 You have had a little bit of eye-opening available yesterday in Caroline's introduction and in that sense I have to say I'm sorry I'm going to disappoint you because what do you take to state everything? As Caroline seems to have implied what I might do so I'm one of the reasons that would take forever and you'd want to get to lunch at some point. What I want to do instead really is opening up a conversation. We are sort of in some ways about to change the way how we are sort of supporting research and how research services run and operate at a wider rate. I would very much like to do this working method-wide a sector partly because I think there's some clear problems and building on some of the discussions we've had today and also yesterday I sort of send an increased interest to think are there ways how we can maybe work together better to be more efficient and in that sense I want to give you a bit of an overview of what we are trying to do thinking about not to say that this is all set in stone but rather say this is what we are working on maybe you can sort of bring to the table what you're doing and then we can see if there are some things that we can maybe work together in new ways. As far as libraries concern some of our thinking sort of goes back to 2015 when we launched the living knowledge strategy for the British Library I'm not planning to explain that in much detail I just want to sort of pick up on one point that our Chief Executive made when he launched it and he was speaking about time of historic disruption seems quite timely this year and last year I think but he made it specifically in the context of global systems of information and public hedging and he sort of set up, there is a challenge here and he said that institutions like the British Library but also institutions that many of you work at should have a role in changing the changes so I thought I'd briefly speak about some of the disruptions that we experience and then sort of move on to some things that we are thinking about in terms of shaping one way, one disruption is that the way our researchers work is changing you can sort of tackle this from many different perspectives I've decided to pick up one number that comes from the survey it was started in 2014 by the Sophomore Sustainability Institute and they interviewed researchers from Russell Group institutions and the result was that second-largest hand researchers say they can't now really in a meaningful way do their research about software that was two years ago there's a lot of interesting data in there but it brings up a few I think quite relevant questions for libraries one is, if software is so important for research are we sure it's being preserved properly if you sort of needed to offer certain research the second thing then is are we actually capable of really looking after all the digital outputs that are created by software and manage them and make them accessible and usable and the third one if our research is increasing your work with software are we able to feed our content so that this is into that software to a way that it's easy to consume and fits into the researcher's workflow and the libraries have been doing quite a lot of work in that space but arguably it's not something that we've cracked across the board in terms of workflow the second question is not just Google but I think it is user expectation where these data probably all of our new researchers expect there will be a nice clean interface between some where that I can access from whatever device I'm using regardless of where I am I can type something in, press a button and some results fall out, I can click on it I get straight through to my result and I have something that I can get on the internet and it can work with our discovery and access services we're meeting these requirements at the moment there's a study that just generated a library that a few years ago called Researchers of Tomorrow if you look through what sources researchers use for discovery you find that Google and Google Scholar or Firewall is popular more than you get citation databases and the others and library catalog and cross-institution library have lots of somewhere here I think if you did the same now or a few years later if the trends are the same you might see this shifting further so we are not necessarily the place where discovery is starting and arguably in some areas we are not also not the place where access is happening here are two tweets from our researcher who effectively said the access solution I get from a certain large publisher is so crap I'd rather go to the legal source even though I could legally get access through my university now it's easy to play a publisher I mean, but only if we should but we should also think we're negotiating these seals we need our service providers and perhaps our access services are quite a smoothie but it's something that we also need to look at on our end a problem that will come up in this context is money I'm particularly thinking in terms of cost of subscriptions is going up budgets, not quite as much speaking from a library here in particular our subscription budget is going down we can't afford to buy as much as we used to so the idea of just buying everything just in case a few researchers would need it because of all this particular problem which we think about value for money because of the budget library buying fairly broadly we have the issue that lots of resources that we buy get relatively little use per individual article compared to what universities do so the cost of providing these items and the value for money I think isn't there so we have to make some significant changes in terms of procuring content for our users in some cases it means if we can't negotiate significantly better deals we'll have to stop subscribing to content which is a challenge in terms of serving our users there's another discussion that will partly help but it's also, I think, confusing in some ways let's see how open inside is generally we talk all day just about solution of your forward points here but one point I want to make in particular from the view of the National Library if increasingly content is published in open access our discovery services are maybe that valuable anymore we don't need to provide access services it's an our role in helping our users doing their research if it's no longer providing them the information in the traditional way I'm not going to plan to speak in all these points in detail but I'd like to raise a few general points to reflect our response to this I think partly is if controlling the content and access to the content is no longer the issue maybe we need to put more work into the infrastructure on which the content is provided and see to it that we get the content out to our users wherever