 A few disclaimers to begin with. I don't have any happy stuff in my presentation. So apologies in advance. And also, I will talk a lot about context during the whole presentation. So it's only fair that I give you context of where this presentation is coming from. I am not a librarian, but I believe that I'm an honorary one by now. And I've been working in libraries for more than 12 years now, something along those lines. I am a computer scientist from that ground, and I'm very passionate about technology. And I've been struck by lots of excellent conversations that we've had so far. So I'll try to integrate some of those in the presentation today. So starting from something that Robin said, I am the head of digital innovation at Lancaster University. But since first of March, I'm also the head of research services, which is not something new. It's just a slight reformalization of the role. But what that allows me to do is link digital with research quite easily. And I think that would be the focus of today's presentation. Another thing I'll be talking about is forward-looking approaches. I won't be talking much about what's happening now, but my feeling about what might be happening in the future. And I think my success criteria for today is if you go back from this presentation, feeling happy, sad, confused, that's fine. That's my success criteria. So I'll cover all the grounds, basically. Right, so starting from what I'll talk about, I'll talk about technology and leadership. I'll talk about some of the key challenges that are coming our way. And I'll talk about the digital skills gap that might be emerging because of that and what we can do to handle that. Also, feel free to tell me to move if I'm in front of anyone. So apologies for that. Let's start with technology and leadership. So my thinking when I started reading a couple of articles more specifically, one of them was digital disruption in academia. Are we ready for universities in 10 years? And actually, there's a lot of similarity between the taxi industry and the academia in a way. We both have slight monopolies on the market. We haven't changed in a long, long time in bigger capacities. And actually, technology has the chance to revolutionize both industries. It just happened for the taxi industry before it happened for the academia. The other one is greater technology and great unburdening of higher education, which is more focused on the banking industry, but it's worth a read. So I would encourage you to read it afterwards. Right. So even before we start on what technology can do is the hype reel. We hear about all these technologies and we're like, it's never going to happen in academia. It's never really going to happen. So let's look at that. So I'm going to show you the Gartner hype cycle from 2010. So Gartner is a large commercial entity that creates the technology hype cycles. And this is the hype cycle from, they usually release it in July, August. And this is the hype cycle from 2010. And I picked those technologies that Gartner predicted will become a reality in industry or become mainstream as per their language in either less than two years or within two to five years. So I've given them a bit of margin for error and said, okay, let's talk about 2017 instead of 2010. And just your recognition. So a lot of my nieces and nephews are playing Nintendo V and Microsoft Kinect. It's getting really popular now. Micro-globbing. Some of a lot of us have been tweeting. It's very common now. 3D-flag-panner-TV, social analytics, private-cloud computing, which is actually happening in academia right now. So Lancaster has a private-cloud computing infrastructure. Media tablets. Hello, iPads. Biometric authentication methods. Hello, Apple Pay, Android Pay, Alipay. There's so many going on. Virtual assistants, Siri, Android, Google Now, et cetera, et cetera. Cloud computing, ebook readers. That was a big shock. Kindle, et cetera. 3D printing, wireless charging, augmented reality. And some of them are astonishingly mainstream. Location-aware applications or mobile stores. It's like natural now. No one even thinks about when they were actually coming in practice. So it's a bit scary that technology that's predicted to be true is actually becoming true. And this takes me back to some of the conversations we've had where Claire was talking about, we can't really relate to digital in the same way as physical in terms of books. And I partly agree with that. But I also think that technology has only been there for about 30 years. Give it the same time as books, and it will be a completely different time period and completely different feeling. Let's look at some numbers. So we in the library are really like to look at numbers. So in 2010, in this world, we had 5.2 billion internet connected devices. There were 9 zetabytes or Zetabytes the way you can get off data. For those who are curious about the zetabyte, that's 1 billion terabytes basically. Content is king. And I think as libraries, we are a path to high quality content and information. That's our core purpose in a lot of ways. And content is king. There is no denying that, especially in 2010. But the content to information transition in 2010 is a lot more simpler than content to information transition that's going to happen now in the future. And I'll talk a bit more about that later on. And the kind of leadership we used to have, at least for 2010, or we should have had at that time and definitely did, is either research or academic leadership. And I will raise this question for you to think about. For all the directors or entities who are sitting here, how many of you have come from research or an academic background? My gut feeling would be about 80 to 90% would have come from that background. Now let's fast forward to 2017. 28.4 billion internet connected devices. So we are talking about roughly 30% of the world population or something like that. 17 zetabytes of data. Context is kingdom now. So content is still king. But without the context in which it's created, it's becoming more and more meaningless. Contrary to understand what this content is talking about, and the fact that it's so massive in its nature, the context is becoming even more important. And the kind of leadership, in my opinion, that we need now and for the next few years, is one that starts with a strong research background but with a good grasp on what digital challenges might be coming our way. And if I try to fast forward to 2025 now, there will be about 80 to 100 billion internet connected devices that's roughly about 62 to 65% of world's population. 