So yeah, Stephen, you are a professor of information services management of the University at Sheffield in the UK, and we are very glad to have you here at our virtual conference, and we are looking forward to your talk. But you are also an associate director of the Research on Research Institute, which is funded by Digital Science and the Wellcome Trust.

And maybe the people that are in the audience also know you as the, I would say, father of the Sherpa DP project. So I guess you have been concerned with open access for a long time, and you know a lot, and we are

looking forward to get a little bit of your knowledge, and the stage is yours. Please take the 20 minutes and enlighten us. Thank you very much, Isabella, for that very kind introduction. Thanks Nick as well. And hello everybody, it's a real pleasure to be with you and a real privilege to be contributing to the conference and I'm looking forward to our discussion.

What I'd like to do in this introduction, then, is to touch on some of the issues I anticipate us discussing later on through presenting to you a study that I and the colleagues listed on this slide conducted last year on the state of the art of preprints.

And to report some of our findings from that study but also to reflect on the significance of some of those findings in the light of more recent developments. I don't know about you, but this time in 2019 seems like a very long time ago, and a lot of water has gone under the bridge since then.

So what I'd like to do is reflect on some of the issues that were raised in our report from 2019, and see how they've been, or begin to see how they've been reflected in more recent developments.

It's often said that major crises of the sort we've seen in the, in the shape of the pandemic, major crises accelerate or reshape existing trends, rather than creating new trends.

And I think that might be what we're seeing in relation to preprints but I'd be very interested in your perspectives on that. So, this was a piece of work that we carried out, commissioned by the Knowledge Exchange, a consortium of European digital infrastructure providers and digital service providers for higher education in various different European countries.

A piece of work they commissioned that I undertook with Research Consulting, a consulting company in the UK, where they wanted us to survey the field in relation to preprints and to advise them on various different policy options that there might be.

So we produced this report in September 2019, we published a preprint through F1000 research, more or less the same time, and then responded to the open peer review process that F1000 run, and so the final article was made available in late 2019.

And of course we were carrying out this research at the time in the context of what we called the second wave of preprint servers. A number of preprint servers like archive or repack are very well established but what we've seen since 2016 or so

is a rise, a second wave of preprint servers being set up significantly in areas, not traditionally associated with green open access. So our study focused on these areas, bio archive, chem archive and psych archive, and the communities that they were serving

to see what was happening in those disciplinary fields in order to look at the impact that preprints were having there. And of course since that time we've seen a rapid rise in submissions of preprints to those services and others, particularly in relation to COVID-19.

And this was relevant then in the context of the debates that Nick has just mentioned particularly around plan S, and the hypothetical plan that was put forward by those favoring preprint providers which they call plan U.

But we've also seen this highlighted in more recent discussions, as well as the development last week that Nick highlighted with eLife announcing that they would only consider articles that are already being made available in preprint form. The same week, just a couple of days later, there was a statement from a significant number

of open access publishers that was arguing that green open access was actually undermining open access progress. And so this green gold debate that has been characteristic of the whole open access space for at least

20 years is still very much with us. And that was the context within which we were, we were working. So for the rest of this presentation, what I'd like to do briefly is to talk about the themes that you can see laid out here in relation to our study and subsequent reflections.

And what I'm going to argue is that study was, well, it was designed to be a stock take of issues in relation to preprints and scholarly communication to inform policy and action as well as understanding.

And what we found was that defining preprints is still controversial and understanding of the possible role of preprints variable in different communities. But nevertheless, preprints are seen by many and increasing numbers, having a particularly significant set of benefits attached to them.

But at the same time, there are concerns about preprints and the risks and the drawbacks that they potentially pose as well that need further investigation and need to be addressed. We saw that there are different possible operating models for preprints in existence and so the clarity around how

preprints will actually operate is still to be achieved and a number of important questions need to be resolved. We saw that infrastructure to support preprints was already in place to a certain extent but still needed development, and that adoption processes were at different stages in relation to preprints as an innovation.

