 Welcome to the fall meeting. I'm delighted you're all here. I do want to take a moment to particularly recognize three groups of folks. One is our international participants. If you thought it was a long trip from I don't know Chicago, I suspect it didn't get any easier from I don't know London. So I'm particularly pleased to welcome our international visitors, members, and presenters. Glad you're here. I also want to take a minute to welcome the clear and ARL fellows that are here. As you may or may not know, CNI doesn't have an explicit leadership development program of its own. But we try and support both the clear and the ARL leadership development programs, both of which are very important, I think, in helping to develop new leaders for our community by inviting the fellows from both of those programs to join us at our CNI meetings and become familiar with what CNI is doing and what our members are doing. And we have an unusually good turnout here. The clear fellows have scheduled their own meeting coterminous with this. So we have a particularly good turnout from there. And I just urge you as you meet these folks at the reception or at the sessions to introduce yourselves and welcome them. I am very, very pleased to be able to host those groups and to, in that way, have CNI play a part in the broader community leadership development programs. I also finally, in welcoming folks, want to welcome a few new members that we have. One is a group that goes by the acronym of JULAC, the Joint University Librarian's Advisory Committee. And they have joined us this program year. A second is Appalachian State University. A third is the Internet Archive. A fourth is Union College. And fifth and last, just joining us very, very recently, is Innovative Interfaces. So I'd like to ask you to join me in welcoming all five of those new members. We're very glad to have them on board. And that brings me now to my main topic, which, got to get my props in order here, which is sort of a year in review and a look at what we think is going to occupy our attention and I think also some of your attention in the 2014 year. You should have all received in your packet a document that looks like this with a shiny cover that's probably completely eligible due to those lights. It is the CNI Introduction and Program Plan 2013-2014. As with past program plans, it contains some background on CNI and a look at our major program commitments and focus in the coming year. It is also available, as if I believe yesterday, on the CNI website. And I'd urge you to share it with colleagues at your institution and beyond. I think it's a helpful document, not just in understanding what we're doing, but in more broadly sketching out many of the key issues around digital content and support of research, teaching, and learning. I would note that it includes a bibliography of recent publications and reports that CNI staff have been involved in and I believe on the web version of this those are linked to the source material itself. If not, you should be able to find links elsewhere on the CNI web page. There are a couple of sort of survey and synthesis pieces, particularly, that I've done in the last few months, one around research data management and one around personal digital archiving that might be helpful to some of you. So I do want to take us through at least some of the initiatives in the program plan and also to talk more broadly about some of the big trends that I'm seeing. There are some things I'm going to not talk about. One of the things I'm not going to talk about, even though it is prominent in the program plan, is some of the work that Joan Lippincott has been leading on digital scholarship centers. This, I think, is very important because it gets at vehicles for forging and sustaining collaborations in digital scholarship among various organizational components within our research institutions. And it also speaks to some issues around ongoing stewardship of materials that come out of that. The reason I'm not going to speak to it beyond that sketch is because she and her colleagues who have been helping her to explore this issue have got a session tomorrow, I believe it is, that will go into this in depth. So I would refer you to that session, which I believe we will also be videotaping for more on that. Another thing I'm not going to talk about very much right now is the outcomes of two wonderful sessions that we had yesterday afternoon and this morning of our Executive Roundtable. The topic specifically for that Roundtable at this meeting has been the acquisition, selection, and curation of e-books at scale by university libraries and also the connections and interactions between electronic textbook strategies and library collecting of e-books and some of the complexities that are emerging in that nexus. The reason I'm not going to talk about that very much is because I'll be doing a summary session on that tomorrow and we will also be putting out a short report as we have done for other recent Executive Roundtables of some of the key issues that were part of that focus. I will toss out a sort of a general challenge question in this area though that did come up at our sessions and has come up elsewhere and where I'd love to hear from people offline and that's this. Are there examples that you can provide out of the recent landscape of books that are being published in electronic format only or perhaps electronic format with a awkward optional print on demand but things where you fundamentally need to be looking at the e-book marketplace that contain high impact content. In other words, are we starting yet to see the