 Mae hyn yn unig phobl i'ch bod yn gyflaen i'r ddeunydd fel y gwbl Cymru. Beth ddych yn dda, ydych yn yn y tynnu'n mynd i fynd y ysgolion umgarwch gyda'r cymniaeth a'r cyflaen arall, felly y gallw i. Mae'r credu. Mae'r pa i, ond mae'r gyflaen o'r collegiau yma yn gyflaenol saysu. Mae yna mewn ddweud eich mynd i fantasticydd cill quartersol, a'i cefnodd i'r argyffeb. Mae eich colleg sydd yn cefnodd. ac mae'r cyd-dynion i'r ffordd yw yn iawn i'r projeg yn shagwrs ar gyfer ymlaen a'r ffordd yw'r blaen ymlaen oherwydd mae'n rhaid i gael ei bod yn gwneud hynny yn ei rhan o'r cyfrifu ac mae'n rhaid i'n rhaid i gael eich bod yn y rhaid i'n rhaid i'n rhaid i'ch gweld i'n rhaid i gael eich bwysig a'r bwysig yw'n rhaid i'r hunain i'r cyfrifu ei wneud yma'n gweithio Dw i'n ddwybod i'r fitnod o'r argyffredig iawn o sydd yma, iddo i'n dweud i'r parerau. Mae'n amser, ond yn ysgolon i wedi gwneud 15 óot, y Farchiologiol archipeligod. Y dyn nhw, fe yna'r ardu cyfan gwyllus ymweld i ddur y gyflogant a'r archiologiol. Felly nid yn ymdweud dod ddweudio arwad a'r data. Pryddiwn i'r ddeill o'r methu a'i fyrraedd o unrhyw o'r enghraith o'r methu, yn cael bod y rhaid i'r ystafell ..y'r cyfliadau yn edrych ar y mychwilr, y tufaloedd cywlaeth tua... ..y'r cyfliadau cyfalwyr ymlaeniau? Mae usual ymlaeniau. mae'r ddweud er mwyn i roi'n seffaith mentir.. ..maen nhw'n sgwr fânwys.. ..y gwybod o gweld, mae'n ysgryfa'r prydu mwy. Mae'ntillogio'n bywyd yn dyma'r gwych... ..a wgwysig o'r casfurau'n yr ydy o'r mewn. Dwi ddweud yng Nghymru... ..y'r ymlaeniau'n gallu gweld... mae'r ddweud yn bwysig, mae'n meddwl yng Nghaerffordd ac mae'n meddwl i ysgrifennol a'r ddweud yn ysgrifennol. Mae'n meddwl i'n meddwl i'r gweithio'r gwaith, yn gweithio'r gwaith yn y gwasanaethol, ond mae'n meddwl iawn yn ei gweithio'r gwaith yn ddweud y gweithio'r gwaith a'i gŵr yn gyffredigol. A'r cyflwyno maen nhw, ar y gweithio'r cyflwyno, yn gallu y sectionol, those are the sorts of concepts that those that standard was covering. When we moved, I moved further, I did some more work and luckily got joined up with these chats in force in Crete, Martin Durer and many others since working on the CDOC CRM and okay they have more complex ways of doing this as you can do, I'll even just reading the things, the concepts that they're talking about and the identifiers but you know at the simplest level those conceptual entities are in that model too and that's kind of semantics you know in a nutshell I think. So moving on to that so the sort of work that then became the basis was some of my AIHRC funded work with colleagues and then was moved forward in in sort of the earlier work of Ariadne which is now Ariadne Plus and will continue to be used in that. This idea we start and other people have been talking more about this in terms of trying to search for concepts within some of our data now in particular the work we are doing was not just looking into the data in databases where you've got quite structured data but I just wanted to highlight this work which is more used in something called natural language processing where you're actually trying to pick out those entities in free text and that's I think sort of where it links in particularly with the nature of this particular session where we're talking about corpora and large text information spaces where we have to try and find at least pull out the sorts of information that we might be looking for. So this was just really just given a flavour of how that stuff used as I mentioned obviously CDoC CRM as a as a course of international standard used widely now based in museums but now we've developed more extensions for an archaeological extension for instance it's called CRM archaeo that's been used in Ariadne. People like the Getty Institute well they're there they used to develop for cabris which has also been turned into linked data which I'm going to talk about a little bit more. This was just actually I think that was just a highlight of that text so I should have put that earlier so you can see the sorts of things that it's starting to pick out concepts like time places objects how those linked together it's not just the simple sort of key matching it's actually more complex things like coins objects found below the pavement that date to a certain date and those are the sorts I think of questions and ideas that really as an archaeologist leave inside all this technical stuff that I've been doing for years that's actually the reason behind why I started doing this is to try and answer questions like where are you know where are examples of the coins that came from that particular period can I search across lots of resources to find them but to do that as Julie and I sort of thought I pointed out you need infrastructures and again this is where I think you know the importance of this without you know we will we I do want to talk about some of the philosophical epistemological issues behind so you need to look up when I say that behind this but we have to structure we have to try and provide some tools to enable I think people to want to discuss though who want to look for that information and pursue those sorts of questions so this is something that came on the back of work using the ontologies doing that and developing the idea that we could at least surface these particularly vocabries I call them heritage vocabries the kind of uh lists of words that the archaeologists and heritage professionals commonly use so these are just the examples that come up from from my work from a historic England perspective my colleague Peter's lurking at that from Historic Environment Scotland who are somewhere on there I hope that they're probably under their old name the Royal Commissioners Scotland but forgive me for that Peter um and he will be familiar with this one because this is work that he was very close involved in in terms of one of the crucial things about this I think from the you know European context is how we oh that's very doesn't do that on mine do you care yeah that was really quite exciting that broke everybody up okay uh the main point was to look at this idea of persistence which I think is again some of the some of the previous speakers were starting to look at this idea of what how we future proof in these ideas and to what extent by using the structures and the structures that we're providing in the infrastructure I've got to move on quick it's obviously given me a thing I'm taking too long the main point about this was behind us that we we incorporated other languages in this label it's the right one it just being a vocabulary that worked in English this enables us to sort of start going away um the main point being this sort of multi-lingual