they are and in whichever kind of form they want it so that in some ways means sometimes thinking about software or machines as our users or at least as our interface is getting into the users and also certainly from the point of view of National Library means we're thinking our collections because the collections we have here are huge but considering the amount of data that's out there we could argue there's been relatively small so if we want to support our users we can't just make the assumption we have these fantastic collections and people will keep coming and coming we need to sort of in some ways we think our approach to collection management to engage in excellent collections and make sure that we sort of surface valuable resources to our users that we currently don't quite do and that means we have to significantly increase our capacity for discovery because there's much more stuff out there giving something new for back to users is difficult and we also need to think very carefully about how we can provide that content and move more to pay out provision of content and some of the means of doing this is we are looking at replacing at least the review involved in some cases certainly replacing pretty much the whole core infrastructure of the library and we need much more of the shelf height systems and we also need to rethink how we develop services that can increasingly mean finding partners either elsewhere or in the industry where we could say there's an issue provided or something really useful we could partner and sort of almost say we as the library guarantee that the data here is looked after properly if that service provider isn't that good anymore in the future we might move it somewhere else but it would certainly help us to be more agile and flexible and be able to respond to some of the user demands the way we are tackling this is through making everything available it's one of five portfolios that we have sort of change management I don't want to talk about it in much detail but we have one strand that looks at making higher attention to digital in particular digitizing collections sound for example that we are using we are working hard to transform the building here and it will be an extension to the new services to develop we have a program to engage more with the public also out to the world and change the way how we work with public libraries partly through that work of public libraries we can bring services to the user community we are looking at renewing our facility and most inspire and effectively expanding our capacity to provide connection management services both for print but also digital on a much larger scale and then there's everything available that sort of from a research perspective in a way the interface for some of these developments where we are looking at improving effectively discovery access and use and also rethinking some of the ways how we provide infrastructures to give a few examples we are currently working on a long term roadmap for changes that we want to do and this is discovery the general idea in terms of time scales short is something that we are working on at the moment midterm is something that we are sort of planning and hope to do a lot of this in the future maybe 1, 2, 3 years and long term is more sort of 4 or 5 new horizons or perhaps looking towards 2023 so very practically for our discovery services we currently have usability study that will hopefully tell us a lot of how we can improve them and will make some changes to our users in parallel we are trying a few new approaches partly working with startups and we are exploring working with the open access box and say isn't there a better way of finding open access content and sort of facing this to our users and how can this be integrated into a library discovery solution because if we can't afford to subscribe to everything anymore we need to expose the open access content which by the way is a good thing anyway because generally it means we need to supply some of the users we are not starting to think about certainly reviewing our current whole discovery architecture and say you might want to replace it with something better in the next few years that's sort of seeing what's the best off the shelf solution that we can get but most of the interest is taking a long term view and saying what our approach is like machine learning and artificial intelligence should we just wait for providers and say this is what you librarians need or should we just say well let's do some research and see if we can sort of check the market indirectly in a certain way and we've started some plotting on what do we think the discovery solution should look like in a few years and how can we research and develop this and one way of helping us is we now have the Arturing Institute of the House it's the National Data Science Center they're separate institutions but host at the British Library and we are starting to explore some collaborations and we've got a very strong staff on the same level for what he said he'd like us at the NTLI to work together and develop something which he calls data-driven library that came in all sorts of things admittedly but it might be a good way to get funding for some of the ideas and try new approaches around areas like discovery and I think this would be a part that would be really interesting to work with universities partly because you have researchers with expertise in this space and you will also have I think interest in seeing something we can do to explore new discovery services the other area that we're focusing on now is access and use and partly means developing infrastructure that at least for the British Library is new I realize university libraries have been doing for a while which is switching from just buying everything just in case someone wants it to having just in time provision and also expanding our capacity to give remote access and that will mean we have to have a much better understanding of different user groups and be able to tailor specific content partly to meet the requirements of rights holders who might say we allow you to give remote access to this resource for that type of user but certainly not for the other type of user so in some ways this is the area where I'm doing the catching up of the universities we're already having but it's something we sort of need to do as a first starting point we're also looking at in terms of how we can open up our collections a project that starting now is setting up an API platform for the library, the ideas that we have a sort of centrally supported space where we can expose all our APIs to the world to give people access into different collections the first APIs that we're going to set up on that platform will be the APIs that we force into the service but