market emerge where you've got to deal with electronically published material because it's not coming out in print in order to get coverage of key recent events or key recent thinking. And here I'm thinking specifically about books and the book marketplace and would give a certain amount of bias to things that are occurring outside of extremely narrow scholarly domains. I would love to hear offline about examples of that that you might come across because I think that is a bellwether for a lot of very significant developments and will ultimately trigger some important public policy issues as well. So I will toss that out as a general challenge. I will mention, by the way, for those of you who aren't familiar with the executive round table, it is described in general in the program plan. It's something where we put out a call for participation, typically a month or two before our meetings and if you've not, if your organization hasn't participated in it, I would keep a look out for topics of interest. Our preliminary thinking at this point is that for the spring meeting in St. Louis, our topic is going to look at some very specific aspects of software as a network-based service, including issues related to continuity of service and the ability to preserve materials that are being developed in these environments. So just something you might want to have an eye out for. Another thing I'm not going to talk about while I'm on the subject, at least very much, is MOOCs. It's kind of interesting, isn't it? If you go back a year ago, you could not convene a group of more than five academics for any purpose and get them to stay on topic on anything except for the subject of MOOCs, it seemed like. And, you know, while discussion continues about them, it certainly continues at a much lower temperature, shall we say. I would just make a couple of comments here. One is that there is some interesting preliminary kind of research that's surfacing that is starting to look at the characteristics of folks who seem to be most successful at MOOCs and at learning from them. And I think that work is worth having a look at. I would hardly say that it has matured to a definitive level at this point. But I think it invites us to go back to much more fundamental questions about, you know, that go back to the old characterization of the library is the heart of the university and the notion that, you know, really if you just connect a learner with a good library, they will be able to learn a great deal and go forward from there. And there's no question that that is absolutely true of some sector of the population. But it invites us to remember the nuanced distinctions between collections of knowledge, collections of open educational resources, which is, you know, to my mind knowledge kind of structured for learning and then actual open education, the delivery of teaching and learning experiences themselves. And perhaps it's helpful to think about MOOCs as a reason to, you know, once again revisit and reflect on some of those distinctions. The other thing that I would just say going forward as something to look out for is that in the early enthusiasm about MOOCs, there was a great tendency to see them as courses by other means. So there was a notion of, you know, the substitutability of a course and a course conducted by a MOOC. And I actually suspect that we will see MOOCs or MOOC-like things deployed for purposes that are not delivering traditional courses, but rather certain other kinds of training methods, things that don't fit well in the sort of traditional academic, you know, everything is a course framework. And I would be thinking about opportunities in that area for using these to enhance and compliment traditional kind of course-based educational experiences. I can't quite get out of my head the fact that Google decided to get into the MOOC business at least briefly. And what they were teaching was really what I'd characterize more than anything else as, you know, information literacy with Google. Just worth reflecting on, I think. Okay. So we've talked a lot about things we're not going to talk about. Let's talk about things we are going to talk about that are changing the landscape and shaping our program plan. And at least from a U.S. basis, I think it would be hard to not give very prominent place to the Office of Science and Technology Directive that basically said that federal funding agencies were directed to develop draft plans to ensure public access to both the published reports arising from research that they fund and also the underlying data that is produced by those research projects. Most of you probably know that there was an August deadline for submission of draft plans from agencies back to the U.S. basis of science and technology policy. Most of you probably know that those draft plans are not public and that, in fact, there's not a tremendous amount known about them, although there is a certain amount of mix of, you know, rumor and well-confirmed rumor floating around. OSTP is certainly, I think, been forthright that some of these plans are a bit more mature than others, depending on the agency. We don't, at least unless somebody's heard something that I haven't heard yet, have a firm schedule for when they will become public. But nonetheless, I think there's a good deal of momentum underway here, despite a government shutdown just to, you know, reduce progress a bit for a while, and that these developments in the agencies are going to reshape the landscape for the institutions that host these researchers as well as the federally funded researchers themselves. I think you can also see similar