180 zetabytes of data. And there's a big reason behind that data explosion. In 2010, on average every human that was internet connected had about four sensors on them at any given time. In 2017 we have an average of seven sensors and in 2025 we have an average of 12 sensors at any given time. And if you're looking at humans like me, they're probably talking about 16 sensors, different devices. The thing that will become very important for us to continue on our pathways to high quality information would be looking at cognitive load reduction, would be looking at context program technologies. There will be so much content that it will become really important to focus people's attention to the content that's very relevant to them and based on the context that we've derived from their backgrounds. And I'll talk a bit more about that as well. And the kind of leadership we're starting to look at in 2025 will be digital leadership, but not on its own. You'll need people who also have a very good understanding of research and academic backgrounds. You'll also need people who are very good with emotional intelligence. And for people who come from digital backgrounds, that's not natural. So here's a generic plea from me to all of you. When you're thinking about digital leaders, start thinking about it now for people who can either be developed into digital leaders or digital leaders who can be developed into better emotional leaders as well. Right, so that was my 2010 plea. Let's talk about 2016's partner lifecycle now. So there will be lots of big buzzwords there. Context brokering, data broker platform as a service, quantum computing, et cetera, et cetera, et cetera. Sounds scary to me. But based on the principle that 2010 became a reality in 2017, is this going to become a reality in 2025? And if that is the case, we ought to be doing something about it now. By the way, this is not the full list. This is the list that Gartner predicts would become a reality in under five years. It's not the 10 year plus list. Now some of this is already happening. So there's a lot of research happening on connected homes, healthy aging, et cetera, et cetera. Gesture control devices is also something that's happening at the moment. In fact, there are a lot of interesting YouTube videos on how to control drones using gestures if you're bored, have a look. But there are some really interesting concepts here. Something that's very important for us as academic research libraries, which is about smart data discovery. A lot of focus that we have at the moment is we need to have all this data. We need to preserve all this data. We need to make sure that that's sustainable. But for what purpose? If you can't drive innovation out of it, if you can't make meaningful decisions out of it, I don't think that data will be that important. And we need to be prepared for those smarter decisions in the future and prepared in the way that that data needs to be stored, that data needs to be preserved or accessible, that those smart decisions can be made out of it. Otherwise it's practically useless in some capacities. There are other areas, natural language question and answering. I'll briefly talk about that and how that has the potential to change some of the nice information that we were talking about as well. Moving forward, if you're not really sure about Gartner, this is the NMC Horizon Report for Higher Education and Technologies. This came out in 2017, by the way. It's extremely recent. And again, there is an NMC Horizon Report specifically for libraries back in 2015, if you're interested. But this was 2017 Higher Education Report. It's even got bigger buzzwords in it. So it's even more difficult to explain how we'll be able to do all of this. But again, going back to the point that Claire mentioned, if you notice in the enabling technologies, there's a technology called electro-vibration. And that technology is all about actually getting physical sensation from digital technologies while you're interacting with them. And those are the kind of things that might have revolutionary changes or flexible technologies. In fact, only about two weeks ago, there was a petition fired from Apple about a roll-up iPhone screen. So you've got flexible media technologies in that way. And what kind of impact that might have for us in the future. This is an American report, by the way. So LMS over here actually means VLE in UK terms, not the library management system. Moving on, I mean, don't get me wrong. I'm not trying to scare you. There are some really good things over there. There's digital scholarship there, which we are very proud of and the services that we've done in that area are preservation and conservation technologies. Those are things we should be very proud of, but there's a lot to look forward to as well. Right, moving forward. So how does technology change our approach towards research support and enablement? And that was a question we asked in the leadership team at Lancaster when we were developing our plan for 2020. And one of the questions that was in the back of our minds was should we be a research support library or should we think bigger than that? And actually, one of the things and a big question I'm not sure whether he's here at the moment, but we actually acknowledge that we don't want to use the word support and I think there's a different connotation associated with that. So in our vision for 2020, and I'm just quoting a specific part from it, but you're very, very welcome, by the way, to look at the whole strategy if you're interested. We've changed the language. We will work with the research community and other partners to create collaborative research environments. We will build effective training programs and provide developmental opportunities. And also, we've not lost the sight of looking back because we are very keen on making sure that our special collections are also a key strength for us. So we are looking forward and looking back at the same time. But looking forward, we're not talking about research academics telling us something to do and we'll do it for them. And actually, we will be giving them developmental opportunities at the same time as well. And I'll talk a bit more about that going forward as well. Right. So that was about technology and the kind of context of the technology that we're talking about. Having said that, we can't really do it all. So what are the different areas that Lancaster is trying to attempt to do to begin with? We've checked about eight different areas that we're going to work on in the beginning in my team. One of them is approach. And again, I've already