And so therefore the future of preprints was still unclear and so a number of scenarios were still possible. So that's the the rest, that's a roadmap for the rest of this presentation. Let me talk about each of those points very briefly.

So as I've said, our study focused on two things, not just increasing understanding of this space but also informing policy and action so it had both a descriptive and a prescriptive side to it. And so what we wanted to do to inform this group of policymaking and service providing bodies was

to look at preprints in relation to the different actor groups to look at what the benefits of preprints were and the potential drawbacks as well and therefore the incentives and disincentives that were operating within the system. And therefore to inform strategy policy development as well as infrastructure and service provision

development as well in relation to the different actor groups involved in this whole ecosystem. And to do that we carried out a qualitative study, the first of its kind as far as I know, which undertook a detailed literature review, but also a set of semi-structured interviews

with a range of different actors in this space focusing on those disciplinary areas I mentioned earlier on.

Using the theoretical framing of innovation diffusion theory that I will mention in just a moment, and then having conducted the interviews and transcribing them, carrying out a detailed thematic analysis of those interviews followed by reporting. Now at the end of these slides, which I will share with the group, we have all of our participants listed, but

this is a snapshot of the sample of participants. You can see from a number of different countries representing different actor groups, including some researchers, both those who had engaged in preprints, that is submitted preprints to preprint servers themselves and those who hadn't.

Actually non-engaged researchers, as is often the case with these kind of studies, were incredibly difficult to recruit, but nevertheless we did consult them as well as other key stakeholders in this whole space. And the first thing we found is that an understanding of what preprints are

and of their role is still variable in the scholarly communication community or communities. And this isn't just an interesting semantic discussion, it does affect a view of what preprints are and

of the benefits that they can potentially deliver and how they fit into or potentially change scholarly communication. But we saw a number of presumptions in the literature and in our data about what preprints are, but these aren't always in sync with each other in the way that they are described. We saw that preprints are traditionally seen as representing a particular genre of communication, a recognisable scientific paper, that's

the presumption of what a preprint is, and that's the scholarly genre in which they normally thought to fall. Timing is important, preprints are seen as prior to publication, and versioning is also important, prior to peer

review. Those two actually timing and versioning are often conflated in people's descriptions, although they are conceptually distinct. Preprints are normally assumed to be open access, accessibility is seen as a key aspect of what preprints deliver. And interestingly, responsibility is often presumed to be,

that is responsibility for posting preprints, is often presumed to be an author responsibility. So preprints are sometimes defined as things that authors have made available, which I think is quite interesting, I'll touch on that in just a moment.

And then the presumption is that preprints are valuable. In some cases, people put the value of preprints into their definition of what a preprint actually is. A preprint is something that is valuable, and it's interesting that value judgement is at the heart of many of the arguments of what constitutes preprints.

So, greater clarity is definitely needed through discussion and further investigation into this whole area and how it fits with different scholarly and disciplinary communities is an important question about how the future looks for those different communities. Now, as far as the benefits highlighted by our research compared with the literature are concerned, we can see that

many of these are kind of confirmatory of what we would normally presume to be the major benefits of preprints. They focus on accelerating and widening the availability of research and the advantages

that come from that, but also preprints are seen as potentially at least, reshaping or having a role in reshaping scholarly communication further down the line and it's that that often proves to be controversial in terms of their long term importance.

And at the same time, preprints were seen as having significant drawbacks. These were focused on what you might call the lack of quality assurance associated with preprints, and the consequent risks that that creates in the scholarly communication environment.

So a lot of these focus on the notion of trust, trust in scholarly communication, and how trust can be developed when the normal markers of quality and the normal indicators of trust are not necessarily present.

