 I'm Cliff Lynch. Let me welcome you to the Fall 2015 CNI member meeting. I am delighted to see so many people here. We have really good representation from our member organizations. We have a number of guests that I'm delighted were able to join us. And we have a very, very rich program for you today. I'd like to start by just taking care of a few housekeeping things. So far everything is happening on schedule, although I have heard some sporadic reports of delays into national earlier today. And we will adapt to those as best we can. There is a message board right by the registration table, which has, and will post any schedule changes there. As of right now, there are no schedule changes. There are, I understand, one or two speaker changes at a couple of sessions, but everything's going to go on as scheduled. If anything changes, check the board there. We've also put a list on that message board of the sessions that we plan to capture on video or voiceover presentation. A number of people have asked us to make that available to help them sort through which sessions they're going to attend. So that is up there for folks who want to consult it. I do need to make the caveat that you never know whether the recording is going to work until you listen to it. So that's a statement of our best intentions, not a promise. I'd like to welcome some new members and a rejoining member. We have several new members. These include the U.S. Military Academy Libraries, Dalhousie University Libraries, the College of William and Mary, DePaul University, and the University of Rhode Island. Rejoining CNI, we also have back the Library of Virginia. And please welcome those institutions and their representatives. I also just want to take a very brief minute to recognize a couple of folks who are in new leadership roles. I'm always a little hesitant to do this because there are always more people who just didn't get on the list for one reason or another because I haven't seen them or didn't think of it or was just being stupid or something. But I would just like to recognize a couple of people who I think are in the audience and I'll just ask them to stand up for a second wave when I mention them. We have with us the new Executive Director of Lita, Jenny LeVon, who I hope is here somewhere. Yes, welcome. We have with us, I believe, although I haven't actually seen him. I hope he's here. Robert Miller, the new CEO of Lyrasys. There he is. Welcome. And I'd also like to recognize Mary Malnero, the new Chief Operating Officer working with Steve Morales at DPN. Welcome. I also would like to just in this vein bring greetings from someone who couldn't be here to be recognized today, but hopefully will join us in San Antonio. And that is the new CEO of EDUCAUSE, John O'Brien. I've had a number of opportunities to chat with him since he took his position and I know he is excited about the role that C&I can play collaborating with EDUCAUSE going forward. So I hope to be able to broker that introduction in the spring, but he did ask me to send his best and say he was sorry he could not be here. And with that, I think all of my introductions and things are done and that takes us into the kind of year in review and preview. And I know you've found the C&I program plans for 15, 16 in your agenda book. I will mention a number of things that are in there, but not all of them because I don't think time will permit. I'm happy to field questions on those either in the Q&A at the end or later. But I want to kind of take a broad look around and put some of the work we're doing in contexts. And I have to say I found myself as I was pulling together this talk in a very strange place. I feel like we are in a time when there's a lot of transition happening, organizational transitioning, policy transitioning. We are seeing at least from where I sit the kind of character of innovation that's affecting our community shifting a bit. There are things going on at the fundamental technology and that will take a while to play out into real things at the kind of consumer market. And there's in some ways not very much innovation happening, but a lot in peripheral areas that's about to happen. And I'll go into more of this later. But we're seeing major shifts in the way people are thinking about scale and about interconnecting things. And I think those are very important developments that we'll play through. I want to start though with some discussion of public access, in part because this is an issue that's very much on my mind, in part also because it's clearly an issue that's on the minds of many people here. We did an executive round table, as many of you know this morning, dealing with funder mandates and compliance and related issues. And actually we're inundated by requests to participate in this to the extent where we put on a second session yesterday afternoon and still found it necessary to turn people away because we didn't have room. And those discussions have been quite illuminating. Where we are is, we've had a long discussion in this country and it's worth noting that there are a number of international colleagues, especially in various European nations who are farther down the road than we are. But we've had a long kind of discussion about public access to publicly funded research outputs that really has its roots in initiatives by the National Institutes of Health and then, more systemically, a February 2013 memo that came out of the Office of Science and Technology Policy on behalf of the White House that directed federal funding agencies to go and make provisions to ensure that both data and journal article publications coming out of federally funded research were publicly available. And then there was a very long delay. There was a long period of silence and wonder when these things