we're also looking for more directly on the human side by realising we have lots of digital heritage collections and we are sort of exposing them in different ways that maybe are confusing so we're looking at bringing this all together and saying he has all the digital stuff that we hold of our heritage collections and he has an easy way for users to get access we're also looking at sort of improving services on the side of particular at the moment we spoke briefly about the digital type reading room yesterday we are looking at actually having a space where we can have very flexible audio and visual setup that will effectively allow people to interact with all the collections that we have that we can't expose online for legal reasons but if someone wanted to come to us with a server plug the server in and do some text and data mining, we need space and particularly skills to be able to support us and hopefully some of these services you might then be able to sort of roll out your building on what the digital library maps have been doing and we're generally also looking at sort of changing our infrastructure for access and preservation key part of this is replacing the library's digital library system it was developed in-house roughly 10 years ago it was set up at a time as a preservation system because there wasn't to be anything on the market so it was developed in-house we currently hold about 750 terabytes that's replicated across source sites and we're now in the process to take really forward replacements that will be significantly more scalable which allows us to bring more content faster people have provision for both preservation and access and it will be multi-tenancy enabled we do this in the first instance because we need to enhance our capacity because we get more content in them than we did 10 years ago we need to be able to process that faster but it also gives us an opportunity to make some of these services available to others if there's an interest and believe it and one area where we are now starting to have discussion is in the repository services area the British Library doesn't have an institutional repository but you could argue as we're doing certain models of research we should have some function of an institutional repository www.bl.uk which is currently set up on a central repository platform here we're also hosting ethos for the sector that's run separately the idea is that we think let's bring this all together onto the central infrastructure so we can run it better in a more efficient way and while exploring this we've now realized that's actually quite a lot of interesting information so we are thinking about getting a repository or changing our repository for a shared service model and linking into this one area that I'm particularly concerned about in the open access space we're not going to get all this open access content coming into our repositories which is great but I'll be confident that we're preserving this but I think this is something where if there's an interest I could zero for the British Library we've also lodged a data strategy that Essence says that we want to integrate data into our services as well as we've currently been doing this with textual collections the different components that we're looking at partly for our own needs data management, data curation data archiving and preservation and particular data discovery access we reuse this is something again we need internally but it's also something where we can build on that capacity and see are there so this is something we can provide that others might want to work with us on in terms of the way we're serving in searches there's a key part where we serve where we search on all sides of our life but like in already offering some services to institutions and also international international level data site is more example or UKRR so we're kind of thinking there might be different roles that we have to play and in some ways we can probably help individual users that we can't get to directly by providing services to others such as data site or via data site or actually interface with users where in all cases of putting us all together into a roadmap I don't want to talk these things through in detail because it's undevelopable but you'll see lots of things I've been talking about about improving our access and discovery services that are already things that we're working on and there are some of the more speculative or we'll try different areas here like improving our capacity for on-site data-driven research and there are some things that we're planning to do long term for example new types of discovery systems or see if we can not really significantly enhance our capacity to enable types of data mining research in our collections but I mostly still see this at the point we're starting to think about these things and really get your feedback and what would be useful to put on the roadmap and perhaps take forward together and with that I'm sort of coming to an end and trying to give a slight different definition of everything available and the only way how we can really give researchers almost everything they want to need if they're working together because these challenges are way too big for one single institution arguably in some cases we can't even do this on a national level it has to be international I don't want to say much more about this because I think we've heard a lot about these types of arguments in John's keynote yesterday so I'll effectively say just think about everything he said yesterday because we've left this across the slide and then you get some of the main points I want to get across but I thought maybe useful to add on to concrete things one thing I haven't really spoken about is skills we need to enhance our capacity and capability in the library one thing I'm quite keen for this new research services department that we are forming is to have work placements and channelings so we can send our staff out to you to learn about how you work and what challenges are and there might be similar ways how we can sort of organize a more regular kind of work placement and exchange system or some other activities that we can be able to help start to expand our skills and very concrete terms as I've mentioned we're speaking to some partners about ways how we can think about repository infrastructure in a more systematic way it's partly links into the UK working group and I mentioned earlier that I've also mentioned that we are thinking of, if there's interest, maybe setting up a workshop sometime in spring where we bring those of you who would like to have a discussion how can we make repositories work better together and do some of these options to sort of end on to concrete examples of what we're thinking of doing at the moment thank you