 examples of how people are using Madhuk and then hopefully kind of have a discussion at the end about next steps and how we can kind of build a community of Madhuk users and potential Madhuk users. So I'll firstly go through the announcements. I think there's too many people here and it's great to see such a large amount of people joining the call but hopefully you can announce yourself in the chat and also sign into the agenda, that'd be great. For the announcements the Naples Conference. Caitlin, do you want to go jump into this? Yeah, I'll try and run through these really quickly since we have a pretty packed agenda but registration is now open for our annual conference in Naples. If you are planning to attend we would encourage you not just to register soon but perhaps more importantly to book your hotel if you have not done so. Additionally, we have two travel scholarships this year for folks who do not have institutional funding to attend the conference. The information regarding the criteria and the application process is on the website. I will put a link in the notes document and then one last annual conference related update is that though we are sold out of gold sponsorships we still have sponsorships available at the silver and bronze levels and I'll include a link to that as well. Great, thanks Caitlin. Okay, on to the main part of the agenda so we're going to move over to Maddox and we're going to start with John Baker. Can I pass over to you? I think you're on mute. So can you hear me now? Yes, perfect. Yeah, thanks. I was going to say, yeah, thanks for the opportunity to raise a bit of awareness of Maddox on this community call. I really appreciate it. What we hope to do today is just to give everyone a flavour of of Maddox with the case studies that we have from Ghent National Library of Wales and State Library of Berlin. It's quite a powerful platform and so there's a lot that we aren't going to cover today even for people using Maddox. They're probably not all aware of all the potential uses so even people who are presenting today might find this might learn something from the other presenters sessions. So we're also interested in exploring as Glenn mentioned how we can build a community for Maddox beyond the initial uses and we have seen this cause a chance to test the waters of what what interest might be out there and we'll have a discussion towards the end of the session about that and how we might maintain momentum for Maddox through a larger community. So we have, originally Maddox was built by Djorati as part of three projects that we had on the go that had a lot of crossover. So there was the National Library of Wales credit sourcing platform called Torf which I'm sure Paul will talk about later. There's the Indigenous Digital Archive and then there's the Royal Society Science in the Making pilot. All of these projects were very diverse in terms of the use cases but we did see an opportunity to build a single platform that could drive these projects. Why did we need a new platform? The key reason really was that the the data that we wanted to capture through annotation was quite complex in terms of structure and diversity in terms of the data models. It's such a thing as such as linked data. And so one of the defining elements of the initial Maddox platform was this idea of what we're calling the capture model which enabled us to define and apply different models as part of the setting up of a credit sourcing or annotation project. The name Maddox actually comes from Welsh folklore. Interestingly the Welsh Prince who sailed to America and founded some Welsh Native American tribes so the story goes. So this was like a nice nod to the National Library of Wales and the Indigenous Digital Archive as founding partners. And the name has just stuck and we've kept with it and everyone seems to be quite happy with it so no plans to change it. So after we delivered those projects they were all delivered successfully by that initial version of Maddox but we ripped through quite a lot of functionality very quickly and decided after those projects to take a step back and rationalise the architecture and make it more maintainable, configurable and extensible as well as polishing up some elements of the user experience. So some of this work has also been done in collaboration with the Genshin University and the Claria consortium we're also presenting later. We did some work with them to create also to create some more configurable workflows to support quite a range of crowd sourcing and scholarly scenarios which Lisa and Davey will give you a taste of later on. The first version we used Amica S it kind of got us going quite quickly but it's been constraining us in a number of ways. So we abandoned using Amica S and the whole platform knows it's independently using a more modular approach to make further development more straightforward. This is the Maddox 2 release which is now facilitated projects for again National Library Wales, State Library of Berlin, Getty Research Institute have also been using it and soon IDA will migrate from the original version to Maddox 2 as well. So in terms of Maddox it's an open source platform. We've kind of been keeping it a little bit sort of quiet in terms of trying to encourage other people to contribute at this point because we wanted to make sure it was stable and get the second version architecture right. But we're kind of getting there now so we will be more open to invite external contributions. The way that we see the operating model for that is that digitality would provide sort of architectural governance and also facilitates take on engagement to drive the overall map to ensure that we have cohesion moving moving forwards. So yeah we're looking to create some, use the work groups in the spring which Paul will talk about shortly towards the end of the session when we discuss the community building. So that's really all I was going to say really just to get things kicked off. I don't know anyone has any questions