 Gwyl, hi Everyone, am ym yn Chris, rwy'n anwyn, rwy'n anwyn yn ei ddweud, ond ym yn ymwneud hynny'n gweithio'r plug oherwydd yng Nghymru yn anwyn yn ymgyrch yn ymgyrch o'r cyfrifio. Rwy'n ddweud yn y bwynt, ond yn ymgyrch ar y ddweud, mae'n gwybod yn ymgyrch yn ymgyrch yn ymgyrch. Rwy'n gweithio'n ei ddweud yn ymgyrch yn ymgyrch. Rwy'r syniad o'r idee ac mae'r gweithio ar y Manwyn a'r maeself, ond yw'r gweithio'r gweithio. Felly, i ddim yn ddigwyddol ac i ddim yn ddim yn ddigwyddol, ydym ni'n byw, mae'n byw, mae'n byw, Manwyn. Rwy'r gweithio ar y project yn Gwyl Inglade, mae'r gweithio ar y landscape enghraeg, mae'r identitiw sy'n gweithio'r llyfridd, ac mae'r pryd ynglyn wedi'r gyflwytaeth Cymru, ac mae'r gweithio ar 2011 yn 2016. Rwy'r ffiddio'r pwblifachau, ond nid i'n gwybod i'r bobl yn ymddangos o'r projekty nid yn nid, ond nid oedd mae'n gwybod i'n gwybod i'r bobl yn ymddangos, sy'n gwybod ei fod. Yn ymgyrch yw o'r unig o'r cyffredinol, ychydig cynnigon cyllid, yng Nghylch, ydw i'n masg o'r 900,000 record, eu bod yn ei gwiswch o gwybod o ymddangos, yng Nghymru ar y blefnodau brydwyr. Yn ymddangos, mae'n gwybod i'r bobl yn fwynt. Felly, mae'n oed o ran o cael eu bod no project yn fusillu ond rydyn nhw'n meddwl, wedi bwysig i'w archeologi i gyd, ac oherwydd y dylai chesaf o mileydigunau data sy'n gan congruiau ar y ffordd rym surtout. Felly, gallwn amod, mae'r pryd yn Ytod G16 a'r ysgol drwy Gymru, ac yn ddiwrnod cael ei gael ychydig sydd wedi eu lleidio. Cynol, i wneud â'r projewdau, argyfael â'r cyhoeddion faviwn ffordd wedi'u ganddeigiaeth gan ein gwyllgor i magnau rhaed i'r Legaidd yw yng Nghymru, ond mae erioed yn fawr yn ratedi. Felly y gallwn gweithio'r digwydd o'r cyffredin llwyddoedd yn ddau'r cyffredin, dyma hwnnw yn ymgynghwyl y cyn Barnasol yn ilisio'r cyffredin llwyddoedd a daethau ysgol o'r cyffredin llwyddoedd. fel ymdegwyd, ychydig, yw ddweud, ychydig, y ffordd, ac mae'r ffordd! Ond mae'n ffordd i'n ddweud y ddataeth yn hystodol i'r hordd, rydym wedi bod ar gyfer hynny ymddangos i'r hoffaeth yn ei bwysig, a'r hoffaeth i'r hoffaeth hynny yn ei bwysig. A'r hyffaeth yn ymddangos i hynny'n cael ei gwaelig i'r hoffaeth yma, gyda'r cyfwyr fânfodol sy'n lanlu cyfraith yw'r cyfwyr yw ddweud yw'n gweithio ar gyfer ddechrau'n ddod Ac roedd yn groesio ychydig iawn ddadech ydyn nhw'n ôl. Fy fydd y cwmaint rwy'n ddod y bydd yn hyn o bobl cyd-slid yn ei ddod yn gynghwm, byddechrau dan yw mynd i'r cynnwysau müssenol credits yw'r hynny. wrth cremonm are under threat in the current economic climate and their continued existence relies upon demonstration of their usefulness in that they are actually being used. So there is reluctance in some quarters we found in archaeology in England to use these secondary datasets due to concerns about their accuracy and reliability but its actually really important that we do use them otherwise they can have to die. The relationships between archaeological engineering entities and the digital records they become mae'r eich cyffredinol a'r cyffredinol eich cyffredinol. A'r lle i fynd ddod yw'r ysgolio arall yw'r newydd yn cyffredinol, a'r cyffredinol rydw i'n mynd i ddweud y bydd eich gweithio'r cyffredinol, ond yn y llif. Dyna'r ddigyddion cenderon o'r archaelogicol yw gwylwch yn ymlaen i'r bwysig yn rhan, ond mae'n gwneud o ymgylcheddau metaforol, mae'r ddim enw yn eu cymdeithasol yn y cwmtex. Felly, mae'r cyfnodd yn cael ei ddweud y ddechrau yn ffordd yn y cyd-dweud yn ddyn nhw'r cyfnodd yma ar y cyfnodd ar y dyfodol. Mae'r cyfnodd yn ystod yn gwneud y cerddurol ar gyfer ar gyfer archaeoligol o'r data, ond mae'n rhaid i'r ddweud hynny. Mae'r cerddurol ar y bobl yn gweithio ar y gwaith ar gyfer Baulfa ac Star. Mae'r cyfnodd yn gweithio'r cyfnodd yn y cwntes ar gyfer y gweithredu ar gyfer gweithredu Ildwch chi'n ei gynllun o'r pryd-dweithio'r cyfnodd yw'r cyfnodd ar gyfer ICD. Felly, gallwn i chi'n mynd i ddiddordeb yn holl o'r cyfrifio ar gyfer ar gyfer i chi'n mynd i chi. Mae'rarchaeologist yn fyddownadau ar gyfer gyfer gweld yn ffynolau cyllidau a'n deffinitiwns, a'n dda chi, ond