 OK. So, I'm Adam Rusbridge from Edina at the University of Edinburgh. So, in a slight change from the programme, I'm going to talk about entitlement information, which represents evidence of perpetual access. So, it's not as sexy as text and data mining, but a bit like keeping receipts in your kitchen drawer. It's important to manage the evidence that shows your permanent rights to access a digital copy. So, libraries refer to entitlement during renewals, and it's often difficult to understand what access is retained if subscriptions are cancelled. Under the big deal, institutions receive access to all-front file material, so publisher platforms become like databases of articles. But should the big deal prove unsustainable, an accurate entitlement record will show what was purchased and so what you can continue to access. It's been hard to keep a record of entitlement, a system to gear it around current access. It tends to be managed in ad hoc arrangements in spreadsheets, email threads and people's heads. We all want to provide the access that we paid for, and when access is cut off, a lot of effort is spent trying to pull together records to provide publishers with accurate information. So, we've built an entitlement registry to assist in the process, both to aggregate historic information and to capture new records as they're produced. There are four purposes here to discover publisher-supplied entitlement records, to upload library records and compare differences, to agree a shared view with publishers, and to build a golden copy record that both libraries and publishers can exploit in the future. We asked libraries how this would benefit them. They said to register what they own in perpetuity, to reduce time spent in agreeing renewals and alleviate pressure on the team, to run our department more efficiently, and to ensure our students and staff have access to everything that they should have. I've been enthused and surprised by publisher enthusiasm for this. Publishers strongly recognise the problem that libraries have here and they themselves find this a problem during journal transfers. Getting entitlement information from the transferring to the receiving publisher can be a messy process for them, and it can be unclear who has responsibility for providing access after a transfer has taken place. So, to dive into the registry, you create an account to not only see information about your institution, we collect publisher titles, start and end dates, and there's typical navigation like you'd expect. One of the main initial use cases will be to identify problems in data, so you've tried to identify the most common issues or flags that might occur. So, data from multiple publishers indicates a journal transfer. So, here Edinburgh has access to a Taylor and Francis title until 2015 that transferred to CUP from 2016 onwards. So, we can make it easy to clarify who's providing access for the years before the transfer. Is it still Taylor and Francis or is it now CUP? We can work this out once and share that with everyone. In the next example, Edinburgh believes it has perpetual access to an EUP title from 2010 to 2012. However, we've discovered that EUP no longer hosts the content and the entitlement record hasn't transferred to CUP. So, perpetual access has potentially been lost in this example. Where we have non-matching entitlements, that indicates discrepancies between two data sets. So, here the earliest dates from library and publisher supplied data, they just don't match up. The reasons for that might be straightforward, but if we can filter down to that very quickly, we can make this scenario transparent and easy to access so it can be quickly inspected and signed off. Gaps in entitlement become noteworthy. So, here CUP report no entitlement for the single year of 2007. That looks odd, is it correct? If it's a flaw with CUP data across all institutions, we can spot that quickly, correct it once and resolve that for everyone, saving individual libraries from duplicating time and effort. We can flag up titles where we only have data from a single source. So, we want to encourage a shared view between libraries and publishers. This comes from aligning two sources of data, or by one party signing off on the other's data and confirming that it's correct. So, this is the publisher's assertion about entitlement that deserves some review from the publisher. This all feeds into strategic initiatives, really. So, within individual libraries, the data can help make renewals more efficient. But at the UK level, we can conduct analysis across UK research libraries to support things like interlibrary loan or print rationalisation. So, we want to make it cost-effective to use the data operationally. We're running a pilot right now to understand running costs, access to data and to prioritise features. We're working with a small set of invited libraries right now. We've asked each to provide library-maintained information, help encourage publisher participation and help with user testing and feedback. We are having discussions with a range of publishers. These publishers are at a variety of different stages of participation. Some have now supplied us with data. Others are scheduling this into activity plans for the coming months. And we're planning on using UKSG to secure a participation of a much broader range of publishers. So, improving data quality is at the heart of this tool. We found that library data quality can vary and it's held in unstructured formats. Our experience is that producing library data at scale is quite expensive. Right now, we're focusing on automating data supply from publishers to populate the baseline database. We've found that this is likely to receive more buy-in from publishers. But what happens when a library and a publisher disagree about the extent of entitlement? So, we've modelled a dispute resolution process, but at this point we need to explore the data a little closer to understand workflows and practice. The first step that we're working on right now is to capture information about disagreements and then discuss how to resolve them between libraries and publishers. We're working with nine library institutions, but we're keen to co-design the registry with a wider audience. So, if you'd like to participate, please do get in touch. Our target is to have discussions with 30 top publishers by the end of July. There's a longer tale of smaller, messy publishers with few titles. So, we've had positive conversations with some subscription agents to get data about them at scale. This activity is all towards a goal of launching a service in the 1890 academic year. We believe that there's strong demand from the library community for this tool trying to understand this data is a real pain point for a lot of librarians right now. So, we're also in discussion with a variety of European countries who've all expressed interest in this. So, we're looking at deployment outside the UK but hoping to launch a service in the UK first. So, thanks for listening. I hope this is of interest to you and if you'd like to speak more, catch me in the break or send me an email. Thanks very much.