or whether we just go straight into the use cases. There should be plenty of time towards the towards the end of the session for questions so maybe we could we could sort of get going and I'm happy to take questions at the end. So just quickly check the chat. All right okay that seems okay then so I'll pass on to Lisa and Davie who are going to talk through some of the Claria use cases. Just having a chat with Lisa about and apparently they're having a few technical problems. I wonder if we could switch to Andrei to do first and then go back to Gens if that's possible. Okay I'll be the next or that would be great if you wouldn't mind. Yeah hi my name is Andrei Buchman I'm from State Library of Berlin and I prepared a short presentation maybe it's better just to start and that you will see some some images parallel to my speech and it will be just easier to just a moment I'll start it. Yeah do you see it? Yes the full screen yeah or not. We can see the presentation for you. Okay I think it doesn't matter it's fine. Yeah just we started to work with Madok around three years ago and as John already said we started with the old one it was the Madok based on Omega S and for our goals it was not enough powerful because normally we have big collections and the Omega S was not very productive to work with the with the Omega and we just noticed that Madok started to to change their system and after some changes we got the new Russian version 2 and after version 2 it was possible to work with bigger collections and now we use the last one version 2 0 8 waiting for a new one and some some data about our collections at that moment how it works normally we have old collections written or old collections and they're placed in different old applications and we need to move it from this old application that never can maintain to a new to a new world and parallel we want not to display only images but to make it possible to annotate data and so on that's why we just migrate our collections from typical database structure to trip life and then just try to find someone who can help us because in this case we try to find a group of people and we found no one and that's why we found Madok Degirati in this case and we are working with Degirati which collections we have at that moment there are three collections one of them the biggest one has 13,000 manifests and more than half a million images you see all these links they are all available and you can just take a look and try to navigate and to search the information there and take a look how it looks like their manifests and so on there was some differences or some difficulties better to say with the first two collections the difficult was or the problem was that we have it in two languages lau and anglish and thai and anglish and all you know both of these languages lau and thai are not supported with a postgres database and that's why sometimes the search are not so fast like and in the supported postgres languages but it's stable it works it's possible to find everything in in those languages too then i want to show some some images some screenshots how it looks like for example with a search one image i want to display with Degirati we edit it's an extra feature we edit the footer the header and the footer we have our private footer and header i'll show how it looks like for example we have footer we have a header like in this like in our family like in another our cross asia pages and footer too maybe it's possible with the existing Russian it's possible just to integrate extra extra buttons like here for example i just integrate an extra dynamic button and you can just go from Degirati from Madoka platform to another one and maybe i prepared a list of a list of improvement list but i think we can just talk about it later in our discussion and what i i was not very sure is the a place where we can discuss it here again some some our improvements list where we want to see in the next version i think paul's going to cover some some of the current plans so you can bring it up towards the end if you like you can have a little discussion about it that that should be fine no maybe later i just stop on it and that's a short presentation of and we're working forward and new collections will come soon in our madoka repository right thank you any questions for andrei or would you be willing to go next i think we're still having some technical problems over uh i think davie's still having a little trouble with his setup i'm here oh you're here right yeah that was stressy beautiful timing yeah so i'm i'm not stefi i'm davie um i'm i'm using a different computer so quickly can i check um so i guess all presentations have been done um we've just done um sbb you're welcome to go next if you would like if you're ready i am actually quickly gonna check um you can do you can go last if you want a bit more time should be okay i think and you see my screen did you get to understand project okay super okay so hi everyone um so i'm davie from from gen cdh um but ironic that my computer didn't do its duty but um i just briefly want to present three projects um the first one so from gen cdh we've been setting or are setting up a couple of projects now the first one is the detergent project which also was a co-under of the pneumatic development so it's actually our pilot project now this project revolves around um witness interrogations so historical documents 17 18 19th century so what we've done in this project is we we have three collections um all uh with different localities or connected to different localities in belgium with these witness interrogation documents and for each of these collections we've set up um different projects so uh this is where you can see our projects um so a couple of have been done this is one that is still ongoing um so this from androp as you can see um it's still in progress uh not a lot have been started but um what i wanted to show you is if we have a look at the collection um a way that we actually or what is really a