developments taking place in other nations and you can see a number of non-governmental funders moving quickly towards very similar kinds of principles. Now, one way to think about these developments is sort of very narrowly within the government sector and that this is a bunch of new government requirements that we're going to have to figure out how to support and how to ultimately track conformance to in much the same way that we're dealing today with data management plans and deposit requirements for national institutes of health-funded research and things like this. Very much to its credit, I think we've seen the leadership in research and higher education and here I'm thinking not just of the association of research libraries but indeed AAU and APLU as well, the Presidential Associations. Really looking at this as a invitation and a challenge to rationalize a lot of what's going on with the production and dissemination of scholarly literature and data and that that really needs to be done anyway. It's needed to be done for quite a while and that this should be the opportunity to really try and step up and do that. I think that a good response here would clearly be one that incorporates and accommodates the needs of the federal agencies but it should be one that goes beyond it. We're seeing a lot of changes in the obligations and practices that surround scholarly publishing today. It's not just federal agencies, as I indicated, we're seeing private funders step up here as well. We're seeing state-level mandates. We're seeing institutional level faculty open access mandates. We're seeing a whole range of behaviors that need to be rationalized in some fashion so that researchers are not left scratching their heads about how to comply with these or expending enormous amounts of time seeking to honor these obligations. This is an environment that's really in need of some serious streamlining. I'd note particularly, again just speaking personally and from my perspective, that we've seen this sort of unfortunate framing of various competing responses to the OSTP mandate ranging from the share initiative that I just described to a set of responses coming out of some of the STM publishing community under the name of Chorus to various governmentally based archives sort of using something like PubMed Central as a model to scale up from. All of these have a place in the psychology and I'd like to really make a plea and a pitch for looking for an ecology that makes sense and that uses all of these where they belong. Just as an example, one of the very useful things about Chorus, one of the things that I think is very attractive about it is that it makes articles available in context to readers within the context of the journals in which they appear and that's something that institutional repositories do not do as a sort of a natural course of event. We need to think about how to strengthen those kinds of connections and collaboration rather than to present it as a competition. I'd also just, you know, in thinking about this, note that a little bit of redundancy in these systems is not a bad thing. We had a very, very interesting applied case study happen a couple of months ago. The government decided to shut down except for its essential services and we suddenly got a real good understanding of how deeply entwined various governmental things are with the overall knowledge environment and scholarly communications environment. To make matters worse, at least in the initial shutdown, as I understand it, a number of agencies or resources within agencies that were deemed nonessential were actually shut down or had their home pages redirected to some machine saying, sorry, the government's closed. Come back later. They didn't even go the, you know, route of putting it on autopilot, hoping the machine doesn't crash and maybe we can at least get, eck out a few weeks that way and they'll get the government reopened. Although my understanding is that somehow later on some resources were moved from the first approach to shut down to the just leave it turned on and unstaffed model. So it's very interesting to look as a case study at what was unavailable during that period and what resources the government actually did deem essential. I would note that at least at some level they did deem PubMed Central essential and that did stay up although as I understand it was not ingesting new contributions. Those kinds of sensitivities to the continuity and good management of specific players within this ecosystem do remind us of the virtue of a certain amount of redundancy. I'd connect that too to some very interesting conversations that took place recently at a meeting with the rather forbidding name of Anadap too that was organized by Catherine Skinner and Martin Halbert and a couple of other folks. This was a follow on to a meeting that was held in Estonia a couple of years ago. It took place in November just before Thanksgiving in Barcelona and it assembled mostly national libraries, some research libraries to look at what it characterized as aligning national digital preservation strategies and that leads you to that horrible Anadap acronym when you pick the right words out of that. But the word alignment there is one that I think bears a lot of thinking about. One can see very clear advantages to aligning strategies at a nation state level. But one also recognizes that there are certain functions that nation states are unwilling to rely on other nation states for. They decide that these are critical enough and core enough to what they do that each nation wants to maintain a certain amount of autonomy and ability to