are gonna come out and really most of the key agency stuff did not come out until the first quarter of this year. And much of it came out in the first quarter of this year with implementation dates that were out in 2016 or beyond. So they didn't take immediate effect. There was some time for people to digest this. So it's really just now I would say that many of our institutions are coming to grips not so much with the broad sentiment of public access to publicly funded research and broader initiatives around open data and open access, but the actual practicalities and specifics of those funder requirements for compliance to their specific roadmaps for making this material available. And this is very interesting because one of the things that is evident at this point is that there's a significant disconnect on the ground between what you need to do to ensure public access to published articles and the sort of conversations that many of our institutions have been having with their faculty about open access and institutional open access mandates. Put another way for some substantial funders simply placing your material for public access in an institutional repository or putting it in an open access journal is not necessarily sufficient to conform to the funder requirements. There are things we can do about this. Obviously one of the things that we need to get much more adroit at and to work with our colleagues at the funding agencies on is automatic propagation of articles from one place to the next so that if you deposit an article one place it can be propagated out to wherever the various funding agencies want it. But there's plenty to do here and I think that one of the things we're going to face over the next couple of years is sorting out a lot of confusion from our scholars about what does it really mean to provide open access to research results and how does that connect up to these funder mandates that they're starting to see. The fact that we have a wonderful assortment of funder mandates from different funding sources is probably not going to make this conversation a lot simpler. Certainly we're already hearing reports of substantial confusion among researchers about some of this. I think that some of what this underscores is the need for holistic institutional strategies, things that really carefully construct and try to be clear about the messages and the outreach that they're doing to faculty, to researchers. I think that we're seeing some very useful and helpful recognition that we shouldn't stovepipe articles over here and data over here that really this is about supporting the research community broadly in opening up access to this material. But I think that we are clearly facing a number of challenges here that I would say really we're not evident as recently as a year ago. I think there's one other thing that we need to be a little careful about too and that's over emphasis on compliance. One of the things that we saw starting probably four or five years ago was a serious commitment among many of our member institutions to work with scholars on research data management to provide the support that's needed for data and computationally intensive scholarly work. And that really wasn't compliance driven. I mean it was to an extent but data management plans really aren't so much a compliance issue as they are a challenge and an invitation to researchers to think about the management of their data assets as they define a research program and a lot of the support structure was not so much in merely writing data management plans but helping faculty to think that through and to give them services to build upon where appropriate. So I think that we want to be mindful of the reasons we're doing these things and that they really are not just about compliance. We certainly need to be mindful of compliance issues here but I think we also need to be clear that we're doing this for a much broader and deeper set of reasons than simply to check off some boxes on somebody's form. It's also interesting to see how some of these challenges of dealing in this new world are pushing various other implementation efforts and challenges. I already spoke about inter-repository propagation and that's an important thing that we've known we've needed for years. There's been some good work done on that particularly the open archives repository reuse and exchange efforts but there's plenty more to do. Versioning of articles and how we talk about versioning through the scholarly life cycle is clearly an issue that's growing in importance and is very messy unfortunately but I think there's no question that we're going to have to take it on and I'm hearing lots of work under way to address this in various contexts. It is striking as I speak to people how much uptake there is now on orchids. It seems fairly clear that these are likely to be the dominant author idea at least for scholarly workflows and since they play pretty nice with ISNES they should extend out into that namespace as well. A lot of institutions are starting to develop or deploy strategies to push them out to faculty on a systematic basis. I guess I would make a plea that you don't stop at faculty. There's a really high payoff I think in getting these out to graduate students early in their graduate career and having those there. It's I think a very powerful strategy to reach out past actual faculty. There are some other identifiers that are lurking in the wings that we know we need that we're collectively kind of looking at and figuring out how to engage fund draft and the whole work on identifying funders and funding programs is clearly going to be essential to coping in this environment. And then there's another one that makes everybody's head hurt. And that's organizational IDs. They're these are