maen nhw'n ddiddordeb i chi, ond mae'n ddim yn rhan. Just like us. Bakr and Star found that the ICD has – I'm going to list these things now – a tendency to obscure the relationship practises and histories from which it was formed. Its users employ the fixed terminologies and standards in different ways. Different practitioners tend towards lumping or splitting data into more or less categories. Many entities evade neat categorisation. mae'r cyfgledd ar y cyfnodau yn gweithio'r chwlad, mae'n gweithio'r cyfnodau am ymgyrch, mae'r syrfa yng Nghaerhau IE, ac ydych yn ôl i'r ffordd i'r cyfnodau sydd yn cyfnodau i gydagio'r cyfnodau i'r cyfnodau'r cyfnodau yn ymgyrch. Yn ymgyrch yw'r ICD, y cyfnodau ymgyrch yn cyfrifio'r cyfnodau gyda gwasanaethau sydd yn gwirionedd i gydagio'r cyfnodau sydd yn gwirionedd Ychydigwyd yw'r gwneud ychydigfyniadau yn ymdweithio'r iechyd yw'r pethau fflaesun o'r modificatio'r ystod yn gweithio'r gwneud. Yn Gyling, mae'r ffordd, oes? Felly mae'r gwaith o'r oedd yw'r oedden nhw ar gweithio'r cyfnodd ar gyfer ar gyfer aelodriaeth. Fy ydych chi'n gweithio'r gweithio'r problemau, fe wnaeth chi'n gweithio'r oed. But Barker and Starr argued that they shouldn't necessarily be viewed in a negative light, as modification and variations are important, making data relevant at a local level for their primary users. They argue that resistance to standardisation is necessary, as that is how information systems are able to evolve. Data managers should, in their opinion, abandon their faith in creating unified systems to describe the world and instead focus their attention on developing strategies to better handle the diversity and practicalities of the world. And Barker and Starr argued for a new kind of data management research, exploring the nuanced topographies and histories of information infrastructures where they described as a plate tectonics rather than a static geology. I'll get it in the end. So the issue that I'm going to cover today, which is one of many, but the issue that I'm going to talk about today is one of the major issues with the data we gathered for England involved the different spatial scales at which archaeological data are managed across England. So what most people in English archaeology would consider the main or the most complete record are the around 80 local historic environment records which are maintained by local authorities. But there are also a number of national level data sets administered by bodies such as Historic England, which is represented here by Keith, or the British Museum, which also record archaeological data. So when we tested this manually through map overlay analysis of sample areas, these showed varying levels of overlap with local HER records. So the top one is the national record of the historic environment, and these were the 30-odd HERs we tested, and this is the percentage of overlap of those records in these little test squares we did. So some only had really low levels of concurrence or overlap, whatever you want to call it, coherence with the NRAG, and some were very, like, inclusive. And these are just a couple of other data sets that I've watched for the investigation project in the DAS. You can see there's kind of a tendency for NRAG things to be integrated in HRI data sets and the tendency for the portability schemes to not be. So all of this is caused by a specific historical circumstances through which the various data sets have come into their current form, which is a topic that's way too large for this presentation. And this issue, in practice, for us, required a multi-scaler approach to its solution. So for the smaller case study areas, we manually checked where overlaps through map overlay and defined them as relationships in our database. For larger case study areas and at our national level, we had to rely on some kind of automated process, and the way I did that was through simplifying all the terms even further to further different types of sites. And then filtering the data on a presence-absence basis using various resolutions of spatial bin. So