 Goodbye, okay it's very nice to be here. So I guess one of the main problems in Bw apida is the reuse and synthesis of archaeological evidence, isn't it? You simply can't get at the data or ac mae'r argymell, mae'r argymell, mae'r argymell i'r anodd ac mae'r argymell, mae'r argymell i'r anodd ac mae'r anodd yn ffantastig o'r tyfnwys a'r anodd i ni'n gweithio. Felly, cyfnodd yn gweithio y gallai cyfnod o ddau, cyfnodd o ddau, ac mae'r ddau o ddau, gweld eich cyfnod i'r ymddangos, cyfnod i'r ddysgu'r ddau. Felly mae'r ddweud yr ysgol yn ymddiadau i'w ddod i'r ymddangos, yn argymell ar gyfer archiologi. Argymell argymell argymell yn ymddiadau. Felly, mae'n ddysgu'r ddysgu, ac mae'n gwybod, mae'n gweithio i'r ddysgu. Mae'r archiologi yn ymddiadau i'r UK, and not within academia so may you can't often access applications because they're behind payrolls. Many of my colleagues who work on exactly the same material in commercial archaeology companies and may you can't pay to access the symptoms that I'm doing on their material. Of course, lots of research funders now mandate making your data and your research outputs open so if you're funded by them, mae dyna fydd tu chyddeudiau. A, mae, wrth gyd, y mwy gyd, amser bryd yn bwysig arlaethau archaelogicol a drylch�u'r pethau, yn ddiwethaf i gyd, ond mae'n mynd arfwyntion. Roedd yn gyd yn gwneud ysgol darllenol. Mae hynny wedi cael ei wneud a phasaf eich bod yn gyd i'r unhysgr ac mae yn cael ei wneud. Ond o'r cysylltu gwirais, mae fydd yn ddwyldiau ond is subject in archaeology, particularly by Doug Rock's McQueen's, and there's a fantastic website with a database of open access journals, and there's been a kind of, I guess, gradual movement over the last 10 years to provide more open access publication. Remus, so we have a number of, that's called in gold publication places where you essentially pay an article processing charge and your article will be open access under CCBY licence forever. So some of these are kind of low cost options, so things like open paternity, journal of computer applications and archaeology, published by Ubic2Press who have a £400 apc and are one in a very sustainable, scolar-led way. We're increasingly having more commercial publishers setting up fewer array journals such as Star, which will have a higher apc, but there's a growing number of options out there, and there's lots of exciting journals in countries in Eastern Europe and Central Europe which are moving towards full open access, which is great, but there's much less movement between the kind of, I guess, top tier Anglo-American style journals. So what we do have is a large number of hybrid journals where you can pay a fee to have your article open access that many of the articles in there will still be behind a payable, so at least if you have research funding you can pay for this, but of course if you don't have research funding you can't pay, and so you're then behind a payable still and you suffer kind of imbalances. And of course the apc's are very high, so what's that one, Archaeological and Anthropological Sciences, it's £2,080 to make your article open access, and there's no clarity as to what these apc's, so article processing charges are, and lots of these journals are being told that they have to set a high apc to kind of retain journal brand and be seen as a high quality journal, and they have to charge a competitive charge, and it's not a free market because we can't choose where to publish really, it's where, who's going like that. So this is a big problem, and it's not sustainable, and one of the key things on the horizon is this thing called Plan S, and I've forgotten what the S stands for again, sudden I think. Anyway, so it was launched in autumn 2018 by a science coalition in Europe, so it's research funders from 12 different countries who've come together and set out a kind of outline for where May 1 research, funded research outputs to be, so by 2021 any research funded by one of 12 single countries has to meet these requirements. So the key things is that hybrid journals are being phased out, you need, but you can publish in closed access journals if there's a zero embargo on your postprint, so that is okay for some journals, such as CUP journals, but it's basically calling for a move towards gold open access publication, which has big problems for lots of society-won journals, such as things like medieval archaeology proceedings of a prehistoric society, botania, where a learned society gets income from publishing that journal, so if that income source goes, lots of scholarly functions that that learned society currently offers need to be funded elsewhere. Anyway, so this is a big challenge at the moment, so it will be interesting to see how many publishers in Europe react to this, but one option which I'm involved in is moving towards more scholar-led consortia type publishing, so one quick example is the Open Library of Humanities, which is a charitable publisher run by academics at Birkbeck University in the UK and they're funded by a global consortium of universities who all pay a small amount of funding, so about £2,000 a year, so a lot less than many journal subscription packages, and then that means that anyone can publish in these journals without paying an APC, which is fantastic for scholars in the humanities and the social sciences, where there simply isn't the funding available to pay £2,000 an article. So it currently hosts 26 journals, which is great, and I added a journal called Trage of a Fluentical Women Archaeology Journal, so we essentially host new journals, so we made an application to turn the Fluentical Women Archaeology Conference proceedings into an open access journal, and they accepted our application, so we now publish through RLH, so we can publish 13 articles a year, there's no APCs through the office, so post-graduates, ECRs, anyone without a huge funding grant can actually make their research open access and others can utilise it. And interestingly, in anthropology, a few of the big journals are pursuing a similar kind of consortium approach, so I think there's a news of a deal with unburgrun publishers whereby they'll kind of come together to host a fully open access series of journals, but sharing infrastructure and editors, because of course there's all costs behind publishing. So to me one now to open data, as anyone who was in the session on Thursday, which David organised with Courtney, there's lots of fantastic big projects across the world, putting money into digital research infrastructure, such as the ADS, the Al-Yadni project, so there's very large data archives that you can use to archive your data and make it available. The problem is many, I guess, individual researchers simply don't know how to archive their data properly, don't have money to do that, or maybe don't see the kind of use or in kind of usefulness of making their data available. So as an archaeobotanist, I'm very interested in using other people's archaeobotanical data, and in trying to synthesise this, I've spent far too much time typing in numbers into spreadsheets or databases, but simply because the data is horribly published and there is monographs and PDFs and obscure places. So I wanted to see how much data is actually being made available. So the largest other study that I know of was conducted