big asset of madoc is that we've used the imported metadata um from the collection to actually let users decide on what they want to contribute so it's it's a simple transcription project but say for example i only want to have we have a couple of metadata so the language the number of pages for example so say i only want um the documents in dutch and say i'm i have a lot of energy and i want something with 15 pages um i can select and then this way um i can find the documents um and so if i go to this one as you can see it's a it's a simple transcription project but um i can start transcribing um so it's pretty straightforward now another asset i want to show you um is so i use them the facet search but actually say for example i'll have a look at the bruce collection and if i do a search again and i do a keyword search and i'll just look for the town where i grew up um i'm actually searching in all the transcribed documents so this all these documents in this collection containing the locality of fields um i can identify so it's it's also a way of um exposing the collection to people who might be interested um so this is the first project the heterogeneous project um now we have other projects and so the second project i want to briefly present is heis von allen um this is in the context of the gantt maps or gantt enriched project which is it which is actually a collaboration between um eight ganttian uh glams so um the first project is connected to heis von allen now heis von allen is the museum for everyday life so they have um everyday life in the 20th century and they use a particular part of their collection which is i really find a very cool project is they use everyday objects from their collection for um memory work as to say you can as for example a social worker or a nurse you can for example use objects from their collection go to a retirement home and for example work with people with dementia and use these objects as a way to instigate reminiscences memories with these people now i'm gonna have a quick look at the at the um collection so as you can see we imported this collection in madoc as well as you can see it's like it's children's toys it's um harler games it can be clocks coffee grinders so every everyday objects now this collection currently they only have an excel with limited metadata so fairly obvious they want to um enrich this collection with more metadata to be able for social workers for example to go and look for objects more targeted so we've or are setting up it's it's set up but we have to launch the project but what we actually did is um made this little capture model um and it's fairly simple so we ask volunteers if they can say what is the team so we made a little drop down um for this also we want to have a more specific date so it's uh decades here so we made a drop down again and then a fourth uh no third a field is just to give a better description and then the last field is an alternative title so it's just best basic metadata um as a way to disclose this uh collection now in a second phase which would be a second project in maddock we also want to use this collection and add memories um to this collection um so actually enrich the collection even further so this was the second project and i hope i'm not using up too much time and i guess i'm i'm quite okay um the last project it's also connected to the um to the gantt enriched so this is from the amsep amsep is here it's the gantt institute for social history um now this is one project was setting up and it's connected to the gantt somerton which was a newspaper and we found or amsep found a nice photo collection mainly 1980s so this collection it's um i think it's over 200 photographs and it's always two sides so you have again a location in gantt um and on the back of this photograph you have some something could could be written it could be a location or the photographer or whatever but so quite unstructured so um with amsep we've set up this project in maddock as well um where we want to again um allow to collect some basic metadata and then feedback back to their uh content management uh system and their own collection system so as you can see none of the um pictures have been started yet because we're still setting up of course the project um but again let me take this one and let you let me quickly demonstrate um what we've done now it's actually quite cool so a basic metadata we want to harvest is actually the location but we've actually um added our own gantt gazetteer so it's actually an external vocabulary with the gantt gazetteer so it's basically like a a dictionary with unique identifiers for each and every place connected or in gantt um so as you can see it's quite a long list you can of course also search in the list so um we connected this gantt gazetteer to maddock um and then we've had or we have another a couple of other fields say for example the location is not in the gantt gazetteer we can ask for another location um then we have uh or we ask volunteers to add a date maybe if some people are um visible on the photograph to identify them make a transcription uh for example of the text written on the back possibly if there's a photograph or photo studio um written on the back so this is again it's a quite straightforward metadata project but um I think it's uh yeah it's a nice use of of maddock for this this collection we found rather small collection but um nice anyway so these were briefly the three projects that I wanted to present to you so I now give um the word or the floor to my colleague Lisa who has been working with maddock in an educational context so with uh students and researchers so Lisa the floor is yours yes if you could stop sharing your screen maybe david and I can share okay great okay so um a project in an educational setting that I wanted to present as we just finished this it's a building upon an educational project that we did last year it's the votes for women project which was a British feminist periodical and so we used maddock