the kind of problem, or however you want to look at it, with this method is that it relies upon treating the inputted data as if they are accurate, i.e., giving the data the benefit of the doubt and assuming that it's unlikely to be wrong in any important way at the scales of analysis that we operated at. So, yeah, that's plain. This is just a video of our web app, which you can have a play around with. You can go to the address at the top. So in practice, when we compared our data against data collated through more intensive manual processing methods by other projects, the only real differences we saw in our data from those were in the detail. The broad-scale patterns were very consistent. In the few cases where we could see larger problems, these were readily apparent as they showed clear step changes at the edge of local authority areas. Creating these multiple data sets designed to operate at different scales as you can see as we zoom in and out on this enables us to approach a wide range of questions relating to past processes that themselves operated at a wide range of scales. And when we opened up this data for exploration by members of the public or other people like you all or everyone else in a familiar sort of online setting, we facilitated much easier access to data which is publicly available but relatively complex to get hold of and also to examine for a non-specialist, so it classifies democratisation. And also by offering the ability to download these synthetic data sets to anyone who wishes, although that would need a bit of GIS skill, we're enabling other researchers to examine a background picture of English archaeology which they should find useful to contextualise their own findings or to challenge ours which I label under disruption. Unfortunately we weren't in the end able to share our raw data because we ran out of time to go ask all the 80 people again if they'd let us put it online and I think we wouldn't have. But this kind of binned data is reasonably precise that goes down to the nearest kilometre and at least lets people have some idea of what's out there either in their local area or if their research is what men are on a wider scale nationally. So we would argue that digital archaeological data sets are characterful with diverse histories, contents and structures and also with gaps, inconsistencies and uncertainties. Moves towards interoperability have helped to alleviate some of these issues but the very characterfulness of data in and of itself can be valuable in terms of gaining insight into shifts in archaeological practice. The many individuals and organisations that have had a hand in creating archaeological data sets over time mean that it's hardly surprising that they're complex and that sometimes their relationships to archaeological entities in and of themselves can appear ambiguous. But to reiterate it from what I said earlier archaeology is not alone in this. So the three major conclusions that we came to in the paper that I referenced at the start are that one Despite their idiosyncrasies, archaeological data sets can perform amazing work so we should not give up on them just because of concerns over objectivity etc. Metadata and ontologies do help here obviously. Secondly, it's vital to continue the quest to understand data better in their own right. This not only enables us to understand how to use data better and how to interpret the patterns that we find but can also provide insight into ourselves as a research community. Given the previous discussion, it's essential that we develop more methods for working at a range of analytical scales. It can be fruitful, interpretatively, to pick the best data for our analyses but it can also be very productive to try and look beyond the usual suspects and embrace available data on a wider and more inclusive scale. This is especially the case if we wish to justify our existence as a discipline, particularly commercially, at least for us in Brexit Britain. Increasingly, economically difficult and uncertain times. Thank you David Cameron.