 So today's session is the second in a series of RLUK events exploring AAAF use in cultural heritage. Our first event gave an introduction to the framework from myself and from Glenn Robson who is AAAF's technical coordinator, and if you missed that it is available to watch via the RLUK website. Our event today will feature two presentations on using AAAF for digital content. AAAF stands for the International Image Interoperability framework and the framework is a set of open standards for delivering high quality attributed digital objects online at scale. It's also an international community that develops and implements the AAAF APIs. Our first presentation is from Scott Bradley and Valentina Flex from Newcastle University. Their talk will focus on their newly relaunched AAAF enabled website of the archive of Gertrude Bell, explorer, writer, archaeologist and colonial diplomat. The website contains a comprehensive and unique resource of newly digitised images of Bell's photographs, letters and diary entries presented alongside transcriptions and other contextual information. Scott Bradley is a library system developer at Newcastle University and works closely with special collections colleagues to promote and facilitate the widest possible access to their rare, unique and distinctive archives and book collections via AAAF. Valentina Flex is a project archivist at Newcastle University's special collections. Over the past two years Valentina has been involved in the relaunch of the Gertrude Bell website which is now AAAF enabled. Her current project Beyond Margin's will focus on the creation of a geospatial temporal interface which allows users to explore archival items within geographical and historical context. Our second presentation is from Joseph Padfield from the National Portrait Gallery. Joe will be exploring the use of AAAF on national scale, how it can facilitate digital infrastructure, enable researchers to reuse digital content and showcase dynamic applications of AAAF in documentation, examination and reinterpretation activities. Joe Padfield is the principal scientist at the National Gallery in London with expertise in preventative conservation, digital imaging and the organisation and sustainability of heritage science data. So if we can go to the next slide I just have some housekeeping before we start. We are going to have time for some questions at the end of the session but please feel free to post questions at any time during the event using the Q&A function at the bottom of the screen. The chat function is also open so please say hello and let us know where you are joining from. If you are on X then we are using the hashtag RLUKDSF. This event is being recorded and it will be made available on the RLUK website. And next slide please. I think we are ready to start so a big thank you to Scott, Valentina and Joe for giving up their time to talk to us today and thank you to you all for joining us and we are going to get started with our first presentation now so it is over to Scott and Valentina. I am just going to start by giving you a little bit of context about Gertrude Bell and the archive and then I will hand over to Scott to talk about the technical parts and then I will come back in at the end just to let you know what we are up to now. So Gertrude Bell for those who are not familiar, she was born in Washington in County Durham so she is quite local to the Newcastle area in 1868. She was an archaeologist, a traveller and a writer. She travelled extensively particularly in the Middle East but also in North America and East Asia. She was later a member of British Intelligence in the First World War. She was heavily involved in the creation of the Kingdom of Iraq and she was also the founder of the Iraq Museum and she created the first antiquities law for the country and she died in Baghdad in 1926. So the archive contains over 12,000 unique records so that is photographs, diaries, letters, reports as well as like photographic negatives and photograph albums. It spans 52 years and it covers lots of geographical areas. As I mentioned she travelled quite a lot so Europe, East Asia, North America and the Middle East and the archive documents Bell's activities so her personal archaeological and political activities but it also portrays landscapes and people and intangible and tangible cultural heritage and as a result of that it was inscribed on UNESCO's Memory of the World Register in 2017. It is a really, really highly used collection and we do have a lot of different collections here in Newcastle University but it is one of the ones that is constantly requested. People are always interested in Gertrude Bell and her material. The archive is complemented by Bell's book collection as well and that's her personal working library and there's a lot of intertextuality between the archives and the books so there's like lots of nice annotations in the books and so it's a very, very rich resource that we have and it's also really important geopolitically as well as culturally and historically. So the Gertrude Bell and the Kingdom of Iraq 100 project, this project was kind of how the new iteration of the website happened. The project was funded by the Harry and Alice Stillman Family Foundation and it basically the main goals were digitization, cataloging, the creation of the new website and the creation of an exhibition and the main kind of focus was improving accessibility and preservation. As I've mentioned it's a really highly used archive all across the world so we really wanted to make it as accessible as possible so we are incredibly aware that it contains material and heritage that relate to lots of different people who can't always come to visit the archive. So during the project my colleague Graham digitized 13,000 items so letters, diaries, photographs and the photographs include the negatives, prints and the albums so these are all now available on the new website. We also transferred photographic negatives into specialist cold storage and there we created new cataloging, that's what I did, new cataloging with improved information and descriptions and as I said one of the main facets of this was then hosting all of this wonderful new resource, so the digitized images and the cataloging metadata on a lovely website and how do we make the most of this resource that we've created which is where Scott steps in. Thanks Valentina, yeah thanks Alice for the introduction as well but just quickly just yeah I'm Scott Bradley, I'm one of the library systems developers here at Newcastle University Library. I thought I'd start with just giving a bit of background about sort of our triple IF journey at Newcastle and I'd also like to apologise at this because it's a bit of a brain