actually as a sort of co-creative way a co-creative platform for students to collaboratively design a crowdsourcing platform and really ingest their own ideas into it and how they would create a crowdsourcing platform so the students started off with limited digital literacy as it's not really uh integrated within our uni currently and so I gave them a two hour maddock workshop and um yeah that actually went went quite well the students which were six in total so not a lot you were able to quickly use maddock and create a project and be really creative as you see on this project page and maybe if it's even too creative um but yeah they did the design to um projects ingested all AAAF manifests and created a capture model to identify poetry or some other information which they thought was useful such as a title other or form so the goal was to design a crowdsourcing project but it was more so an educational use of maddock and then maybe if I'll quickly show the back ends that's where they um yeah basically collaborated together imported to manifests created collections and created projects so I think it was really positive that uh that's the students were able to work with maddock quite quickly after two hour workshop uh and do all this stuff as admin themselves so it was also a bit of a user test in in that sense um and so we also found out you could really set up complex capture models using maddock uh that's another thing we got from from these workshops and uh yeah that's something that does really well complex capture models so the next project was a scholarly project so we've at again Center for Digital Humanities we've actually had quite a lot of researchers who want to create a simple scholarly edition using some of the AAAF benefits such as deep zoom high quality and then basically annotated and we've tried to use maddock as a sort of way to simply um create a scholarly edition edition and publish it online and I think there can be a lot of potential for maddock to to become a sort of digital scholarly publication platform um but currently uh we're still lacking some interactions of maddock on um basically defining the annotations and defining some better interactivity first steps have already been set with the annotation styles but um yeah I think it's mostly just gathering the annotations that's going really well at this point but um it's displaying the annotations within maddock itself in an interactive way so for example for this project it would be um it would be the yeah it would be annotations and translations and transcriptions of uh Japanese text um so yeah that's still something it could improve on um we've noted and it's also takes quite a lot of time for us to set up these enrichment projects and to monitor it uh and we've currently rather limits a bit of of at times of resources to support it all um and then finally I wanted to uh quickly say when doing a workshop on maddock at the european conference so with a lot of cultural heritage institutions presence uh they found that annotation participation AAAF metadata and crowdsourcing they deemed that as the most important features of maddock so AAAF was quite up high there as as they really saw that as one of the most positive aspects of maddock even though the workshop was not really about AAAF so I think there's a lot of they saw a lot of potential in the collaborative aspect of maddock and how it could improve the amount of metadata and that's my dog but yeah we currently are a bit limited to support it okay I've tried to see it all quite quickly I'll stop now thank you Lisa that was that was really good um is there any questions for for Lisa I think um you know the fact that maddock is built from the ground up on AAAF is a is another key differentiator which I could have mentioned in the introduction actually uh so yeah I'll pass on to uh to Paul um if you're if you're ready Paul yeah ready to go just can you uh see my screen yes we can yeah see my notes as well or just uh we can see the source you can yeah that better yeah yeah that's it brilliant right so um I'm going to run you through a brief overview of the type of projects we've been doing with maddock here at the National Library the past couple of years so back in 2015 we'd been working on a number of different crowdsourcing projects to enhance our collections all of which were very successful but also required a lot of resources and none of them worked together or were linked together all the data that was created was and still does to this day live in silos. The projects on this page are the whales at war which was a linked data project looking at school communities adding data about soldiers from their local war memorials. Kenevin was a geotagging and rectification project of tithe maps and currently 1900 was a project to create a gazetteer that washed place names from the 1900s so it was part of our continued uptake and development of IIF and as a way to access our digitized collections we created a project to transcribe their Welsh Book of Remembrance using a version of Mirador we had modified slightly in house I say we glended most of the work very unfairness this allowed users to draw a box over a name and fill in pre-selected fields such as name rank etc related to each soldier once complete we had a surgical list of every single soldier in the book this form part of the whale sent for peace remembrance activities whilst this worked great for this project we used it again for a university admissions project for their centenary but we realized it was very specific to this specific need and wasn't a robust solution it also wouldn't allow us to do all the things we had done in previous projects we needed something more flexible that we could throw trip life manifests into and create different projects with different outputs so at the end of 2016 we were able to secure funding from the Welsh Government we put out a public procurement for a bilingual crowdsourcing platform for Wales did you want the contract and I believe that's why