hard and they're hard for a lot of reasons. They're hard not just because of the technical issues but because they come with a whole pile of sort of legalistic baggage like who defines the substructure of an organization and what are legitimate departments within a university or centers or labs. How do these change over time? Who maintains these sort of identifiers that almost imply a org chart with history and how do we tackle the historic elements of organizational evolution? Now these are not unknown problems. I mean these are problems that show up in authority control. They are issues that show up in archives. There are people at organizations like the National Archives and Records Agency which have to deal with this because they're managing records that are attached to an ever evolving structure of agencies. Yet we can also see these fascinating connections between these identifiers for individuals and organizations and the way they're updated and the sort of world of linked data and of kind of omnipresent factual biography and of being able to represent currently active scholars but also the work of historical figures. There's a very interesting project that's starting to deploy out of NARA that had its roots in work that Dan Pitti and others did looking at archival name authorities and how to build those up. I think that these are going to create a very interesting kind of an environment where we have these sort of diffuse analogs of older biographical dictionaries and things like this and we can see some of the effects of this rolling out with Wikipedia for example which has more biographies in it than certainly than any biographical dictionary that I've ever heard of. The numbers I've seen are well over a million biographical entries in English Wikipedia at this point. That's a lot of biographical entries and one of the things that you may or may not be aware of is that there's this whole sort of quiet movement that's intertwining structured data with the textual entries in Wikipedia. So all of this biography in there kind of begins to connect to these name authority collections that are being developed. There's an interesting piece of policy attached to Wikipedia which you may or may not have ever looked at and that is a set of editorial guidelines about notability. Put another way, who gets to have a biography in Wikipedia? How's that for a contentious issue? Now, I think that we're clearly moving into a world where everybody gets at least a short factual biography who wants one. I mean that's sort of what happens when you populate your orchid profile. So I predict we are going to get into a set of very interesting challenges here about who gets to have biographies and where they go and who gets to update them and who gets to decide they don't have them. These tie into issues about the right not to just to be forgotten but to be known and remembered. And I think that we can expect some really hard issues showing up there as we go down this path but it seems to be an inevitable outcome of a lot of the thinking about identity and the sharing of scholarly activity and other kinds of activities and the sort of link data approaches and here I'm speaking very kind of broadly about the notion of interlinking data in various places and I don't want to bicker about RDF representations or something, I'm speaking on a sort of a conceptual level here. The specific details of linked data as all of you who've worked with it know are seemingly eternally moving target. So perhaps that's a good place for me to move over and talk just a little bit about some of the issues that are going on in security and privacy as we look at this landscape. And this is a world that is getting unbelievably strange and I think it's getting strange in ways that some of us haven't really had an opportunity to step back a little bit and consider. Let me just give you a few examples and data points here. So once upon a time, for those of you who studied computer science or cryptography, you had these very simple sort of threat models where you had Alice and Bob and then eavesdroppers and people trying to mess up the conversation between Alice and Bob and occasionally you'd also have Alice trying to talk to Charlie and you'd worry about Charlie and Bob conspiring in various ways. It was a pretty constrained set of threat models. The Internet Engineering Task Force earlier this year which has really in some ways become I would say almost traumatized by the Snowden revelations and really has launched a sort of a major effort to reassess and revisit security and privacy in the Internet environment. They actually issued quite a remarkable document about new threat models in the Internet and they actually identified threat models that included sort of pervasive monitoring. For example, if you could watch all the traffic that's going in and out, right, even if you couldn't see into the traffic, you can do a variety of inferential things about who's talking to who which fall very much outside of these kind of classic computer science threat models. It's really kind of remarkable to me to see the kind of foundational ideas here being readjusted for a very new kind of environment. Then you have other weirdnesses going on. All of a sudden, quantum computing is very much in the news and you have people talking about maybe practical quantum computing isn't as far away as it ought to be. One of the things that is just a little side effect of that is that if you could actually get quantum computing to work, it messes up a good deal of the cryptography that underlies the network. There are now programs underway, both in the States and in Europe, to look at essentially what kind of new cryptography do we need if and when quantum computing works. And one actually hears tell of people