dump of PowerPoint slides so please do bear with me and I'm also conscious that my Wi-Fi isn't great and there are some videos in this so if it's a bit glitchy once again apologies. We started using triple IF around 2017 when we launched our then new digital asset digital asset management system by the tongue twister content dm, content dm generates all the required triple IF manifest for us and then we just sort of consume those into different applications and since then we've gone on to develop a lot of sort of different triple IF components that can be used across a host of different library applications from your you know your traditional websites to kiosks and nearly everything in between and if anyone wondered this background image here is a triple IF manifest from one of the bell photograph albums. So what we use generally we generally use a customised mirror door viewer for the applications we develop just because we find you know it has a lot of the features that we require sort of out of the box and it's relatively easy to customise. We actually use a stripped down mirror door viewer you know on some library web pages and our on our archival catalogue Atom as well and there it is in Atom that's a customised widget within the Atom interface that one of my colleagues developed. Again we use a customised mirror door viewer to display triple IF generated content on kiosks in our exhibition space within the library it we you know we find it allows for a much more immersive experience for exhibition visitors and we find it works really well when we have it alongside sort of printed materials it allows the visitors to see that you know the physical items in a secure lock display cabinet but then in a way they can flick through the item digitally and you know zooming in to see the finer detail of the album. This is our current exhibition Newcastle University at 60. It shows pictures over the years once again using mirror door and triple IF. Like I mentioned earlier one of the things we like about mirror door is the ability to customise the interface relatively quickly and effectively. You can see here for example we've gone with a black background and sort of a kiosk mode as they call it with a thumbnails along the bottom but with mirror door you can easily tweak that with the buttons at the top right and you can easily as well easily go into sort of a gallery layout as well for what I think is sort of easier navigation. We also work closely with the museum in Newcastle the Great North Museum. Last year the museum hosted one of our Bell exhibitions and once again using triple IF to display some of Bell's photograph albums. I don't think the picture on the left anyway does adjust this but this kiosk once again sat alongside one of our display cabinets and similar to what I said earlier it allowed the user to see a page from one of the albums but then using the kiosk they could click through the whole album. However we don't just keep our triple IF work to ourselves. Exit University currently have a couple of our kiosks with a Gertrude Bell exhibition while they've got a Gertrude Bell exhibition going on and once again it's displaying one of the Bell's photograph albums. There it is in a slightly more detail. We've also used triple IF for an application where branding is turning the pages. It allows access to some of our standout collections you know just with the click of a mouse and once again it allows for that rich immersive experience something you know that was we don't think was ever really possible in the past. Okay so yeah moving on to sort of the Gertrude Bell website in its overall the site was in massive need of an overall the old site sort of predated me starting at the university and that added up the other day I think I've been here 11 years now and it required a lot of updating from sort of the basic search to the way the image and the metadata was presented which you can sort of see here this being the old site search results page which I'm sure you know I'm sure we all agree looks you know very dated and not very inviting and easy to use at all. This was the and you know this is the resulting this is the new site search results page it employs a patchy solar as a search engine and it uses facets to help you find items you want quicker and when but when you know planning out the new website we knew we needed to incorporate the the triple IF because like Valentina touched on it allows for this that sort of enhanced accessibility. Triple IF allows website users to really interact with that material in the way and I keep saying it in a way that they haven't before and it's especially true for sort of the bells diaries and letters for example here is you know where Bell is scribble notes in the margins you can now really zoom in and see that deal and you know maybe discover something that we we might have missed. One of my favorites from the collections and this is a video so apologies if it is a bit glitchy at your end is where Bell's used tape to stick pressed flowers in triple IF really brings those details to life and zooming into these editions really makes that the the item a lot more you know richer and more immersive you can also see here on the right hand side there that we denote with the triple IF logo that we offer the user the ability to grab the triple IF manifest that we use to generate the metadata in the image and then they can use that metadata and image sort of you know anywhere else. Another great example of triple IF is this Japanese drawing of when one of Bell's diaries entries you can really see the detail I think in the gold and the red threads when when zoomed in there and even the even the pin and how that's sort of degraded over time. One last favorite of mine from is one of Bell's dance cards from the Royal Mail ship Parra on our way to the Caribbean date 1898 zooming in again and even if you go sort of full screen it really brings out the detail in that that printed card and the colors used and and because it sort of has you know faded slightly being able to zoom in sort of really brings those details to life. What triple IF also allows you to do is sort of resize triple IF generated images sort of on the fly so instead of having to either use something like you know Adobe Photoshop or even sort of CSS to achieve it you can pass image coordinates to a triple IF URL and in return it'll give you like a cropped image sort of which you can see here so that the original image being the one in the background that's blurred slightly and then the new image in the foreground and all that was required was to where you can see we've highlighted in yellow there which is to add some coordinates for the image and like I say it returns that that cropped image