mad at was born since the initial platform was created from the brief we have been through three iterations of the mad at platform the original which I think was version 0.5 version 1 and now what they call version 2 which is actually only going fully live with us this week over the years we have run a number of different projects on the platform this one was our original and the first project we ever launched it was specced in the brief as one of the main outputs we wanted to ask people to transcribe the information from the Caudiganshire war tribunals this was completed relatively quickly and was one of our most popular projects it was visual it was quick to do and it was interesting to a lot of people once the project was finished we were able to use a viewer and our nlw website to allow users to not only view the records for the first time but also to search them the names, addresses, dates and occupations of the people within it feedback on the war tribunals project identified some issues such as not being able to search across manifests the fact we have to embed the outputs and existing web page and the reliance at the time of having to export all the annotations and convert them to store in a separate place one of the hopes we have for the latest version of the platform is that we are hoping to be able to go from a crowdsourcing project to a presentable data set be much quicker, smoother and integrated this is the William Livingstone Evans photographic collection and probably it's the longest running project we have done it started on the first platform and it will probably be live on what we call version 3 we're looking at implementing more of his images in the near future it was an opportunity for us to get information about people and memories of communities in north wales we were really lucky with this project to get a couple of volunteers who lived in the area and remembered a lot of the people and events and they have almost single-handedly identified everyone in these photos currently the data from this project is not available to view sadly and it's just sitting on the platform at the moment on our live site which is running mad at one what we call version 2 we have three live projects ongoing and quite a few hidden or older projects would have similar types these three sort of outline the three types of projects we've been predominantly running over the last couple of years the photo books are a collection of albums with very little metadata we are using wiki data identifiers in both Welsh and English through the platform using the wiki data API to outline and tag parts of the images that should make discovery user in the future and as the collections are so varied it may allow us to start a database of generic marketing images for example if we want an image of a dog we can search and find the crop image of a dog from within the photo books quite quickly originally we started off using the Library of Congress subject headings but found that people struggled to find the correct tags for example if you want to tag a car you need to use automobile and that also covered other vehicles whilst wiki data tags are varied they are at least consistent so as part of our ongoing work with wiki data we moved to tag in most of our projects with wiki data tags we are also looking at other projects using the maddox platform with wiki data for tagging up wiki commons paintings the second project is a continuation of the good in limestone ovens photos from blind of stineog and the third is a basically text only transcription project transcribed in the diaries of Wales best known artist Cuffin Williams all of these are near in completion of three quite different types of data collection models however looking forward these are both screenshots from the latest version of maddox running on our live server but on a beta domain both of these projects we are hoping to launch to a closed group as soon as possible should have been last week or probably be next week the first one is the chain meteorological records which will be transcribed this is our first foray into scientific data and to be honest the data is too complex we made a mistake picking this to be our test first project on the new platform it's taken quite a few months to get the changes we need with different workflows and terminology and getting this project just right has been a bit of a journey mainly due to the fact the content varies so much from page to page and having to work around that and trying to fit it into a content model has been a learning curve second project is our first partnership project this will host collections from three archives in south wales which have little to no metadata the aim of this project is to enhance their collections metadata through transcription and wiki data tagging the outputs will then be fed automatically back into their catalogs to enable better discovery for their own users one side effect of this project is the introduction of AAAF to record offices these and hopefully wider collections in the future looking further ahead we are excited to look at the possibilities of using maddox for research projects with individual researchers we're hoping to use the sites feature of the project to start displaying outputs from both these projects and previous projects and we are about to embark on a project to develop a central annotation store to link our crowdsourcing projects with our other AAAF annotation projects we currently have projects planned with the natural history museum for an academic research project and next year we're going to launch our biggest project today which is to transcribe the women's peace petition which currently resides in the smithsonia institute but will soon be making its way back to wales this will involve geotagging and transcription as expected to have a lot of interest both in wireless and in the states lastly we're also looking at training an AI Welsh language of voice to text transcription model on BBC audio content due to the complexities