collecting up encrypted traffic now that they can't read because they figure they'll be able to read it someday. Very interesting kind of approach to things and one that's quite different, I think, in many ways than what we're used to. Another thing we've seen is just systematic security failures. Just scan the headlines over the last year or two. Look at the marvelous job that the Office of Personnel Management at the US government did. Look at what happened with Sony. Look at the numerous corporate events where they've had data breaches. Look at our universities and it just seems to go on and on and on. At this point, you have to start asking some questions about, can we really keep things secure over time? Probably not on the scale we'd like to. And once you start crossing that line, you have to start thinking about resiliency and contingency and what's really important here and where to focus efforts. And I think a lot of the thinking about security and privacy is going to end up reorienting that way. Let me just note another phenomenon here that is just fascinating to me just briefly. So used to be that most data breaches were either criminal and intent. People were trying to steal credit card numbers or money or you know, siphon off accounts, identity theft, this kind of business. Or they were espionage like in character, maybe industrial espionage, but taking secrets. So now what you have is you have these breaches where material is purloined and then made public. And it's often not entirely clear why, what the motivations were of the people who do it. It's clear that at least in some cases this has a very large impact on international politics in diplomacy and public policy. Think of again the Snowden data breaches for example. Very clear that that material is going to be consulted for a long time by a lot of people as evidence in various discussions. Think about the Sony data breach. It's really not clear though that the material that's being put out there publicly by these folks who cause these data breaches. Is complete or unenhanced. It's really often very hard to tell what you've got there. One of the most amazing stories for those who like to follow this sort of thing was the Ashley Madison data breach. I see a few people nodding about that. So Ashley Madison was a company that actually was just getting ready to go public at a significant price that ran ad campaigns essentially depicting themselves as an online affair arranging service. And somebody basically pulled the database off of this. All the profiles and put it out on the internet. And all kinds of interesting issues ensued. So remarkably the first thing you got to understand is database. And I'm going from memory here about 35 37 million people in it. Astounding. Now not clear it's a real high quality database because you can sign up for it. Apparently just by putting in an email and there's no handshake. So maybe it's your email maybe it's somebody else's email. Maybe you just thought it was fun to sign some other people up for it in case there was a data breach later. So people start analyzing this thing of course. And they notice well gee we got a lot of members of parliament in here. Got a lot of Gov addresses all kinds of good stuff. Nobody knows of course whether any of those are legitimate ones or not. So and who knows. Who knows if the people who pulled the database added a few just because it was fun. Or took out a few addresses before making it public. We'll never know probably. Other people have been analyzing this file in other ways since by the way. And there's now a sort of a small statistical industry in writing papers about anomalies in here. It appears for example that while something like 10 or 11 million entries were coded female that almost all of them were computationally generated and imaginary. There were really like 10 million imaginary entries in there. Interesting as an insight into something. But I just leave it there. The question I would ask though without putting too fine a point on it is there is a long history of Perloin documents of various kinds. Changing public policy. Changing history. Having big impacts on society. And being very very legitimate and important ongoing evidentiary material for the kind of scholarly work that people in our institutions do. Think back to some to things like the Pentagon papers for example. We're now in an age where we don't publish these things they live online. And I think you can ask some very confusing questions about who should be looking after these when and if archival commitments should be made to maintain these. How we attempt to validate and understand the quality or accuracy of these kinds of things. I don't think it's good enough to just say oh this is messy you know it's these this is Perloin data we'll just wish it goes away. I think I think we're in a world now where it's not going to go away. And we need again to think through a nuanced response as a part of this whole rethinking of the security and privacy landscape. I need to keep moving though. There's so many things to talk about. I'll say one more set of things about the privacy area. As I think many of you know CNI has gotten quite interested in privacy especially as it interacts with things like the widespread and somewhat profligate use of analytics of various sorts that user tracking in various contexts and I think we'll hear a little bit about that at the closing plenary has moved to you know levels that are really hard to imagine. We're building all kinds of potentially creepy things ranging from you know e-text books that tattle on you all the way through companions for children in the elderly that you can talk to that send your voice out up on the internet so that it can be processed so that you're you know talking Barbie or whatever