for us. We found this tool really or this feature of triple IF really useful so what one of my one of my colleagues again did was they created this small little this small little app sorry that you just need to enter the original image URL and then as you can see here you can drag this box over the image that you want and then in return you've decided exactly how you want it to work and then you copy the cropped URL and in return it will give you the image coordinates of the new URL and then from there you can you know you can paste that into a website or into the browser. Because of this sort of this triple IF feature 99% of the images on the bell's website or triple IF generated images meaning we only store very few locally reducing duplication ultimately saving on server space as well. I will however you know be fully transparent here just in case others are sort of starting out on this triple IF journey and you know maybe somewhere you know about a year in maybe behind us and some of the problems that we came across when we were when we were sort of implementing the new site the problems we came across weren't necessarily due to the well the word due to the triple IF framework it was more the tools that were being used to generate the triple IF. As I spoke about earlier our digital asset management system content DM serves that triple IF manifest which we then consume into the bell site and like I said in others however when we started to plan out the new site and the functionality required we realized that the triple IF that was generated didn't always meet our exact needs so for example what I mean by that is bell has a lot of diary entries which have been catalogued content DM is what content DM calls compound objects or sort of a multi-page document meaning a diary could be tens if not maybe hundreds of pages long however with the new bell site we might only want to show sort of a subset of those pages so you know for example pages 10 to 17 of a diary as that diary could have been written over sort of different geographical locations or over different time periods etc and you can see that right here with this 260 page compound object however with this example you only wanted to show pages sort of one of two and that picture there in the foreground is a screenshot from the actual bell site another requirement which was one hour wish list you know was to show item level metadata so you can see here from within mirador you've got your item level metadata at the top and then below that you've got your object meta object level metadata meaning when you sort of you're flicking through a multi-page canvas manifest you when we started out you would only see in the metadata for you weren't sorry you weren't seeing the metadata for the item you're on you were just seeing the general object level metadata because content DMD didn't allow for this at the time what we ended up doing after sort of speaking with both the developers of content DM as well as well as the triple IF community you know was something quite straightforward in a way something I wish we thought about earlier we call the content DMAPI so when the manifest is requested and before it sort of served to the user we make a call to the this content DMAPI passing I think it's the collection and canvas ID in return we get sort of item level metadata for that canvas so all we had to do then was rebuild the manifest including this new data one problem we did encounter with this was solution however you know sort of due to the number of API calls that we had to make especially for those items you know with maybe many canvases long so many pages long was longer than maybe average page loads times so we developed sort of a basic caching mechanism to store these cached items as basic text files on a local server and you know this caches cleared periodically and in case colleagues have made any updates to the metadata excuse me made any updates to the metadata we do sort of have plans to further develop this caching to a more sort of a digital object repository where we can bring in and store trip live objects alongside you know other digital objects like TEI however that's still very early days so it's something you know I might be able to cover at a maybe at a future event and I think that's all from me for now the volunteer will now sort of briefly cover the next phase phases of the bell project and sort of how we plan to further develop it including sort of TEI and and maybe interactive mapping thank you and so going forward and we're going to be embedding HDR so it's handwritten text recognition and TEI text encoded initiative and into the into the site this is part of a kind of ongoing project called evolving hands where we've explored both HDR and TEI using Bell's letters to create transcriptions and then encode the transcriptions with contextual information so eventually we hope to be able to incorporate this into the site alongside the the IIIF image and the current project which was mentioned at the beginning in our introduction beyond the margins is exploring geographical breadth of the archive and by using a geospatial interface so we have basically kind of two strands going on at the moment we've got the development of the interface which will sit on the website and will incorporate the metadata and the IIIF images that are already in the site as well as metadata work that really involves myself and some students working on tagging all of the items in the archive for the location so eventually we will add all that metadata into the interface and we'll see all the items from the archive as pins on a map so you can see this is like a little work in progress screenshot of our development site and this is pretty much what we're aiming for which is representing the item on the left hand side so using IIIF viewer being able to zoom in and zoom out as much as you can on a full view like Scott demonstrated whilst also incorporating this map view and we also hope to have a little kind of section where we can have exhibition type of material so stories that link to items on the map and show a particular journey or a particular theme within Bell's archive. Good afternoon yeah my name's Joe Padfields one of the scientists working in the National Gallery looking a lot of how we work with data I've been exploring options for sharing images on the web for quite a while now but we've been looking at IIIF for a good few years and I'm just going to try and summarize some of these issues in the next 20 or minutes and I apologize for it seems a little rushed but a lot of it is commented on in a report I will mention so the detail is available now IIIF as you probably already know fairly obvious is the international image interoperability