search which is obviously something that Julian and colleagues in Ariadne are going to use more and more but it's this potential it's not just us putting these things up to say right these you've got to use these it's more about this mapping process the potential of people to do the semantics to sort of say well that one looks like the sort of thing that it might and we have finds of that type but actually we call them something slightly different but but they are we are talking about the same sort of concept or much more likely as well that there will be concepts which are similar but because of cultural differences we don't want to mediate all the cultures into one sort of label we actually want to be able to map lots of different cultural references through through a kind of spine of potential searching tools so this this again this is just again tripped down memory lane for me but i think one of the challenges though is as we do this is for me and I love to surface this issue that actually understanding who is then using these things has become quite difficult personally I know that there are people like these people that do oceanographic research down in Antarctica who have used some of our terminalities just in their in their link data to match up things to do with types of boats and types of historic boats but there are people using it all around the world but the next challenge for us is to really kind of understand not just what archaeologists are doing with this but also I think how what other domains related domains like biology in terms of human remains environment science climatologist who I was in a session very interesting session before I came here or climatology work and how they want to use some of the historical data but they need it in a form that they can actually find it and then when and they kind of know that there that our data is when we're talking about certain things we're talking about the same things so they can use it finally I get to a new slide to Julian cheer Julian up this is just really again to show that that this is an ongoing process I've been doing this now longer than I'll confess to I think on tape but we are now finally hoping to plug we'll be plugging these in to this online resource this one Oasis has actually been running in the UK since at least 2004 in various bits before that but it's now a system that after 15 years it needs upgrading and as I think my friends and journey was saying about the you mentioned the point about don't type in the words by hand well I'm afraid that you know for 15 years people have been typing when it came to us because we had these we had these the sore eye but they would you know the web at the time we developed this the web the web facility that the actual online technology just wasn't good enough to actually have the whole of the for sore eye up online so we still had to do it but finally now the technology has moved on and we can just reference the link data so here's just this this is a sneak preview of this system which they will be out in March 2020 won't it Peter but again just to make the point that semantically this is looking at event types if you see we're off again in the in the wonderful world of bright purple or something or other it's looking at the excavation and obviously it's referencing the point because it's linking out to the heritage data terminologies again and we actually people just click on it and that links you in so in the time remaining what is it I would like to just think forward a bit because one of the lovely one of the issues again this is the the real point about the sort of semi controlled nature of people typing in and even in the field work I know for myself you know when you're recording something in a dirty muddy hole you have to say it's kind of ready brownish but for people doing the work with the computers that's a nightmare and as this chap said he found a reference to snuff because this stuff you know what snuff is probably but who knows what colour it was and the point being in 150 years or more time is anyone going to actually sort of I mean I doubt I've made that link data persistent and there is this fundamental thing that it's a persistent uri and it will persist but the technology I'm convinced will change in terms of how that is persisted in the same way that we have to plan to migrate our data our data archives our digital archives and one of the agencies core strategies is migration it's not just preserving it on a disk it's actually keeping that data alive and so I think that is is probably something that I want to think about from the from the point of view perhaps coming back to the first the first set of questions is to what extent is our data sharp fantastic this is the sort of thing that I am at and start thinking about now sort of projects that we're looking forward to and hopefully I'll be getting some money to do some of them in terms of sort of how we're going to preserve things like you know not just not just the link data which is something that I have a strategy but but all the other sorts of information which I'm trying to categorise in this very amorphous thing called heritage data and how how will we actually keep some of that alive and again that raises this question of the more the the internet is amalgamating the cultural differences what are we going to actually choose to keep alive again I think I'll skip through this in the time available raising questions around sort of data analytics as well which I've just wanted to surface in terms of because of those control vocabaries I know that there is some responsibility behind that that's the sort of stuff that people are going to start doing the analytics on and it is an important responsibility as well that we actually do think about how we want that data to be used and plan ahead responsibly now obviously you know we can we can do some of it in terms of creating the infrastructure and the systems to to to make those things possible but there are responsibilities in terms of actually making that work in a way that is suitable to different audiences and that's why I'll put that middle one up there again to some of the whole question of how certainly the linked the linked open data is going to work also has a responsibility to everybody who's maybe not so interested in how the technology works or anything else but it's more interesting in just answering some of those questions but if you want to answer the questions that you have you have to be I think prepared to share the data you have in a way that allows other people to question your data and I think that in some ways is still a big challenge for our sector and with that I will stop thank you