can say the right things back to you and maybe tell your tell your parents what you've been saying to your doll or whatever. We're into a level of things that listen and collect data that is is really almost unprecedented and I think there are a couple of places where you know this falls pretty squarely into areas that we ought to be considering here. The whole question of reading analytics I think is a very real one for example. CNI held a small workshop which I'm still trying to get the write up done for this spring which looked at a few of the privacy and security issues and one of the one of the things that we came out of that meeting with as a very strong consensus and I'll just say this in a kind of a general and equivocated way is that people should have an expectation of reasonable privacy in their interactions with various kinds of networked information resources and in cases where that can't be met they should be at least informed that you know their expectations are not being met and that is being collected and it was fairly clear that you know we've been generally doing a reasonably horrible job of those ranging from you know real no-brainer kinds of things like opening up connections to eavesdropping by default by not using secure HTTP and you might think well gee you know is he gonna get his tinfoil hat out next? You know how much thread is there really in tracking this stuff? Have we got spies crawling around you know listening in on on people's interactions with science director the local online catalog or something? Well you know actually given how pervasive wireless is for the last mile this is no longer you know the world of spies and wiretappers and things like this it's so easy to collect this stuff up off of unencrypted Wi-Fi that these you really are holding these conversations in public if you're not encrypting them end to end but at least as serious as the technical stuff is the need for a hard look at some of the contractual provisions about data reuse and collection by some of the some of the information providers that our libraries and our institutions contract with and I think there needs to be a fairly careful look at that. NISO with some funding from the Mellon Foundation has carried forward some parts of this conversation throughout the year and just the other day literally I believe put out the final version of its effort at distilling some principles that are relevant and and perhaps a helpful start here but it's it's really clear I think that we have some homework to do in this area and that there's also a reasonable amount of confusion in in conducting these conversations for instance I found widely differing opinions about exactly what the state was of authentication and authorization handoffs between institutions like universities that do site licenses and content providers like journal publishers or database providers. We're going to actually go out and collect a little data on that and you will be hearing from us with a couple of questions on that sometime in early 2016. There seems to be enough disagreement about what the facts are that maybe it's time to get some so we will be moving ahead on that. I just want to say a few more things now about scale and about innovation. I get very confused when I look at the landscape now we see a lot of striking innovation showing up in areas where information technology is getting combined with other things you know drones autonomous vehicles various kinds of machine learning driven things robotics of various kinds the personal companions I was talking about a minute ago those kinds of things but in terms of kind of direct technical innovation I can't point to a lot of things I've seen in the last year that are like this is really new and it's a game changer maybe some of you have different opinions about that but I actually feel like a lot of the kind core information technology that we've been relying on is now in a mode where it's doing a lot of aggressive four stops lessons it's generating a lot of churn it's costing a lot of investment in individual and institutional time to stay current but with often rather limited payoffs in terms of what you get for doing all that work we'll see whether that continues I think it's fairly clear that if it does we're going to need to start struggling with how to slow down the hamster wheel because just running really hard and burning a lot of resources to stay in place is not where we need to be I do see huge progress though on the scale side we are starting I think as a community haltingly and painfully to think about operating at scale and about hard issues of genuine interoperability trying to consciously and thoughtfully reduce redundancy when it's not useful for some reason I'm very encouraged by some of these developments like the international image interoperability framework that it was striking to me in looking at the submissions for breakouts at this meeting not just the ones we accepted but the ones we didn't have room for at how many things now we're employing or building on that to interoperate across collections and pull down silos I think that you know again we're seeing people think about how do we interconnect A and B as opposed to having a silo and we need a lot more of those conversations I want to conclude just briefly with a few conversations about where we are on preservation and stewardship and I think when we look at you know the sort of fundamental technologies of preservation there is good progress happening here there is progress happening on everything from the managing the ingest side all the way through trying to get emulation kind of environments to the point where they are a genuine reliable production worthy kind of tool in the in the world of preserving materials I had an opportunity a couple