framework as mentioned in the previous presentation and it's an international plan or initiative to design to allow images to be dynamically displayed together from multiple locations without downloading so the concept is store once use multiple times and this if you if you went to the previous meeting will be fairly straightforward to most of you now in practice what does that mean so we have as was mentioned also in the previous presentation this concept of a manifest which is almost a recipe which describes information about images about groups of images in a standardized format that can be understood by a number of different presentation viewers to allow you to explore the images so if we look at this snapshot of a manifest on the left for the six images that are on the right if you want to share these the URL for the manifest is 72 bytes the manifest itself is 9.7 kilobytes and the compressed high resolution images are 180 megabytes so it's a basically the concept of efficiency of sharing a lot of information and this is just for six images so if we're talking hundreds of thousands of images the efficiency becomes quite obvious now from our point of view if we're working in the National Gallery we have an image store including large numbers of triple life images and a triple life server so this is the software that takes especially prepared image and will give you small image tiles and zoomed images as required and we explore them using a variety of different Mirador based image viewers Mirador is mentioned also in the the previous presentation now if we take another institution the A.O. Centre for British Art they also have an image store they also have a very nice triple life image server and they have a public viewer so it's not an internal one where you could explore a lot of their conservation images now you can go in there and take that small text file or the link to that small text file and drop it into the internal image viewer within the National Gallery so if we're working in this case in Mirador 3 we have an example of a manifest describing images of a painting by Turner but also then we can drop in the manifest from the A.O. Centre for British Art and within the National Gallery research environment we can be comparing both the images at the same time so triple life obviously allows users to efficiently present and compare images from two different institutions at the same time while maintaining the connection and this is the important bit is that those images are still connected to the institutions they came from these institutions can see that they're being used and where they're being used people didn't need to download the images you're maintaining that connection this becomes even more important at scale say if one wanted to create a virtual national collection so without a standard image framework a potential national or even international collection would need to maintain a huge number of different custom metadata and image service connections so if you think on the image on the right there you've got all of these institutions that potentially a lone little researcher down the bottom marked in yellow and they're presenting either structured linked data and endpoints or static data dumps or potentially working through services now a national collection would have to work with manage and potentially host a lot of that this is a common problem with data aggregation systems and it also leads to a disconnect between the image source and the presentation as I mentioned before now triple F obviously solves that so if all of these institutions and that lone researcher is making use of triple IF a national collection would only need to aggregate the metadata or even just reference the metadata published by these different institutions or services now there are a few alternatives for researchers to publish triple IF content at the moment I believe there's been a lot of development in the internet archive in the last few years where people can host images but there are other solutions coming along so you don't necessarily need to be an institution to do this yourself now the other thing to think about this may not know as well there's something called the fair principles now across data management in general across multiple fields the notion of fair is used to try and help people guide them in making sure their information can be found and reused as a series of principles you can go and explore this more but the top level four are making sure that your information is findable accessible interoperable and reusable now triple IF very very nicely supports all four of these so it makes it a very robust standard for working with data and working with research environments and sustainable environments going forward so triple IF enables precise metadata URLs if they're persistent for people to find digital images they're obviously accessible it's an open standard so it's clearly obvious how you use them interoperable is in the very name of triple IF so allowing sort of cross collection and cross institutional integration and obviously reusable but sustainability can be an issue because obviously there is a lot of development of triple IF a lot development of how institutions use triple IF so the notion of the difference between triple IF as a presentation tool and as a research resource is an interesting issue to be discussed now between the years of 2020 and 2022 the National Gallery led a particular research project focused on the practical applications of triple IF this was funded as part of the Towards a National Collection Program funded by AHRC and our main aims were to showcase some of the benefits of triple IF investigate virtually connecting organizations together examine the potential and use of existing tools and then potentially look at what else we might need going forward now there's a lot of output from this project it went on slightly longer than originally intended due to COVID but we had quite a large amount of outputs the final report which you're all welcome to go and read I'm just checked today and it's almost been downloaded a thousand times now which is quite nice we held seven events contributed 16 others a number of surveys and all of our webinar and seminars were videoed and recorded and they're available through Zenodo or YouTube and the work also fostered a number of very interesting discussions about potential future work and we also provided a clear list of recommendations for the Towards a National Collection Program in general and the ongoing research funding from arts and humanities research council in this area so it was a very successful project but in addition to the networking and discussion what I want to sort of show you today is we put in a