And so some of our research focused on what those markers might look like in a preprint world. Now it's interesting and important to say that many of these drawbacks as they were described to us in our interviews were often couched in hypothetical terms. In other words, people saw these as risk rather than actually manifested as problems, and often highlighted the need for more

evidence to see whether these risks would be realized, and therefore what kind of drawbacks they would actually accrue in reality. One of the things that remained uncertain though that was highlighted by our research was the question

of operating models in relation to preprints and this impacts on the wider scholarly communication ecosystem as well. So as I've said, preprints are traditionally associated with authors posting pre refereed versions of articles to preprint

servers which are normally standalone platforms in some kind of way, in other words they're not associated with publishers. And yet, as part of the second wave of preprint service we've seen publishers become more and more involved in managing preprint servers

or setting up preprint related services themselves, and at the same time becoming involved in posting on behalf of authors preprints as well and this whole space then becomes more ambiguous and more fluid. Are we seeing the rise

of if you like of a publisher centric approach to preprints, rather than a researcher centric approach, and what consequences does this have for preprints and a system of scholarly communication involving preprints. Some people saw the entrance into this space of commercial providers for example. Some people saw this

as potentially dangerous and reinforcing some of the dominance of large commercial providers in this space, others saw it as creating the potential for sustainability, and therefore as being useful but it's an open question.

One thing that surprised us about our data though, and this is quite interesting because it touches on a number of the papers that we're going to hear from this afternoon, is the importance of social media and particularly Twitter. Some preprint service providers really emphasize this, that they saw the success of their preprint

server being accelerated and amplified by social media and particularly the role of Twitter and the role Twitter was now playing in scholarly communication in allowing authors to publicize their preprints

and allowing some kind of exchange and debate around preprints as well and that was those were seen as incredibly important and Twitter came up again and again in our data which we thought was particularly interesting in terms of the role that it's now playing to aid discoverability, basically, of preprints.

Having said that though, Twitter you could say is part of now the infrastructure related to preprints. Although there are different components of the infrastructure in place which we describe in the report, in particular in some detail,

there are still elements of this which are not in place or are not sufficiently joined up or interoperable. And so there are a number of problems associated with this whole space that still need resolution. Just one example, it's still not as easy as it should be to clearly associate a preprint on the one hand, with a formally published paper on the other,

and to show that they are different stages of the same work that is still not as easy as it should be and so the standards and the infrastructure in place to address that kind of problem still need to be, still need to be developed.

Now, as I say, we used innovation diffusion theory to frame a lot of our analysis and indeed inform the design of the interview questions in the first place, but what innovation diffusion theory does is show how major changes can be adopted by communities going through innovation adoption stages that are listed here on the left hand of the slide.

And it was, it was fairly clear that different communities as well as different individuals are very different stages in this adoption cycle. And many were still not beyond the decision making stage. Knowledge was variable.

People were, to a certain extent, being persuaded, but the decision making stage seemed to, a lot of people seem to have got stuck there and it's really important to observe that very often in the adoption cycle, decision making is affected by a behavior of peers.

So this was seen as something that was happening, could only really happen bottom up with communities, gradually taking on board the benefits and adopting, therefore, this innovation.

And that then would lead to greater experimentation, implementation and confirmation where some of our interviews have got there, but it was definitely clear that many had not, and many of their communities had not. Now within innovation diffusion theory, there's also this idea of the rate of adoption being informed

by these different factors which are listed here, attributes of the innovation, communication channels and so on. And once again, many of these issues were in place. But because not all of them were, it was seen by many of our interviewees, particularly those representing policy organizations

and funders, that mandating preprints was extremely difficult and so therefore what they had to do was encourage bottom up uptake rather than mandate top down, although developments such as eLife with publishers mandating are an interesting intervention into that

whole, whole space but the nature of the social system and the way in which the perceived attributes of preprints are seen as, nevertheless remain particularly important. One thing we did see that was a correlation in our discussions between general sympathy for open access on the one hand,

and uptake of preprints on another so people's presumptions about the value of open access was definitely seeming to inform their activity. And so a number of different scenarios, because of this uncertainty is still possible in this whole space.