carry out those functions rather than getting into one of these interdependent collaborations where you are absolutely reliant on someone else to do things to carry out your own core functions. This is a set of tradeoffs where we've talked a lot over the years about interinstitutional collaboration and when that crosses the line into interdependence where all of a sudden you are critically dependent on another institution rather than just doing things that are collaborative and build collective value. And I think casting it in the framework of delivery of services and preservation is a very provocative and constructive discussion that deserves our attention. I think as we think about how to design ecologies like the one that is shorthanded by share, these are principles that we need to consider a bit. So share, I also want to remind you, going back to the OSTP drivers, is not just an opportunity and a challenge to straighten out how we're handling scholarly publication. It also deals with data. And I want to remind you that there was a second executive order that took place over the summer which for reasons I don't understand did not get the same kind of airplay that the OSTP memorandum got which basically told federal agencies that as you are designing data systems the default thinking unless it has to be modified by considerations of security, confidentiality, personal privacy and other things, the default consideration and assumption should be you are to provide public access to data, to your data, to the data that your agency generates and exercises stewardship over. Another very powerful driver towards opening up data resources. Now I note by the way the word public there which is a very popular one in governmental circles. They are not using the rhetoric of open, they are using the rhetoric of public. And I don't want to read too much into these choices here other than to note that when you talk about public access and a commitment to make the public of the United States have use of data that the government produces, you can construe that very narrowly as you know there's a bunch of bits up here for FTP or other downloading. All the way through we are going to construct systems that help the public understand what in some cases is rather abstruse data that comes with a very complex contextualization. And I can tell you that there are people in government who share that commitment to making data public who are really struggling with where to go along that continuum that I've just outlined particularly in light of the helpful additional language in all of these directives which basically says oh and by the way there's no money for this take it out of existing resources and do it in your copious spare time. That adds an extra element of challenge to trying to take the broad interpretation of that. But I think we need to be mindful of some of this as we make our own plans for how to handle data. And I want to say just a couple of things about issues emerging in the data area. We spent a lot of time on that at CDI and other venues and you know there's a great deal of good work going on in this area now. It's quite striking to me how the research and higher education community is beginning to you know seriously mobilize to address at least the front end of these needs through a bit preservation services. I would highlight two areas that I think are going to be very very difficult. One is data that is constrained because it contains personally identifiable information or personally reidentifiable information. One of the things we seem to be learning again and again in different areas is that anonymization while it's a useful tool is a tool of very limited power and somewhat limited application. That it is frighteningly easy to de-anonymize data that one might think is anonymous and that we're going to need some really fresh thinking I believe about how to handle personal privacy and personal data in a world where we want to gain the maximum advantage from the free reuse and recombination of research data. The other area that I want to put on the table and it's one I've been thinking about a lot lately is the move away from people talking about open-ended commitments to preservation of data and the kind of thinking instead that you frequently are thinking now encapsulated in data management plans like I'll keep this data for ten years. I'll preserve the bits. I'll make it available. Probably within a ten-year life cycle it's manageable to document what's in the data so someone can reuse it. There is a better and even chance I'll be alive if they really get stuck and they need to ask me or one of my graduate students. It's quite a different proposition from I want to pass this data across a thousand year chasm and still have it reusable on the other end. So what they say as I indicated is okay we'll make arrangements to keep it for ten years let's say as bits. There are a number of commercial services institutionally based or consortium based kind of bit preservation services that are in place or going into place where you can prepay ten years. Ten years is a manageable thing to run a prepay service for. It doesn't suffer from the same kind of fiduciary you know sensitivities to interest rates and other stuff that 50-year commitments or other really long term things do. So you're seeing people step up for that and the general proposition is we'll go ten years and at the end of ten years we'll see what kind of use was made of it and go through some kind of process where we'll either decide to throw it away, fund it again for another stretch of time or perhaps find it a new home as part of that