number of tools and demonstrators but the fun things you can do with triple IF and I was going to show you four of them to go through here now the first one was a simple triple IF discovery system now many institutions are starting to offer APIs so if you have time and you are of technical bent you could go and read the documentation and explore what they've got and search for content some of them include lists to manifests or triple IF enabled images so what this little project did was looked at how can we unify these across multiple different collections so each of these collections have a different API they have different ways to present triple IF content so what this does is it produced a secondary set of endpoints that all outputted exactly the same formatted data you give it a text word and it gives you back manifests or triple IF resources to present automatically within a triple IF related viewer in this case open c-dragon but mirror door was also used within this project so we connected together a large number of projects now occasionally I have to turn off some of these institutions due to updating of versions and stuff the Smithsonian it often needs to get turned off because you kind of swamps everything else because they have a huge number of resources there and the National Gallery is currently down at the moment because we are changing our endpoint but it allows you to do free text search and return to triple IF content across multiple collections it's really good fun another one so manifests were mentioned before and depending on how much you know about triple IF there is also the concept of collections so collections are groups of manifests or even groups of collections of collections and we at the National Gallery have been presenting a set of high resolution images to support exhibitions or publications so if you've ever read a journal in the past and it says as you can see from the detail and then you see this very small image that you can't really see and you're out with a magnified glass so zooming in on the pdf these days but what we were able to do is present the pdf and supporting visible images at the same resolution that the authors had access to this was done by a combination of tools pre triple IF image presentation tools and a number of databases and what we did was we tried to do the whole thing with triple IF and Mirador so effectively this whole system now presenting content from a range of different publications i think it's like 10 15 different publications or exhibitions on there now are all done based on cached triple IF manifests and collections so there's no database there the system all just works on the manifests and you can browse through have a sort of breadcrumb effect be able to go up and down the series of different publications explore read the pdfs and explore the images so it's a very nice way of doing it we're hoping to do more with this to make it more generic but at the moment it works really well with this particular triple IF collection another one was the idea of data repositories so we've talked and i think sort of has to be presented the idea of a given collection presenting their content to tell a story or to engage their audiences with their own collection but if we're starting to look at generating research data so things that support publication or want to be the the foundation of future research you want to create data sets that can be reused so Zenodo if you have not heard of it is the catch all data repository for the EU it's free to use you can upload any some data set you want and you're given a DOI for it and you can reference your work quite happily now that works on a piece of software called InvenuRDM and within the practical triple IF project we were able to work with one of their developers data futures to enhance the triple IF aspects of InvenuRDM so that we could stick triple IF-enabled previews directly within this data repository so for a given set of tiffs in this case within a Tudor portrait repository there was a series of conservation images of this particular painting and the preview allowed you to just zoom in and zoom out on this simplified viewer so that worked really really well and a lot of that technology has been integrated into the newer versions of InvenuRDM and Zenodo so that was quite good now the other thing to think of is that if you're an institution you say well this is my manifest this is my list of images but what happens if you want a different list what happens if you want to explore alternate lists users generally do not want to look at these manifests they're written in a format called JSON and you have to be of a particular mindset to want to go in and edit them by hand so we needed a more user-friendly approach so the project also supported the development of a web-based user-friendly tool to make manifests so you can pull in existing manifests cut and paste them together and create new presentations to tell new stories so it's all open source it's on GitHub it was produced by a company called Digerati and you can go and have a little look but this is sort of a rough idea of the screenshot they have where you can very it's very much a drag and drop you can see exactly what images are available and added metadata for them it's it's a very nice tool which we're hoping to use more in the future um the other thing to think about as well is that Shiplier doesn't have to be static so we often talk about the idea of I have a collection I've made my manifest these are my images please have a look at them but there are many examples of use where you want these manifests to be dynamic an easy example of that would be search results so if you wanted to encompass and save the results for certain search in its own manifest that could be reused these search results you could generate lots of them you can also have changeable groups of images or even sort of annotation based situations where you're adding to an existing manifest as an example instead of images of an object or a work by an artist think of an institution might want to present a manifest of a given room or a given location or a given building or even given geographical space thinking about what was discussed in the the previous one previous presentation but within institutions object move so if you ask for the national gallery group gallery 54 manifest that might be different one day to the next so there is a certain aspect of dynamic nature and potentially versioning that may need to be considered here now as a example of annotation we use triple f enabled sort of technology in this customized documentation system within the national gallery so in case you don't know when paintings are examined we often take very very small samples