by Ben Marwick and Susan Pilaberg in 2018. Simone looked at articles in the journal, in several journals, so archaeological science, archaeometry, journal filled archaeology, etc, and they were looking at LIFFIX data and basically found 53% of 48 articles did make their primary data available. Simone's 47% didn't, when they emailed all of us to say, can I have your data? So no, it's yellow, and clearly most people said no, or I can't find it, or I'm still working on it. So I spent some time reading any articles that published primary archaeobotanical data over the last 10 years, and I mean counts of seeds, and that's a very widely accepted data standard in archaeobotany. So what we do is we take a sample from one context, we count seeds to species, we identify them, and then what a lot of people do is take that data and somehow crunch it into summary data by phase or by site, which is useful for interpretation and analysis, but it is not useful to other researchers trying to do new analysis. So basically I went through every single article, so about 239, and it said okay, is this kind of data available to me, and in what form. So basically the red bars are no, that data is not available, so that's the highest category across many journals, so Vegetation History and Archaeobotany, which is our subdisciplinary kind of main journal, 50% of journals do not have articles, do not have their data, PNAS, only three articles, but none of them had their raw data, which is ridiculous. Plus one also surprising, and Coffinery International, yeah, so there's a lot of articles being published, which are the main reports of that assemblage, they're claiming new research findings, but they're not providing their data, and there's only one article which has its data in a repository out of 239, so something's clearly not kind of percolating. People are sharing other types of data, it's often summed per site per phase, summed per site, on a shared graph, but not with data, or the supplementary data tables have disappeared, so it should be there, but somehow it isn't. Um, why are people not doing this? Well, why are archaeobotany's not doing this? There's a big lack of training in archiving data, at no point in postgraduate training was I told or encouraged to share my data or in what form, should I do it as a CSV file, or should I just make it into a PDF? Who knows? Often specialists have to publish their data set in the site monograph, because that's what the excavation director wants, so you're then kind of stuck between trying to push out journal articles, being told that you have to put your data in the monograph, so then you're kind of conflicted. We don't really think about future reuse until you do it and then you realise how much of a pain it is. It might be someone that's stopping you, such as a heritage agency, or a developer, or your PI doesn't want you to. Peer of reviewers often aren't asked about data availability, and they are increasingly in cross-filled journals like Nature and Science, but in journal, well-plogical science, you're not asked. Cost, or no, some cases are free, so lots of universities now have data archives which are free for now in some cases. Finally, if you do make your data available, you're not given any credit, which is a big problem. So to come on to my final point, we also need open methods, because otherwise there's no way to test what others have done, or kind of build on it and make research and synthesis better. So I also read 107 synthesis articles using Arkham Town for Data over the last 10 years in 20 journals, and basically said, okay, aren't I just citing their data? So not even are they making it available, is the code available? It's another citation, and I think it's about 23% overall, do not cite their data. So the main article in my kind of sub-research field, which found these major patterns in plant food consumption in Roman Britain, has zero data citations or site lists, so there's no way you can do that without doing the whole process of data collection again, and then you wouldn't even know if you had the same sites, because, and yes, I could ask to be offered, but you know, at some point, they are going to leave academia, be uncontactable, and then what was the point of all of that research funding? Anyway, I'll stop. And also over time, there's a bit of improvement in data citation, which is good, but there's still a lack of citation. And it's an interesting thought that lots of the citations, a reasonable number are in supplementary data, which means you don't get the citation link on Google Scholar. So even if someone cites your data paper, it doesn't help with your pitch index or your citation count, and you can't trace the citation links. And that's without even thinking about things like access to methods, you know, how many correspondence analysis plots do we see where we have no idea how a data reduction or a categorisation works. There's a few papers now making code available, which is great. So one of my favourite examples is a paper by Arun Parahani, who I think is at Nirwada. So this is a pretty typical archaeological primary data article, and his data is all available, but in mental data, but never mind, our code is there. So that is all lovely. So I guess in my own work, I've learned a lot from speaking to people who are not archaeologists. So we have in the UK a new organisation called the UK reproducibility network, which has nodes in different universities. So we have on Arun Oxford, where I talk a lot to psychologists, social anthropologists who are doing wonderful things. And then there's also, of course, a centre for open science in the US, which is really great for bringing people together. And there's, I think these kinds of things could have really big implications for archaeology, like preprints are just kind of catching on, but would usually speed up a research process, but I think most journals don't have preprint policies yet. And then also things like registered reports. I mean, so much of archaeological publishing comes down to, did you find something new or early or exciting in your assemblage, which doesn't reward people who are just doing very good work. So what a registered report does is you say your research design, and how it fits in the kind of research agenda. You submit that to a journal, based on the quality of your questions and your design may say, yes, I'm going to publish the upcoming paper. So if you get negative results or you don't find the earliest seed or the latest seed or whatever, may still publish it, which is great. So it takes away this kind of gaming of just finding sexy things. So I'm going to stop talking now, but in conclusion, now our big news coming towards open access, which are great, but have lots of power inequalities and challenges, which we need to think through as a discipline. We're still stuck on prestigious journals, despite the fact that most institutions at signatories of door, it doesn't seem to be made in effect. We need to teach post graduates and undergraduates about research data, and that CSB files are great and PDFs are bad. And there's a lot of exciting open research approaches coming through, which will have big impacts. And if you want to learn more, there's a really exciting conference taking place in Oxford in March. Great, thank you.