funding. We have no process for doing this at the moment. This is a process that will need to engage all kinds of scholarly communities and remember that at various points in data life cycles the different scholarly communities may be interested in the same data or alter their priorities for keeping it available. This needs to cross institutions. We're going to need registry mechanisms for this. Constructing methods for doing this are going to be absolutely fascinating and incredibly challenging. And I think we're going to see these in other settings and I'll return to that in just a minute, but I would certainly nominate this as an area that it is not too soon to start really thinking seriously and systematically about. Those ten-year countdowns are going to be upon us before we know it. It's going to come really, really fast. So this whole, this is an example of a broader issue that I've been looking at quite a bit lately and that I think we need to understand much better and this is transitions of stewardship. The notion that somebody's been taking care of something and now their commitment is expiring. Maybe they're going broke or maybe they just need to reallocate resources or something else has happened, but we need an orderly way of putting this information resource before the broader community and having a conversation and perhaps a decision-making process about is it worth continuing to keep it and if so who's going to step up to it? Huge challenge there. And one by the way that is growing some fascinating new dimensions. I'll just suggest one. We're getting very, very good at making digital surrogates of things. In two dimensions, we're really great at that now to the point where we can replicate essentially a good deal of fine art and other sorts of material and the big difference between the replication and the original is provenance. It's getting real hard to tell the objects apart. We're making substantial strides with three-dimensional works as well. I don't know how many of you noticed the conference that the Smithsonian put on a few weeks ago highlighting the work that is going on in three-dimensional capture and reproduction of various kinds of objects of interest to different communities. But that's a significant development. So now we actually have the opportunity to potentially kind of peel off some of the scholarly side of artifacts that are of interest for both, you know, sort of collecting and exhibition reasons and also as objects of study. And that's going to come into play, I believe, in some of these choices about stewardship transitions as well. The last thing I'll say about this area, and we could talk a lot more about it, and I hope we will over time, is that there are lots of cultural memory organizations or things that embed cultural memory organizations that are under tremendous stress. I'd simply invite you to watch the discussion about what is or isn't going to happen to the materials that are stored at the Detroit Institute of Art. There is a theory, at least, that is being put forward that since these are worth some substantial hundreds of millions of dollars, selling them off in order to help pay off the creditors to the now bankrupt city would be a very sensible idea. That, for example, these could be used to help meet pension obligations that the city is unable to honor. Without getting into the, you know, public policy pros and cons there, I think the important takeaway is just the sort of thing is in play, and there are a lot of, there are a lot of organizations under severe financial stress. So we need, we need, I believe, to be thinking about these kinds of transitions more systematically. This takes me to a final set of things I want to say in the digital preservation area. And these really deal with rethinking where we're assigning resources. And here I want to recognize a wonderful, succinct talk that David Rosenthal and Vicky Reich gave at the Barcelona meeting I alluded to earlier, which at least for me crystallized a lot of thinking I've been doing around this general topic. There are two things that are troublesome. One is that we don't know how well we're doing with a lot of our digital preservation work. We have some sort of individual snapshots. For example, the work that Bob Wolven and Oya Rieger from Columbia and Cornell shared with us a year or so ago, and we'll see an update on that at this meeting, where they tried to look at what part of their journal subscriptions at Columbia and Cornell were actually covered in the various mechanisms like portico and locks that we've put in place to try and preserve that material and got a disturbingly low number. Another area that has seen some academic investigation is how much of the web gets covered by various web archiving services like the Internet Archive, but in general we don't really have an understanding of the classes of things that are out there, and we don't have good estimates of what part of these is covered. We also don't have good estimates of where the areas of highest risk lie. We're very much in danger of recapitulating the old joke about the guy crawling around under the streetlight looking for his keys and he's asked why he's looking there when he lost him over there and he says, oh, the light's better. There's a certain tendency to go after the easy stuff here rather than the stuff most at risk. I'm starting to feel like really part of our strategy going forward here needs to become much more systematic. And that jumps me actually to the last couple of things that I want to say. I'm going to skip some other speculations in the interest of having at least a few minutes for