microscopic samples from the surface of the painting to study the materials used to create it and in this case we have a mirror door to version of version two of mirror door interaction with the database where people can zoom in annotate particular point right on the particular crack where the sample came from and then fill in a related form that will allow them to define what that sample is now the eagle eye of you may have noticed that some of the sample details are not tremendously accurate this particular painting I use as a test case so it's easier for me to display but this is a password protected administration system used by one department now if we wanted to share that information with other departments or potentially the outside world what we then do is repackage all of that information and present it within mirror door three for the rest of the institution so here if you do a search in our internal collection image viewer for that particular painting and g1234 you can get all the paintings all of the samples and then a specific image presenting the sample sites so again this can be read only presentation of the same information so not only using triple f in one place we're actually reusing the triple f information in two places to get more so you can see a very nice interaction between the samples and location and the information that we have the other things we start to use triple f for is other exploration of images so it's not just the presentation images but it's the study and examination and reinterpretation of these images and this is one example of which is written up in the technical bulletin I think number 41 where this particular painting they're exploring what pigments have been used and a particular analytical technique called x or f has been used to try and identify which elements are present in which paint passages and then the images can be registered and effectively Mirador 3 allows you this option to change the transparency of one over the top so this particular image here on the left is a blending between the visible image and the copper map to show you roughly where copper can be identified in the surface of the painting but then you can compare that obviously with Mirador with just the straight visible image or x-rays or infrareds as you need so this is something we use on a day-to-day basis as an actual study tool rather than just a presentation tool the other thing to consider is how we group images so you can just say well here are all the images but as was commented in the previous presentation you may have hundreds if not thousands of of images against a particular object and for certain applications you may want parts of that or all of it or subsets or different presentations so here is an example of all of the images coming back from a particular painting the Giovanni Bellini, a Dominican monk attributes of Saint Peter Martyr and here we've got visible images we've got x-rays images we've got infrared images we've got photo micrographs samples and some XRF going further down and different users within the institution may want to see different things so we are exploring the idea of having a different default manifest for say a curator than we might have for a scientist or a conservator or even the education department so we're playing with how we package this we're trying to look at this concept of trying not to hide everything but just trying to promote the things that are more readily useful for people and triple F allows you to do that which is very functional now just give me an example these these images of Peter Martyr is a wonderful little painting it was edited somewhat in the past to make it slightly more saleable and someone put a big knife in the person's head and a big sort of dagger or sword in their chest but these selection of images where you've got the original visible image you've got an x-ray and infrared and then some compositions that have been created to try and re-establish how the painting might have appeared in the originally before these additional attributes were added so again this is a study tool but it's sort of a presentation of a story of the painting and this set of registered images can then again be explored this way or presented in different structures so just to finish off triple F is an international standard for presenting and working with images and AV and soon 3D it provides a reliable efficient foundation on which other systems and tools can be developed whether that's internally within one lab or whether it's internationally across multiple institutions or large-scale research infrastructures and it can also be a lot of fun. Thank you very much. Thank you so much Joe that was fascinating video one you can see me yeah thank you to Scott and Valentina and Joe for all of their presentations there's so much to work through here we have had a few questions have come in I am going to start with one that's been addressed to Valentina because I know she has to step out a little bit early for another commitment I've got a question from Ruth who would like to hear more about your HDR and TEI plans for the project yeah so I cannot speak as like the only person that was involved because it was a couple of different people involved in different levels and they evolve in hands project but I can give you kind of an overview so the project was our first kind of exploration into HDR and TEI and it was like another member of interdepartmental I suppose you would say project with one of our colleagues in the School of English who kind of specializes in digital humanities and there are other institutions that are involved in other kind of parts of the project so they're doing basically the same thing but with different material so what we have done or we have what we did was we started by recruiting student placements and they used the digitized images of Bell's letters which we already have because of the digitization project and they used transcribes software and they scanned that through and thankfully we have transcriptions already of a lot of this material because people did it a long time ago thankfully saved us all that work so we used the transcriptions that we had we used the digitized images that we had fed them into transcribes and created a model to read Bell's handwriting then my colleague Alex did an amazing job helping the students with that and learning how to use this to teach them I was more like consultation for the TEI part when it comes to contextual information but I'll get to that so and that's what Alex and the students were doing in the first instance and then we used we selected kind of like a batch of letters they consulted me as to what ones might be most historically interesting and we scanned the