done the investment in digital leadership but that's one thing I will keep doing through and through. But we are also writing something along the lines of library futures. We have a digital futures paper that we're currently composing which is going to be a hybrid of macro and micro elements. So we're looking at the seven to eight years timeline but we're also looking at what we can do now that has the potential not to revolutionize but also to add value to our existing services. And we often lose the sight of adding those values and as Mike was also saying, those little changes can make a big difference as well. We're also looking at behaviors and behavior is quite interesting and I'll actually have more detail slide on this. In my opinion, we talk a lot about being a data-driven library or an evidence-driven library. Actually I need to think bigger than that. Data in its own right will give you a very concise picture about a particular scenario for a particular person but it actually doesn't connect all the data points together. It doesn't tell you about the behaviors that you're interested in and that only comes when you apply data with context and then deliver those behaviors out of that. And I'll talk a bit about that but the initial steps we're taking in that way is we're developing a large-scale analytics infrastructure in the library and also we're developing an intelligent library building which actually not only tries to gather the traditional senses of data but also the contextual data for those people and try to merge those things together. We're also looking at consistency. It's a very, very uncertain environment at the moment financially, politically and technologically. The best thing we can do to make our staff feel comfortable and to keep our value in the industry is to provide that consistency but at the same time, again, I'm not a librarian so I probably feel like I can say this, I'm sick and tired of library saying we will adapt to the circumstances. We've done adaptation for a long time. That's what we're good at but let's think about creative solutions that not only provide consistency through that but go beyond that and actually provide efficiencies through a lot of different things and how do you provide that consistency through that creativity and it's about the kind of workforce we have. Let's develop a workforce that's more innovative, that's more entrepreneurship and it's got that quality in it and who believes in partnership, not collaboration. Again, we like to collaborate where one party does all the work and the other say, have you tested for you? No, let's develop things together. I think we need to have a stronger model of partnership not collaboration anymore. And be prepared to introduce diversity of thinking in your teams. I think diversity of thinking is the key to creativity and innovation. So add roles like research software engineers Patrick was mentioning earlier or data visualization experts or data scientists or research scientists in your team in the future. If you truly want to be a research library of the future we need to start incorporating that diversity of thinking now. And context I briefly mentioned already but we need to start investing in cognitive load detection area. We need to start investing in context broken personalization relevance. So as librarians we have this thing saying we want to see consistent results coming back from our discovery services every time you use the same search words. That's no longer going to be the case in the future. Actually it depends whether you're searching it as a librarian or as a student in a particular situation in a particular environment, studying a particular course why give them all the other data that is not really applicable to them? They've got too much to look at already. So that will become really important in the future. The other four areas that we're looking at is ecosystems. So we have this tradition of investing in products. We buy one product for one task, one product for another task but we don't really look at how they integrate together how they've seamlessly provide that user experience at the end. And unfortunately we've done that for too long. There's no golden solution to that except for the fact that we need to be a bit more proactive in incorporating ourselves into the procurement processes to challenge vendors and say are you based on open standards? Is your product going to work with all these products and prove it before you buy your product? I think we almost need to do that in order for us to become that interoperable in the future. And again that's a big problem to speak to anyone and they talk about interoperability. We need to embed ourselves in a different stage of research and I was really pleased to see Oxfly was talking about how well they've embedded themselves into the research processes but we at Lancaster have not done that very well and I'm sure others can do that more effectively in the future too. So one of the things we want to do is get out of that time frame or that mind frame that we've done the research here are the outcomes. Can you do something about it? Can you store it? Can you preserve it? Can you publish it? Can you create an impact from it? No actually we want to integrate ourselves into I'm about to think of a research project can you work with us to develop it? I think that's the question stage that we want to integrate ourselves in. It's not an easy thing to do but we pick two areas where we believe as a library we have course trends where we can integrate that and one of them is digital health and the other one is digital humanities and those are the two key areas we are going to integrate ourselves more. Developer culture of experimentation at Lancaster we have a strong philosophy of generate ideas fast, try them fast and succeed and fail on them fast. We don't compare how many times we succeed and how many times we fail and people who want to know whether we failed yes we have failed multiple times and I can give you concrete examples of how many times we failed on what but every time we've learned from it and we've not let go we actually have developed something out of it and I think that's the core strength in anything else. The areas where we are working on is natural language conversations so one of the technologies we are looking at is whether we can train Amazon Echo to help in different parts of the library about simple questions so can a student just go to a library call and say what are the library opening hours for today instead of coming all the way to the A floor to the information points or say