of the content and the AI models used and we're not yet sure if we're going to be using malloc for this but we're open to that's a very quick seven minutes 54 seconds overview of what we're doing at the library and I'll stop showing my screen if I can thank you thanks Paul um any any questions for Paul in the chat okay quite an impressive range and depth of activity going on in terms of crowdsourcing in this library of wales um hopefully it comes across you can everyone's getting a feel for the breadth of applications from from from from these examples um now we'll move on Paul do you would you like to take the range and talk a little bit about where we're at in terms of current developments yep sure um I will just share my screen did everybody see my screen okay yep uh excellent okay um all right so my name is uh Paul Mollahan um not Paul McCann but uh we share a few of the the letters so it's always confusing when talking to Paul but yeah before I kind of dive in obviously uh my role uh with Maddox over the past 12 months or so has been uh trying to help with the product management of the platform um and uh yeah just get the grips with all of the different use cases so it's really good to see um those examples from across the work that Andrew is doing at SBB and Lisa and Davy and obviously Paul at National Library of Wales and then as you say John hopefully it's given everybody a bit of more of a sense of what Maddox can provide um there is more information obviously available and I'll share some links at the end of the the chat so in terms of what I'm going to cover it's just going to be really brief but um hopefully give you a sense of what we're doing at the moment in terms of in-progress work um some updates that are coming um and then finally really just to talk a little bit about some ideas that we have around the next steps to try and build a amount of community so to get some feedback from yourselves and ideas and input um that's what we're trying to hope to do uh today okay so in terms of where we are right now well the team have been working on a number of different strands of work that will form the next essentially the next release of of Maddox so I think obviously as Paul mentioned there's been a series of iterations and obviously some some users are on the latest versions which are 2.0 and others are on different ones but um what we've been doing as part of the upcoming 2.1 release which is almost ready to go um is to introduce a series of uh project improvements that relate to some of the work that uh Leece and Davie have talked about from again perspective also some of the work that's going to support Paul's new projects going live at Nationalite we have Wales and also some work that we're doing for the Indigenous Digital Archive which I'll talk about uh in a moment we've also been addressing some technical improvements um which uh you know address technical debt but also looking to try and modernize and improve some of the shared components that Maddox itself uses and to try and refine the implementation of the platform as we as we progress so some of these things include a new API to support the capture models which obviously people have been discussing um and I guess the uh the main thing is that that improves how that whole capture model process is managed in the back end and it simplifies future enhancements and so on um we've also recently introduced Canvas panel which is um another open source viewer component that we use to render resources so again it's IIIF native um and um I think the the key thing with this is obviously Maddox integrates with and uses other open source components and libraries and and in that way it's benefiting from some of the work that we're doing and the community are doing more widely in terms of new features and functionality that that are available to support um some of the different aspects that Maddox itself provides there's also been a series of um other new improvements like adding in new page blocks to allow people to customize the sites and the the pages that they're developing as part of those sites there's new theme options and so on um some of the uh I guess the larger areas of visible UI and functional enhancements that are coming uh include um a new version of the reviewer dashboard so this is to support uh those reviewers either you know in a crowdsourcing project or in a research project um so we've been working uh with and building upon an initial iteration of this reviewer dashboard that we completed last year with the team at Ghent and we've been improving that in a number of ways so looking at how the data that surface for the reviewer users is presented and giving them some more tools to allow them to find and navigate their way through their work more easily we've also been doing a lot of work on improving the reviewer area itself so providing more consistency in terms of how information is presented and providing a new and improved UI to support their work and and also as part of that giving them a kind of a new focused uh view uh to really you know narrow in on and focus in on some of the detailed reviews are required for maybe for transcription projects and so on um what else so we've also um been working on uh some export functionality now Madoc has APIs to do an awful lot of different things um and supporting getting data in and out is something that that's been able to provide but in this case both the National Library of Wales and Ghent had some similar requirements to support uh or to allow them to kind of gather data from the platform to allow them to import it into other systems for reporting and so on and by combining their separate requirements we've been able to build a fairly comprehensive solution that allows data to be selected and exported in different forms so we've got a new UI to support all of that and that will yeah that that will hopefully provide an easier way from an administration point of view to actually get grab this data and make that available for others to use okay