the publisher and the library sit? So I'd like to leave you to think about that question because I'd really, really like to know what you think. And I did wonder, looking at this picture, we can see the noble librarian crowned in laurel. We can see the worried poet. It may be that the writhing figures are actually publishers, but that might be unfair. So I'll stop now because I'd very much like to know what people think. Thank you so much, Charlotte, for your thought-provoking talk. And Jeff, we'd love to hear what you think and please do pop your questions in the Q&A or raise your hand if you would like to ask your question in person. So I'm going to ask you one now, Charlotte, from the audience. In the digital full text world, is the object not the catalogue? Yes, I think the catalogue. I think catalogs are wonderful, wonderful things. There's nothing more exciting than a catalogue. And I hope I've got rid of my screen, by the way. Yes, yeah. I've stopped sharing. I've stopped sharing. Good, good, good. Yes, cataloging is so important, is such a creative task. And I love, what I love about that Xenon catalogue that I showed you is the idea of the catalog shared project. And the catalogue eventually, do you know one thing I'd love to do? I'd love to stop students being taught to write bibliographies. I hate bibliographies with a deep loathing. Individual ones. Whenever I have to examine the thesis, I look at the bibliography because they're all told to produce one. But there's only one thing any academic ever looks for in a bibliography. What does an academic look for in a bibliography? They look for their own name. That's the, you see yourself doing it and you think, hang on, this isn't good enough. There's a problem here. So I think rather than teaching people to build bibliographies, want you to be teaching people to use online catalogs so that you could, if cataloging hadn't started, if ISBNs hadn't come in so late, we would actually be referring to everything by ISBN, wouldn't we? Do you think there's a role, you know, the interconnectedness, you know, how can libraries and researchers and students together create that interconnectedness because it's different areas of expertise, isn't it? Absolutely, absolutely. So you could take my camels and you could go to a classical library and talk to them about what they have on camels, but then there'd be input clusters of specialists in different places, but all contributing to some central kind of conversation. Perhaps a catalog should be a conversation. And yeah, I've got another question here. Given what has been said about interconnections and the role of librarians, do you think the concept of a collection or collections more widely needs to be redefined? Oh, absolutely. I find it this battle to find a collective known appropriate cluster of knowledge, which I suppose is what we're talking about. Using library to describe, for example, a series of books, it seemed to make great sense at the time. What interests me, I think, is making sure that we retain a sense of responsibility that a particular library is responsible for the care and cherishing of particular resources. And part of that care ought to be making crystal clear to other people, flagging them, making clear that they exist, telling people, we've got, we are responsible for this resource. And this is what it'll do for you. This is how it works. You're welcome to use it, but it's my job as the librarian in charge of that particular cluster to make sure that people know and understand. Yeah, and you don't know what people are going to use. No, you don't. That is what's so exciting. One of the interesting things about publishing online is that you can't, you know, it's like, you can't use acronyms, as it were. I mean, we all protect our professions by having our own set of acronyms, which other people, and you hope that nobody else understands your acronyms, because then that makes you, and you put the stuff out there, and I put online a Byzantine, an 11th century Byzantine text, and it has to be admitted that not the entire world is completely plugged into the life of the 11th century Byzantine. And what was interesting was thinking about what backup material, what links do I need to provide for people so that they can go and follow something up that they don't understand without having it shoved down their throats if they do understand it. That's one of the great attractions of publishing online is that if people want, don't know who the Emperor Augustus is, you can give them a link so they can burrow off and find out all about him. And if they do know, they just carry straight on. And creating that is quite fun. Yeah. And that's one of the things you can do online, isn't it? Absolutely. You can create layers of knowledge. And one of the things I sometimes think is that there's a danger in, well, I'll come back to the research library question, there's a danger in a sort of hierarchy of knowledge being delivered in separate blobs. So you see sites which say this is for junior schools or something like that. This is for school children of a certain level. But actually, what you really like is for there to be an option for a particular child to go further in or not go further in, that layers of engagement, layers of understanding should be accessible from one another. Again, it's not a business model though. And I suppose the other question that I have failed to ask, which you'll have to have in another seminar is, what is the business model for libraries? We know what the business model is for publishers. That's why they survive. But what's the business model for libraries? And