digitized images of those ones so that transcribes would create its own transcription and then on top of that we used the TEI to add contextual information to particular things within that transcription so like names to explain who people are places or just kind of concepts and the chunk of letters that we picked was kind of around 1921 and around the 10 of creation of the Kingdom of Iraq and the crowning of King Faisal so there was a lot of things going on and a lot of concepts in these letters that if you didn't have historical context you would you wouldn't know what they were talking about and a big thing was like reducing that barrier to access the project was heavily kind of like focused on the methodology so how we did it and what problems we encountered so we could then report back about how that worked but as I was kind of like more on the content side of it I was very excited just about the fact that we could embed this information and directly into the the transcriptions and so that was kind of like how that worked and then now this phase of the project we're looking to embed it onto the site and so that we can present the IIIF image of the original item the transcription that was created by Transcribus and then kind of like layered on top of that would be the TEI and Scott would have to tell you a bit more about the technical side of it because that's not really my area. Yeah I think you've coded in great detail there, Valentina so yeah like Valentina mentioned it's still a proof of concept at the moment we're working with colleagues in the research software engineer and colleagues to create sort of a reusable widget for this TEI so yeah you would have the IIIF image on the left hand side and then the TI document on the right so you can sort of compare like for like and like I said still fairly early days I think the project is summer of this year if not the back end of the summer so hopefully by then we might have you know something a bit more you know to show anyway. And then we do have plans to do a similar project with other material which we kind of haven't really started on yet but that kind of like has been ruminated so as Scott said this is like a proof of concept and then we can go ahead and replicate this with something else. That sounds so exciting and even just the work that you've done I think to put a AAA for you were into Atom I think it's a very exciting development so yeah you guys are spearheading lots of exciting questions about that. Yes and a little technical questions. Which version of Mirador were you using? We used version 3 the reason why and it's a bit of a lazy answer this was because it integrates with content DM if I remember because we've been using it for a few years now if I remember correctly it was because it integrates better or it did anyway at the time with content DM like I said a bit of a lazy answer it does have I think if I remember correctly again sort of additional accessibility and additional features of obviously version 2 yeah that's the that's the the quick answer I think. And are you using presentation version 2 or 3 on your AAA FAPI? It's version 3. And what do you what's your feeling of the difference so you you happy with that choice? Yes yeah like I said in the in the presentation I think because we know I have used version 2 but not in anger necessarily yeah but version 3 as I said in the presentation allows for it it already offers so much of what we already wanted and it allows for that customization really easily yeah so yeah it does everything we want. And what's it been like working with content DM and the question was do you share code publicly so I'm wondering how much do you how much have you built on what content DM does out of the box and how much? Yeah so well not necessarily with content DM but I think to answer the question around I think it was about the widget within Atom yes once again I keep looking I feel like I'm a broken record here but that is also a slightly improved concept we're near in the end of that but once that is in a position to go live yes we're more than happy to share that whether that should be via or via you know via you know via colleagues yeah we very welcome news. I can't take credit for that there's a colleague you know a colleague on my team that that's working on that but but yes we're more than happy to share that. Brilliant I mean the other theme that's come up has been around copyright and image sharing so this obviously depend on your own approach at Newcastle and what your your kind of cultural appetite for sharing images is but this is this comes up a lot when people talk about AAAF and very sort of frightening this idea of just putting everything out there. What's what what is your approach at Newcastle and how do you feel about just people just being able to download everything? Well the thing is we have this kind of the Gertrude Bell archive is quite unique in a lot of ways for example the Evolving Hands project we were really lucky to already have transcriptions and to have the disguised images to use with that and equally our old site that Scott showed like very briefly in his presentation and there have been scans available online of these images for a long time so that website predates us and they've been on there since that website went up so that's quite wild and I know that a lot of people do take images and use them don't necessarily ask for credit and I know that because I've done a lot of reverse image searching and this work and Facebook has been incredibly helpful so I can't I can't like denounce anyone for using a lot of local history groups in different places around the world and use them usually we're like we're pretty open and we do like ask that people credit us and because the copyright such as there is does reside with us and but equally the thing is like people have been able to take them for all this time if they want to and there is an aspect of like if someone really wants to use something without seeking permission or or referring to us or whatever they will just do it of course and and I think yeah so the main thing was just just making them accessible which is what we have done and it's kind of like we've just sort of had to accept that that's okay. This is what AAAF does so beautifully though isn't it so it allows you to share images in a far more responsible way than has ever been possible in the past so you've got a viewer there where your images are packed up with their metadata and where they come from and somebody can share a link to something that they found on your website and it doesn't matter if they put it on their website it's it'll all be linked back to you eventually and it's not about controlling the images it's about so people would