uh so that's what we're working on right now and that's part of this 2.1 release which we're we're testing and we're almost ready to push the button on in terms of the the roadmap and what's coming next well I mentioned the Indigenous Digital Archive earlier and obviously John you mentioned the fact that it's a founding project it's currently we're working towards bringing it into the new version of Madoc but a big part of that project will be enhancing Madoc further to bring more discovery elements and making those available within Madoc so we saw earlier some of the different ways in which searching and faceting works within projects but this enhancement will bring in the ability to develop thematic groups or topics through either through manual tagging or machine enrichment and then enable people to to search and find and navigate through that through those collections using those tags. Our team I think some of whom are on the call including Heather and Steven and Alex have been working through the current user journeys across the platform and this is partly in response to issues that have emerged from user testing but also known problems that we need to address so there's a series of things looking at improving just the hierarchy of content and labeling the on-screen information to basically to better support and orient users on the platform. There are some other things I just added one or two little things in here I know obviously we've been talking to the guys again about potentially introducing polygonal selectors to Madoc to enable selection of different shapes and canvases to support upcoming some upcoming project requirements but there's also a series of technical improvements to address some known issues and also some feedback I mean I know Andrea's been sending us some issues that he's had and I know obviously there's a few things that you want to discuss as well on that but yes so the the final bit really was just a highlight obviously you know we've got all of these roadmap items that we are aware of but really what what we're trying to do at this point I think is is to really start to kind of look at developing a wider community of users for Madoc and to try and achieve that what what we're looking to do really or what we're thinking of doing is running a couple of user group workshops later in the spring maybe in April and May which will allow us initially to share more information on Madoc and hopefully allow people that are interested to get a better understanding of the platform's capabilities showcase obviously further how it's been used now obviously we saw some great examples of that today but it would also then allow us to obviously identify some common gaps perhaps or improvements that might actually help support more use of Madoc and once we've kind of established all of that we would then like to have a harder workshop where we iterate on what we've got in terms of that current vision and roadmap for Madoc and and to help identify shared priorities on the roadmap that we could then prioritize and also we would be looking to think about you know different ways that we could identify funding opportunities and that's really where we'd like to get some input you know we're really keen to get some feedback and suggestions about that process and approach and whether that makes sense and obviously if people have got other ideas we'd really love to hear those so yes in terms of next steps if you are interested in in learning more and would be you know our interest in being part of that workshop or getting opportunity to see more of that please do get in touch my email address is there in terms of some useful links there's some links to Madoc documentation there obviously Madoc's in GitHub so all of the the projects are there and available the there's also we have a kind of a marketing website which is which is about really just kind of giving you an overview in more detail of what the platform can do so you can visit that and we also have a Madoc Slack channel which to be honest hasn't been used all of that much but it's certainly there and something that potentially that we might look to use in the future and that was it that was always going to say John excellent thank you thank you Paul so I think does anybody have any questions or comments for Paul maybe anything in terms of the roadmap or the the current developments or even things that aren't on the roadmap that you'd like to highlight at this point now's the time to do it if anyone has any comments I've got a quick question about geotagging georectification mapping that side of the stuff what's the what's the future where are we going with that Paul do you want to comment initially on that or even Davey might have an opinion well it's a good question Paul we I mean it is something that we have done a reasonable amount of work on brevity just looking at some doing some discovery on what the options are we spent a bit of time with the team again and with a number of other projects as well looking at possibilities around geotagging now in terms of next steps on that we are looking at possibly having some discussions with some other tools that are already out there that kind of provide some of that functionality but certainly in the short term we think that maybe the first area is obviously looking at the the improvement to introduce the polygonal selectors as a kind of as a first step perhaps in that but it's also will it give us a little bit of time to kind of investigate the different options that we could take with that further I mean I guess we know that there are some project requirements that certainly again have I have got for for that area so there are definitely things that they would like to do at some point but how we achieve that we're still trying to work out what the best route forward for that might be jazz I have a small comment or someone else go for it I want just to to say that there is many different interesting projects really that I didn't know before if it's possible just