Alexa can you renew my books from me my card number is this and that's all done for you why you actually waste or allow them why not actually make human interaction so much more simpler rather than actually having to log on to a PC then actually typing and going to the interface typing the username and then renewing it and then some error coming back or something else so let's make it really easy visualizations, text and data mining link open data and deep and machine learning are three areas that we are quite keen on so what are we actually going to do with this smartly data discovery if we can mine all this information if they are all NC licenses not ND what can we really do out of it what are the potential possibilities for it data visualizations and I'll show you an example of that why I think that's a skill that the library should be investing in now in my opinion personalization I've already talked about I think that would be really crucial because information overload would be massive mixed reality so again going back to the topic about augmentation of physical and virtual one of the ideas that came from our innovation group was actually when people are browsing the physical shells of the library they don't really know what ebooks are available for the same items wouldn't it be nice if they have a tablet in their hands they just put it in between two books and says here are the ebooks related to these books that we have why not that sounds like a good idea why not let's do something about it experiment on it semantic matrix so when we can text mine when this open access is in all the kind of metrics we work with are no longer going to be the standard with the metrics there are possibilities that could come from that about different kind of metrics and there's some work being done recently on semantic metrics that I think is quite interesting and also measure yourselves so again for far too long we say we are good at search library yes but how do you prove it so we are actually measuring ourselves and adopt industry practices again look at NPS scores look at CSAC scores we really like to say we got customer service excellent roles do we measure that in any way do we say that we are measuring how good our services so we introduce CSAC scores back in early February and every time a researcher is very salt they get a survey and we might be like survey every time that's bad we ask them only two questions good and satisfied bad and not satisfied and they literally have to do one click and our response rate on that has been 52% which knowing academics is a very very good response rate about out of those 52% another 60% have actually given us comments which we were not expecting at all and the unfortunate thing or possibly fortunate thing is we've got 100% satisfaction score at the moment so we can't improve anymore but at least we know at least we know now right so moving on we also want to brace the gap between libraries and IT and this is specifically on the third side of things and talk a bit more about that so analytics infrastructure so data to intelligence generally is quite interesting in our opinion and there are four layers to that the very first and very key layer is analytics we need to grab the data in the first place the next one is the context and that's about the location data the interaction data for that person how many times that person has used the VLE how many times that person has moved in the library what physical locations they've traversed through in the library were they talking to their friends at that time what's their favorite study space yes sounds like a big brotherish but actually we are more concerned about privacy of our users than they are if they can see a meaningful use out of it they will give away their privacy and we did conduct a short survey to discuss that and 89% of students said that's fine we're happy to give you this data if this ends up in useful information based on that analytics and context we then drive behaviors and based on those behaviors we need to make business intelligible decisions I think that's the ultimate goal that we want to reach and just to give a flavor of what kind of analytics infrastructure we are investing in so in the library our analytics infrastructure is possibly bigger than most small universities in some capacity so we've got about 60 virtual machines 150 virtual CPUs 400 GB of memory and 10 terabytes of space you might think it's just 10 terabytes this is not our research data research data is 300 terabytes replicatable going up to 2 terabytes soon this is just library analytics data that we are going to calculate and based on and we also use Amazon services a lot as well now we also use technologies that are cutting edge so we are using Google technologies a lot we are using Kubernetes which allows us to do a lot of this automatically other digital initiatives we are working on so study space occupancy those boots are extremely popular at Lancaster University as well and again I shame this promotion we also went through a refurbishment and we believe it was really good so if anyone is interested in coming along and seeing all those fascinating very busy boots please do come along but one of the issues that come with that is as soon as we open we are literally full so how do we then get or avoid that kind of atmosphere of space occupancy or hoarding and I think we are now looking at technology to do that the context data so we are running a research project with English about how can we evaluate how people physically move through the library so it's associated with their mobile phone device and lots of little weekends that we put across the library and it just literally pins them all the time so how long are they in a particular environment are they sitting next to their collections are they sitting next to their friends how long do they take to move from a particular part of the building to another what are their habits and that's the behaviour later we want to get more from that mixed reality I gave you an example of augmentation of physical and digital but you are looking at other ways of doing that as well natural language touch point so one of the innovation ideas that came from our students was can we not have some staff desk on each floor so we don't have to come all the way to A floor financially not possible but why not think creatively why not put some natural language touch points why not just put an echo and say you can ask about 70% of your queries there and if it's more difficult then come to the information point gesture control so we have these nice little display