see an amazing document and think I want to see this archive what else is there right there and they can find you right so I think the benefit of having people contact us for permission like we always say yes like we have a stop answer we're like yes of course this is we ask that you credit us this is how we ask that you structured it whatever and like someone recently asked if they could use an image on instagram we were like yeah of course and thank you for asking how nice to ask yeah yeah I know I was like bless like bless them for like taking the time to ask and but the good thing about that is that it's nice for us to know where things are used just because it's nice to know that people are using them but equally you know yeah sorry I don't know if that's much of an answer that's right do you want to jump in Joe um I was just going to say that if you are able to share some of the code of what you're doing scott there's a very nice process that if you share them on github and then you make releases from github you can hook that up to zinodo so you get a doi for your releases all right effectively it's very very sort of easy and nice for people to build to then reference your work and to say where it's come from and which version of your work you're doing if you're developing it and it's just a nice way to maintain that sort of sightability or attribution really to be able to say who's done what um so have a look at that it's it's it's quite nice you you can set it up so automatically create your new doi every time you do a release it's sort of pretty streamlined really I think historically for no reason but we we haven't really pushed that you know a lot of the stuff that we've done out to get our similar repositories but it is something that we definitely want to do you know going forward we've like a colleague who's really starting last two years is all for it so yeah it is something that we're we're looking into yeah there's also some triple i f score a community in zinodo as well okay so effectively you can connect things that you do to other things on triple i f within zinodo it's uh that's happened in last few years I can't remember exactly about thank you I think there are a lot of people very interested in this atom integration um there was a question on conservation this is a collection that's largely kind of late 19th century did the age of the material make it easier to digitize that content considering that it was in maybe like better condition than older material or were the things that kind of created challenges for the digitization process too I would say take this answer with a pinch of socks I didn't digitize it so when I say don't think there were any issues my colleague yeah um was it more was it was it that helped I think that sometimes if something is like around that period it can be more difficult because it can be like like there are a lot of items that we have that are older that are in far better condition because of the material that they were made with I mean I'm not conservative the answer that I give as well yeah so like sometimes that is actually easier when it's older because it just lasts longer it's better paper is better quality yeah um so I don't think there were any issues there were things that my colleague Graham had to be careful with for example the inserts as we've seen that scott showed um I very carefully replaced the pins with entomology she used just like old kind of metal pins and they were rusting um but there were like fiddly things um the photograph albums which you can see on the website are really big they are kind of like um bound together by like almost like a kind of material canvas buckle to buckles but Graham actually unbuckled them and did it like page by page loosely so he was able to do that I don't think there was really any issues in digitizing but I do think that the big obviously I state in the obvious but the big thing about digitizing is that it's a way off in terms of preservation because photograph albums did have to like they are aren't taken out very often but very unwieldy they're very difficult to handle and they're all going away for conservation now but even even stabilized are quite difficult and and this just means that we can like reduce reduce the handling but if anything actually I'd say the book collection would be more problematic but that hasn't been digitized yet so that's what's maybe next on the horizon for you guys um talking right next on the horizon Joa was wondering what do you think next next for the towards a national collection outputs um you know from all that work that was carried out over the last few years you know where where next well this this um my understanding is that the towards a national collection uh the project I talked about was a foundation project and then there was a series of larger discovery projects that came out which are coming towards an end now and the program is then putting together recommendations for future funding so this could be fairly sizable uh sort of bands of funding that would come out of AHRC into the multiple millions of pounds but I don't know specifically sort of numbers whatever but a number of the recommendations that came out of that triple I project have made their way I believe into that final report um so the the recommendation that sort of triple F should be used more as a sort of a standard way of doing it within heritage rather than a an interesting thing if you've just happened to speak to the right person so I think working with sort of funding bodies to say all right you're doing a project with images have you considered or even you will need to uh I think could be a very nice way of looking to it there have been some discussions about trying to push triple F as a national standard um but no one's sort of found the team to do that yet um jumping through the administrative hoops um but that's something that has been discussed as well um the other big thing on the horizon there is um another big funding structure called riches which is building the research infrastructure for the UK and one of the funding outputs is to build a UK data repository and it may well be wonderful if a free UK data repository for heritage included triple IF of course so we'll have to sort of see how that's done and who's going to be putting that together but that could be a very nice opportunity if small institutions or researchers were able to then deposit their research outputs in a UK repository and then automatically have triple IF manifests and uh resources for it so I think that type of ability for everyone to do this rather instead of just the people with large supportive IT departments I think is where it will become so much more um ubiquitous basically