we've had a comment, maybe the biggest difference between libraries and publishers is a sense of care and responsibility to knowledge generation and sharing. Yeah. And I think it comes back to that conservation as well. But it doesn't have to be a difference. I do think that, especially I think if you look across at the world, say, of university publishers and particularly some of the American university publishers, some of them have a very conscious, they were rooted, they used to be there in order to give people access to the research done in a particular university. So I think there is a tradition of care and offering access in the publishing tradition. But the market model has rather elbowed it out of the way. Yeah. And I've got another question here. It's a bit long. So there is a vast gap between what the publisher used to do, so produce, edit and distribute, and what the library used to do, keep copies forever. So they don't just get a digital publication like inscriptions of, I'm going to say this incorrectly, probably, I can't hear, falls into that gap. Who distributes in the long term when the original web host can't guarantee to who keeps copies in complex, evolving digital infrastructure? Is that what a digital library should become? And that's always a risk, isn't it? Well, it is. That's why it seems to me that that at the moment, nobody's terribly keen to take responsibility for the curation of the core product, the original product, the core data. And that's beginning to come in with research data management, because in a way, that's what that's about. It's complicated by the fact that we as humanists actually want stuff in an immediately accessible. I want to be able to read it this afternoon. I don't want to have to go and access a lot of data and then not know what to do with it. And because it's come in through the hard sciences, that model, it's been, it's been, humanists are slower to engage with it. But I do think that somebody, that there's a question of responsibility. Well, I used to think the argument, the answer is lots and lots of distributed model, example copies, and which is why the book is such a success, because there are lots of copies. And if there are only six copies, actually, they will all disappear. I do think that because people aren't downloading, there will be downloading of stuff, very often to download it and reuse it and enhance it. And that's great, too. But I think that the responsibility for a published body of data probably needs to sit somewhere with an institution. And I think that that's what research data management is starting to make clear, that someone somewhere has to take responsibility. But again, I have no idea what the business model is. And so another question that follows on from that is, why do you think that the fair principles to which you referred to are normally a most considered and followed within open data, but are basically ignored or seen as irrelevant in the traditional digital publication landscape? Can publishers be fair? Well, where's the cash? I mean, it's quite simple in a way, isn't it? It's, it's, we haven't worked out who pays. And it's a really interesting and difficult question. And things that strike has struck me so much over the last 20 years of doing this is that digital, the online things are much, much cheaper to produce than books, but they're not free. And there is a cost, it's not a huge cost, but it's a cost. And nobody knows where to put it. And I think probably that it falls to libraries, but again, I'm not absolutely sure what the business model for a library is, because after all, the library at Alexandria was the proud product of a king who wanted his thing, you know, wanted to be terribly famous, wanted to draw attention to himself. It was, he wanted to demonstrate the Hellenic heritage of his dynasty. You can't always rely on that. So I don't know if this strikes me as a question to which I do not know the answer. I suppose in the end, we are all actually paid for by the taxpayer, aren't we? In fact, and that's where libraries being more open and giving more access to citizens is a related issue. It's interesting that the online ODNB is available free to almost all public libraries. I think they have to sign up for something, but you can get it through your public library. And that just raises that just an interesting thought. Does a civilized modern society ensure, like those people that I was telling you, but like all those people at the beginning of the second century, do they say libraries are an essential public good? And somebody must pay for them. But that doesn't mean it's less clear that Elsevier is a public good. That, you know, very expensive publishers, that's a different matter. That may be about, they may have a business case, they may say, well, we employ lots of people, but I don't think they're a public, necessarily a public good. And we've had another question in. So responsibility for storing published content, national libraries and legal deposit could be part of the answer. We do download digital copies and conserve it because digital content is also covered by legal deposit. And I think that pulls into this sort of network, what we do above the individual library and as a collective and a consortium. Well, I was quite interested when I published the first one I showed you in 2004, the second electronic second edition of a published book. I got in touch with the British Library and said, what do I do about legal deposit? And at that in those in 2004, everybody was running backwards very fast. Nobody wanted, people weren't really very interested in it. And in the end, we sent