screens that revolve around about 10 or 12 different slides but if your student is about to catch a bus and the bus timetable is just moved away you have to stand in front of it for about 2 or 3 minutes to see that slide coming back again why not just embed some depth sensors and say move move move I am interested in the other one and that's another area we are looking at now move away spaces so we are already looking at phase 2 and phase 3 and it goes back to the well being point can we not actually say a student comes into us individual study can and say I want to study and it changes the light frequency to fast blue light it changes the background music it changes the environment for them so they are actually more proactive there or if they want to come and relax it changes the light to slight red tint etc etc etc those are well defined research outcomes out there but we are not using that very effectively in our environments personalization I want to talk about it much because we have spoken a lot and unique and distinctive collection of digital humanities is an area where we are really working on at the moment so lots of people have mentioned triple IF the platforms transcription platforms etc etc so all of those are on our agenda as well right I will quickly move on because I really don't want you to not have your tea or coffee but new areas to invest in oops I have given away my secret already now right so this is a statistic from our data repository since yesterday so I have just found one dataset two dataset three the difference between our most downloaded dataset versus the second most popular dataset from 1.4 to 822 I won't ask you to guess now because you might have seen the other side but the reason behind that is actually not because the first dataset is amazingly described or it has extremely high quality metadata and comparison to dataset 2 no the reason behind that is the first dataset has a very nice visualization attached to it that many people can actually infer that data from while they are looking at it so but let's get out of that metadata questioning all the time let's look at other ways of doing something that explores that data more properly and this allows you to explore that to give credit where it's due it's a very interesting project by the way it's about predicting smart from operating system from your personality and individual differences so again encourage you to go have a look at that if you're interested right moving on reaching the gap of datasets there's a lot of challenge in front of us there's a lot of technology change how do we approach that there's no civil bullet here there's no golden answer here I think we need to adopt we need to try different approaches here but to give you some kind of assurance I conducted a survey and some of you might have seen it on this on it as well and asked 5 questions to most people and one of the questions was how confident are you that you have the skills that would be relevant to the needs of research lives in 2025 and the good story over there is a lot of people are either very confident or moderately confident very few people are worried some of them are about to retire so they don't care so I think that's okay but at least about 19% are not confident so I think there's a potential area where we can work with the second question actually has been better results which is how confident are you to be able to cope with technological changes that might come in 2025 and the results are very confident moderately confident only 11% not confident and very few who are worried so again there's a lot of potential here working with staff who are already in the libraries who are willing to learn new technologies as well and actually that takes me to the third question which is how willing are you or how quickly can you learn new technologies and answer over here is fantastic because actually the key thing to look at is know here very reasonable answer because you don't have to learn all the technologies it depends on what kind of technology you're working with how relevant it is to you at all so yes and sometimes it's actually very positive here I have a personal opinion on this as well and my personal opinion is that you need to tackle it in four layers and those layers are initially digital leadership again I said that to you before I talk about it a lot invest in it please confidence building constraints but I actually believe in confidence more than capability when you are in meetings please don't ask people in fact let go of your critic hat that's what I would say start encouraging people to come up with ideas and again I think it was very good saying that don't underestimate the power of a suggestions box because those are really important things we have this tendency of having meetings where some people are naturally dreamers some people are naturally realists some people are naturally critics and we almost need to curb that criticism before that idea has flourished to a stage where realism starts working all coming up we also need to start investing in digital skills and court clubs but we need to do it in a micro learning style I think personally speaking go on other days when you can send staff to a very long week long workshop or even a two days long workshop or even a day long workshop I think what you need to do is take 20 minutes 10 minutes sessions and build on that continuously micro learning would be key for the future and it's equally valid for our students in fact recently I was reading a paper where they were saying that the reason mobile was not so successful for education was because of the way delivery was done it was all online it was through these 3G 4G channels at that time and it was long and I think the concepts are changing a lot in the mobile world as well right so that was everything that I needed to say I will now again show you an industry statement and this is from who is the managing director of Accenture he said there is no turning back from digital education digital demand is real and H institutions must start making the grid or risk its relevance in the digital era now I wouldn't go as far as that personally but I think there is a strong element of truth in there and from my personal viewpoint the question is not of an F it's of when so when do we want to start embarking on this is it after a new diversity has happened or is it before that I think that's the question first upon and upon so thank you very much for listening as I've mentioned please feel free to ask me questions now later or tweet them that's fine so happy to take questions thank you very much