to put it somewhere on a madock page or in a trip live community that for everyone who are interested in madock can just go and scroll through these projects and just just to see a real examples for example we are still not using annotations we are still in the displaying of images but where of course interesting of annotations and all the things and for us for example it's very interesting to see how our colleagues from other libraries already done this this work and yeah that's it yeah that's a good suggestion you know I was thinking that we would do that maybe to follow up to provide a set of links of the projects that have been demonstrated to today maybe and all the other projects as well that would be a good idea we could add those and distribute a just on the google doc or something so everyone can see them it's a starting point and then we have we have some case studies out there on the on the on the digital arty website but we don't have an extensive number really and so yeah there's more we could do on that front how does everybody feel about the idea of a of a work group would would there be a lot of interest in in such an idea so that we could sort of bring everyone together establish a baseline understanding so everyone understands exactly what the what the platform does now and what we intend it to do at this point and then we could sort of do some planning off the back of that just does that sound like a good sensible approach yeah great we've got a few positives there brilliant well what we'll do is if you can if you're interested in being part of that that initial work group so that you learn more about the platform then please email Paul and then we can we can get in touch and with dates etc and hopefully we'll do another work group after that just trying to understand more about what you'd want to see in the platform in the future as well but I think getting everyone on the same page would be the first step that's what we what we think anyway Chris it is we built quite a configurable and powerful platform but that does hide sometimes some of its potential under under a bush under a bushel so we'll have to sort that out sooner rather than later um a few more minutes I see there's a lot of activity in the uh in the chat now just ask a question just while you're reading through the chat and so is is Maddox generally a hosted solution or do people install themselves or are both options available is it kind of up to what people want to do both options so we we people can just take it and host it themselves and we can if you want if required we can provide some support to on on their own hosting environment or we can host it for people as well if they'd rather not get get the hands dirty on the technical side so yeah those both of those options are available yeah if anyone was interested in it in just trying it just let us know as well if drop a top of line and we can get some trial versions set up for people to have a have a play okay um what else have we got and there's some discussion about technical components that Maddox is built on um I don't know could somebody else pick this one up or the mat or or Stephen I answered the question in in chat so I had asked if what the technical components were I mean the front end is basically a react application with a kind of no middleware layer that I can jump into if I'm getting that wrong but then there's a number of kind of um data management that happens in a kind of python jangle rest framework layer but they share a postgres database across all the services so things like search and some of the kind of enrichment like ocr or natural language processing are done in the kind of jangle rest framework apps but the core application is all JavaScript right I have like not maybe it's not maybe a question but in my ideal world in the workshop it would also we would be able to sort of put all the projects together and maybe learn from each other because we as a dh center we provide digital support across the university but we're not actually builds to supports crowdsourcing and that's so that's why we're collaborating with a lot of cultural heritage institutions but what you said Paul on the project of the national library of whales I'd be really interested in just maybe hearing more about yeah your experiences working with with volunteers and them interacting with maddock and yet the indeed what also resonated a lot was that sometimes it can take quite a bit of time to really configure the the right capture model we've also had that due to the flexibility of maddock we found that it does take a lot of time and to set up stuff especially if you want to do more complex things so yeah I think we can find commonalities in such a workshop under from each other maybe yeah that sounds great I had a common sort of points of friction in using maddock that that would be really useful for us to help you know to sort of polish the the UX experience further for sure so what maybe what we'll do then is we'll float a sort of a rough agenda what we think that workshop should cover and get people's feedback on that to see what we can what we can do and put some structure behind it so yeah we'll take that on as a as an action to pull that together and see get some feedback if that sounds okay great well we've run out of time I don't want to keep it unless anyone has any other further questions I just want to say a huge thank you to all the presenters and John and everyone for pulling this together it's been a really interesting presentation to see kind of all the developments and so I think John you mentioned yeah who should who should we email is it Paul M and it's can you put the email address in there google doc yes the other point yeah yes great it's really encouraging to see such interest a great turnout so I appreciate everyone's time so thank you very much everybody and look forward to hearing more talking to some of you more about what you might want to do with it next great thank you thank you take care cheers thank you bye