them, I think we sent them a CD that tells you, but you know, what the past was like. That was how it seemed. We do the concept of legal deposit ought in a way to be the solution to all the research data management problems. I mean, you just say, well, this stuff has to be deposited. But of course, it creates, it's an awful lot of hard work for the deposit library to receive all these kinds of things. I haven't attempted to deposit my later publications. I've not come across instructions as to how one does it. Actually, does anyone know, are there, is there a request, a requirement from the deposit libraries? Not that I know of. Yeah, I think I can't speak for the British Library, but they do harvest the UK domains. And you can harvest, yes. But they don't say, oh, I don't write to them say, oh, yesterday I published a new publication that there's no process that I know of, though a very good thing if there was, where I would say, look, I've just published this, please harvest it. Yeah, I think there's definitely a way that you can request that they do harvest, and it may well be that they've harvested it already. Yes, yeah. And it was some, yes, Gabriel said, I'm sure they still have that CD. It's never been looked at by a user or copied onto a hard drive. Absolutely. And Chris has put the link to the DL's legal deposit page. Thank you, Chris. Well, that's something I should do then, because I've just gotten, I've got another publication coming out. So I ought to be looking into what my legal requirements are. And yet, and then we've got another question here. So from your point of view as an experienced and successful scholar, would you be happy if the only publishers out there who could make your work accessible, findable, reusable and preserved where libraries? I think put another way, I just think it's quite likely that that is what will happen. You see, at the moment, people think of these things. These are two kinds of entities who have the job of making things public. The library makes things public. The publisher makes things public. The sort of fundamental function is closely aligned. And they've simply emerged as different kinds of entities. And the publishers, I'll tell you what the biggest problem challenge is about publishing open access online. What do you not get? You don't get any publicity. What you lack is people going round to all the gatherings and, you know, with a stall and offer the whole business of telling people that it's there is something that publishers do. Therefore, publishers have a much higher profile. And what will we call the library when it's also a publisher? Will we still call it a library? Or they'll probably invent, I'm terribly terrible now. That's right. Now I'm involved in one institution where they've got a knowledge center. I don't think we want to be a knowledge center. But it is an interesting challenge, that one. But it's more that I just think it's an evolutionary probability that these things will come together. And it'll, like everything one forecasts, it'll take infinitely longer and be infinitely more painful than one would hope. Yeah. And as Manuel said, sorry, things are moving on my screen that library presses when they don't become fully commercial entities with the marketing, et cetera, often disregarded by academics because of notorious reasons of metrics, rankings, lack of reputation, et cetera. And it does take time, doesn't it, to build is it? Yes. And the whole business of journals sort of complicates things, doesn't it? I mean, the fact that over in the sciences, everything's happening in journals and that we're still over in the humanities, we're all still wrestling with monographs. Yeah. And then Chris Banks has put whatever you call it, students still call it the library for my minutes. The library is the new library and we continue to involve. Oh, I think that sounds, Chris, I think that sounds completely right. And it's a very positive word. Notice that people like to describe their product as a library because library is a word that people like, that the associations are so good and we really have to hang on to that. That's why my librarian is Virgil Muttl, it's really important. Perhaps all librarians should dress as Virgil. Yeah. And then I think Chris said what I was thinking, library is a word that is trusted. It's trusted and impartiality. Yes, still, still. Whereas you wouldn't say that about publishers. And I wondered, sort of looking back at when you have published things online, is there anything now that you sort of would do differently and have, like, as you've gone through later versions of putting your work online? Well, it's, I mean, one thing is that because I've been doing it for 20 years, there have been dramatic transformations in the possible. So there wasn't really, back in 2004, when my first thing came out, there was no data linking because there was actually nothing to link to. So the publication I'm working on at the moment is absolutely stuffed with wiki data connections, identifiers and things like that because you can. So I would have been glad to be able to embed things more. I think the next big thing, I mean, one of the things that interests me a lot is the multilingual. There is no reason why something shouldn't be in more than one language. And I'm also experimenting with something which I think we're not quite here, which is the invite other scholars to contribute and comment before you go public. That nowadays is much easier than it was in the past. I've been experimenting with it and some scholars, younger scholars are perfectly happy to engage and older ones can't understand what it's about.