months ago to preview a version of the talk that David Rosenthal will be giving later in this session about I'm sorry later in this conference about emulation he got some funding from the Andrew Mellon Foundation to take a look at this and the progress when you look at these systems is really quite substantial there is still a world of headaches but most of them as with so many things in the stewardship area are now not primarily technical they are about licensing and intellectual property and all of this kind of wonderful stuff one of the things though that I think is really troublesome when you back off from the base technologies of preservation to the sort of broader picture of stewardship and the work of stewardship organizations is the way that our memory organizations are increasingly being cut out from collecting various kinds of material this just gets a little worse every year it seems there's more music that's digital only that organizations can't legally collect and add to their collections because the license agreements don't permit it there we're seeing more and more video the growth of streaming models as ways of moving around cultural materials I think there are some genuine you know positive things that are popping up I was in many ways very encouraged to see the new random penguin ebook policy which actually you know really talks about we're not going to simulate wear and tear or have our books self-destruct these are intended to be long-term you know perpetual licenses essentially and that's a really really encouraging development I think the I just wish we had more of them to point to I also think that there's another problem that's getting increasingly clear and it's about preservation and stewardship but it's also about accountability and maybe I'll close with this because this is a really fundamental kind of conceptual problem that I'm not sure we have got good traction on yet so once upon a time you used to be able to say well you know if I could capture the content of the database I would have captured in some sense the behavior of the system so you might talk about the news a newspaper moving from printed sheets to a database and so you say okay as it moves what I need to do is capture all the stuff in the database with the appropriate time stamps about when it went into the database and when it was modified and I'd have a pretty good sense of what's going on well if you look at the digital newspapers such as they were of 10 years ago that's probably true but now you have an algorithm interposed between you and that database that is very complicated it's not just a pull out these three things that are the headlines and then you know everything else it's it looks at who it thinks you are what your history is with the site where it thinks you are geographically it looks at a whole pile of things so you essentially are getting a personalized experience with most of your interactions with the web now to a greater or lesser extent and you are getting that personalized experience mediated through an algorithm that is hugely complicated and that in some cases is getting to the point where nobody really understands it or understands how to document it or the changes in it and they're actually doing a b testing to kind of tweak the algorithm and find out what the tweak does because the effects of tweaking the algorithm are very unpredictable we have no idea how to document the evolution of these kinds of complex network based mediating algorithms and this is a preservation problem to be absolutely sure a stewardship problem but it's also an accountability problem given how many things now are using these kinds of algorithms to identify you for attention of various kinds establish a credit rating for you decide whether you're you know at risk for dropping out of school you name your favorite thing these algorithms do a lot more than show you what they think you believe is the news that you would be interested in and I think we're going to need to think really hard about how to come to terms with this kind of an environment where algorithms and algorithms that most importantly are both complex and dynamically changing personalized experience I don't know what the answers are here but it is absolutely clear to me that as we look at documenting the digital experience interacting with various kinds of content resources going forward that this is going to become a more and more significant practical rather than simply theoretical conundrum we'll continue of course to push on that kind of an issue so I hope I've given you a little tour here of some of the curiosities and developments that are fascinating me over the past year and some of the issues that I'm taking away from them that we need to be thinking about on our agenda there are certainly plenty of them and as I say I think we are in a time where there's there's a lot of transition taking place and where in some cases we're being challenged to if not change our direction at least change our approaches or revise our strategy in response to unexpected forces that are are reshaping the the directions we're going forward on whether they're funder mandates or real genuine changes in our ability to be capable of securing digital information at scale I would be very pleased to take a few minutes to answer questions about the program plan what I think is going on in the world things that we should or shouldn't be doing or anything else so I thank you for spending some time with me and the floor is open thanks oh come on we must have some takers here comes somebody good afternoon joy sogburn appellation state university I think your spot on cliff and thinking about algorithmic mediation or whatever we want to use I'm working on a chapter for something right now where I touch on that but