questions at the end. Another place where we're seeing a kind of an emerging set of activities that need to turn into a system very broadly is in an area that I'm now thinking about as distributed factual biography. This is this whole massive issues about author identities, citations, aggregation of citations, interchange of these and compilation of these in various ways. It's connected to compliance issues for federal funder requirements that we talked about earlier. It's connected to academic assessment in various ways to social networking among scholars, to identifying important publications through citation networking and things like that. We've seen an enormous amount of activity in this area that CNI's been trying to bring back and a great deal of siloed work where we've tried to help promote and broker some conversations across silos. But really what this is creeping up towards is a place where we have factual biographies, we can fragment them into small pieces, reassemble them, distribute them, update them in little bits. We need questions we need to ask about this evolving world including what degree of assurance we want in it, how easy is it to fake things, to what extent do things need to be computationally confirmable. We need to think about privacy and what role privacy plays or should play or does play in the notion of factual biographies. Given that you publish things and that has its roots in making them public, is the fact that you published them a secret or should it be able to be? I don't think so, but as with anything in this area you can find someone who will argue that it should or it at least should be the author's option. There's an issue that comes up in that realm that's really interesting and that's what I'd call noteworthiness following Wikipedia. Wikipedia has this whole very complicated set of criteria for deciding whether your biography is worthy of Wikipedia. And actually this is subject to great debate apparently as some entries are proposed in there. And if you look at it, this actually has a rich, rich wonderful history. It goes back to the people who were preparing national dictionaries of literary biography in the 19th century and because these were going to be published volumes and cost a lot of money to research, they were quite constrained in the role they could represent and then had to figure out who was worthy in there. So you actually can make some notion about when does someone become a public figure and tie that noteworthiness into the privacy debate. I think there's a tremendous amount of work to be done there, but this again is going to be a landscape reconfigurer. So I want to close my comments with a sort of a final large trend that is showing up and this is a notion that I think was originally framed kind of elegantly by Chuck Henry at Clear. And that is this notion of what he calls coherence at scale. And he's actually set up a working committee to explore some of this, but it's the notion that we are moving past an era where we build fairly little systems and then federate them that in fact we need to be thinking at scale and need to be thinking about how systems interrelate to each other, how they in some cases depend on each other. You look at developments like Hottie or you look at the share ideas that I've sketched for you. You look beyond academia as well, but to places that are hospitable for at least some discussions with academia, Wikipedia, Google, Microsoft with its research, with its Bing system, Internet Archive. There are very fruitful discussions that need to happen across those boundaries as well as just within the pure research community. Look at the incredible accomplishments that DPLA, the Digital Public Library of America, has pulled off in the last few months as Dan Cohen has taken the helm there and really led them forward. And I would suggest that one of the most important things he's done to move that effort forward is not just to talk about what they are going to do in the near future but to be very clear about what they're not going to do in the near future and by implication saying someone else needs to do that we need to make some other provision for that that is consistent with DPLA and the broader community that DPLA aggregates. I think we're going to need a lot more discussion like that going forward and I genuinely hope that these meetings can serve and other activities that C&I will seek to advance can serve to act as a forum for those kinds of conversations. I'm really of the opinion that the sort of scale of engineering that we're looking at to manage scholarship, to manage research knowledge is crossing some fundamental thresholds and that we're going to need to do things very differently in the future than we've done in the recent past. Examples of this are all around us. I think you only need to keep your eyes open just imagine this one as a sort of simple example. Some of you are probably old enough to remember the Pentagon Papers back around the era of the Vietnam War, Daniel Ellsberg collection of government documents that he made off with which are very fundamental reference now to people writing accounts of the politics and geopolitical strategy of that time. Very important documents from a scholarly perspective that go very far beyond the immediate incident. That was a book, it was a fairly fat book if I recall, about a thousand pages but it was something that the research community knew what to do with as soon as it was published. What do we do with things like WikiLeaks which may very well take on similar properties going forward. What do we do with these sort of massive data revelations? I just throw that out as one of many, many examples it's just an area that's getting a lot of press and we could have a very interesting conversation about what are the right strategies in this particular area but I think it serves as a very simple example along with the records of any recent political administration and how those compare to the records of administrations 50 years ago about how we need to think differently at scale. With that I invite you to read more details and more thinking in our program plan. I invite you to work with us to continue to shape ideas like this, clarify them and advance them and I hope that I've at least given you a little bit of useful perspective on some of the places where I see the landscape changing in some pretty significant ways today and in the pretty near-term future. Thanks for listening. And I think technically we're out of time but I'm going to take one hopefully two questions anyway if there are some. There is a microphone here or if you just want to shout I'm happy to try and repeat the question. Comments are also welcome as well as questions. Any takers? Go for it Michael. I'll use the microphone since my voice is failing a bit. I'm Michael Siedl from Humboldt University in Berlin and I was particularly struck by what you were saying about the archiving and I need really to show, to prove that we are succeeding in this. We have a project, a national hosting project in Germany where we are really going to do some serious testing with various systems to have some actual concrete benchmarks to say what they can do and where they can't do things that are important for us in terms of the archiving. So I applaud that piece of your speech especially. Michael. Yeah, I it's hard to figure out here there's a tendency to keep improving stuff you have so to think in terms of how you can make things that are already inside the sort of general grip of our memory organizations and archives more reliably archived so we actually are seeing this sort of strange phenomenon now where we keep layering on more and more backup redundancy layers of copies and this does improve the reliability of it at some level and actually, you know, quantifying that is very, very hard for a lot of reasons we don't have time for going into right now but I'm starting to get nervous about how much resource we put into that investment versus how much resource we put in gathering things into at least some kind of stewardship environment in the first place. I have a feeling that that, at least for large classes of material that first gathering in and I don't even want to call it ingest because you know, that adds too much of a systematic flavor to it. I suspect maybe the highest sort of cost return investment we can make now in some areas. I think this really needs to be looked at carefully. We'll take one more in the back. Hi Cliff, Jeremy from University of Arizona. I really appreciated your mention of this sort of actual biography along with issues around research identification, social networking things of that nature. Have you seen any activity and what are your thoughts around the increase in faculty annual review systems that are being instituted in a number of research institutions and elsewhere and libraries involvement in both those activities and how they might leverage the data coming from those activities. Yeah, I mean these sort of faculty, you know, what have you published, what grants have you gotten, achievement forms. They are a lot of organizations collect them. They are one of the most hideous examples of silos I can imagine that are actually operating in the field today. Just to be as insulting as possible they usually collect the material in a really helpful form like PDF that makes it hard to repurpose or match against things. You're asked to give very similar data, sometimes the same data on grant applications. You need to go chasing around cleaning up your personal bibliographic record in yet another set of silos. I really see those as a fundamental component of both a source and a destination for this data we need to exchange around. Along with grant management systems and grant submission systems along with bibliometric systems we've got to bust down those silos. They're all moving the same data around fundamentally and I would say that in some other countries where the government in particular has gotten much more systematic about counting and quantifying faculty achievement as a condition of passing out budgetary allocations in higher ed. So think the UK with the research assessment exercises you do see a somewhat more sophisticated generation of those systems deployed than our common place in the states yet. Although I am starting to see institutions in the states who often have these things and they're often very much homegrown duct tape and rubber band kind of systems saying it's time to really look at this and either build a good one or purchase a good one. So I do see some investment starting to grow. One of the things we really need to look at hard and quick is interchange standards in this area which are kind of a mess. The Europeans have done some useful work in a set of standards called CERF, C-R-I-F and there's also some very good work coming out of the Vivo community that's worth checking out to do in that area. And now I've managed to keep us over time for which I apologize but thank you very much and I am just delighted you've all made it here. I wish you a great afternoon and a great day tomorrow and I just remind you we all need to be a little flexible with schedules in the weather and do check for schedule changes now and again. Thanks and good meeting.