I think how machines are now writing without us knowing whether they're it's a machine or a person writing and whose authority and and accountability all those things are all bound up together and I think it's a worthy subject for us to all look into because it affects scholarship and people all over the world to him various ways thanks yeah there this phenomenon of machine authoring is something else that these algorithms are doing I don't know if you know but a good goodly number of the sports reports for example the small time stuff you know like high school football games is actually being written by machine now you just yeah feed it a couple of data points and it spews out a little article suitable for the wire services I think they're also starting to do this now with things like quarterly reports from corporations from time to time so it's getting to be an interesting world out there David hi David Rosenthal from Stanford so I want to go back to your point about who's responsible for custody of leaked information this is a problem that we've encountered at Stanford uh Stanford lawyers in pounded a librarian's hard drive because it had wiki leaks on it and I believe that the reason was that that's classified material and Stanford does classified research and it cannot have classified information outside government approved areas and I think that this is a this isn't just a problem about classified information I think people's lawyers will be nervous about information like the Ashley Madison database and so on for liability reasons and so on so I think this is an area that needs looking at it is and it's it's very complicated I mean I have heard similar stories about material being pulled the Snowden material seems to be a particularly good case in point here because lots of people refer to it or might show a slide from it as part of a discussion about security practices or eavesdropping practices and apparently all it takes is you know one of these slides in a video of a guest lecture someplace to all of a sudden cause hysteria and the guest lecture to be removed in some cases it's it's very it is very problematic and it's it's something that the labels we put on things make a big difference for example a number of our our great research institutions have in recent years built up collections of human rights violation evidence they've made that an important institutional priority and you know when you frame it like that it sounds like it's much better but in fact depending on who you ask those materials are very controversial and certainly have been subject to ongoing attack from many sources so I think there's really quite a nuanced discussion to be had here but it does seem like there's a lot of this is material that research libraries should be providing to scholars but it may be very difficult for them to do so yeah and maybe that maybe some of the solutions here are our international ones as well as national ones um but uh you know this this is a problem that's that's nasty but I don't think it's going to go away it's going to get considerably um more complex I fear other questions comments oh Peter uh Peter Bernhill University of Edinburgh um I'd be interested in the extent to which you think that the concept of published heritage is useful or not because one could have thought that libraries in one day used to spend their time with well published material which they then organized and made available and then there was that gray literature stuff which was somewhere else but the idea of that which was not published in some sense was not principally its concern except in special collections and therefore you sort of get towards the archival side but now clearly so many things are made public through the web one way or another some very deliberately by the author or publisher or those that had rights to make it available but just sticking with that rather than the inadvertent so to speak um where is the role if you like of the research library vis-a-vis the published heritage or is that concept now disappeared because nothing is really well published these days or at least so many things are variably published so so um I need to fess up that I'm not hearing the term published heritage for the first time um Peter and his colleagues um at Edina and uh the University of Edinburgh put on just an amazing uh small conference and workshop in September that among other things looked at some of these issues and um uh there's a report from it that I um I strongly commend to your attention um I think this is this is this is a wonderful and absolutely core question and um you you can work on it from two directions you can ask what does it mean to be published in the digital age or you can work at it uh you you can work it from the point of view that um you know thinking about published is no longer useful and um we need to think about much more about sort of holistic um uh social outputs cultural outputs um uh very very broadly defined and there's probably some merit in um in in both of those um uh you know published has always brought along at least some baggage of favouring things that that work inside an economic sphere and certainly there are um enormous amounts of of cultural production that that don't fit in there and in some senses the internet has really enabled some of those things to thrive on the other hand um uh the notion of published as in made public and that using that as a um you know as as a selection line um maybe is there is a is still a really fruitful idea so I find that phrase um really really resonant and and one that um that deserves a lot of thought and scrutiny um and I appreciate you bringing it forward okay just one more I think people are ready for a little break thanks so much and let me wish you a wonderful conference