We talked about three possible scenarios, the one of retrenchment going back to the well established preprint servers. A patchiness scenario talking about different fields adopting preprints, and maybe adopting with different rates and that's something maybe we're seeing now with the advent of bio archive and med archive and

others, or the possibility of a ubiquity scenario, that is, with preprints widely adopted across different fields but I think we're yet to see that that emerging, so that the patchiness scenario still looks the most likely, at least for the foreseeable future.

So the final slide then, just to highlight some of the questions that we pointed out in our research as being relevant now to be addressed. This idea of the researcher centric model of preprints

and its relationship to the publisher centric model, and to what extent that was acceptable and what the implications were of that. And the implications that has for sustainability and the monetizing of preprints services which could potentially be

a long term consequence of some of this, as well as issues like preservation being important as well. We asked the question how the advantages and potential disadvantages could be further explored by the community evidence gathered and that was

certainly something we saw the need for, although there's been a lot published in 2020 on preprints which is good to see. And therefore how different subject communities could be encouraged to consider the value of preprints in their own terms, and therefore what that means for policymakers and service providers in terms of

the policies they develop and the directions in which they encourage communities to go in terms of adoption. So I hope that's been a useful introduction to some of the issues associated with with preprints as we saw them. You're very welcome to go

to the reports, go to the paper, and have a look at that, and I'm very happy to try to answer your questions. Thanks very much. Thank you very much, Stephen. That was a wonderful talk, and we have time for questions

and there are also some questions in the chat and I will post them to you. So the first question we have is from Iraçe, I hope I have pronounced it correctly, and she asked whether you have noticed any country specific differences in terms of use expects because from the like the 38 researchers you asked, they came from the, what is a global north.

So do you think that there are differences when compared to the global south? That's an interesting question. And because our research was a qualitative piece of research, we

can't sort of claim generalizable results here in terms of being able to track that accurately. What we did get was a kind of richness of data associated with people's motivations of thinking, and the vision people have for the preprints. We, because of the funding of the research, we did agree with the funder that we would confine our research to the global north countries

in which they have a presence and so that was an intentional part of the research, but a number of people we talked to did discuss the relationship of preprints with global south providers, and particularly there in the context of Plan

S, which was, as it is now, a point of debate, a controversial kind of approach. The contrast was made by some of our participants to Plan S, which was seen as expensive in the way that

it would have to be implemented, compared with other models of scholarly communication which used more kind of green centric approaches, which could be seen as creating cost savings in the whole system. And some of our participants pointed to examples, particularly in Latin America of

Cielo, for example, of a publisher platform, you might call it, that has some functions rather like a repository as well as a publishing platform. So people were conscious of the different models so there are definitely implications there and I think the contrast

with Plan S is the thing I would pick out as being the most prevalent in our data set. Yeah, so we have a related question from Anup who asked, what are the challenges in Plan S? Is there required funding available

at the institutional level for implementing Plan S? You touched that a little already, but maybe you know a little bit more about it. What are the challenges of Plan S? That's a big question. Well, one thing I can say is, well, I can say a number of things. As somebody

who works within an academic institution and is a member of our institution wide open research committee, we are now grappling with the implications of Plan S for us at an institutional level and that's quite interesting because it does require, at least if the funders adopt Plan S in a robust way, as some have already done,

like the Wellcome Trust in the UK, they are now very clearly spelling out the implications of their implementation of Plan S and there are some crunch points there in terms of

potentially constraining where authors can publish or at least if they publish in certain journals, having to retain the rights to deposit their work in a repository, what that means for funding and so on are really quite significant. I have to say, as a general point, I'm a supporter of Plan

S. I think it has been a very, it has moved the whole implementation of open access to a new place in terms of its robustness, in terms of the level of coordination across different funding agencies. One of the things I argued for a number of years was that

there was insufficient coordination of funding agencies and other policymaking bodies in the area of open access that was leading to what I call the dissensus, in terms of particular policy approaches, with a lot of people agreeing with open access in principle, but then finding different policy levers to pull and it led to confusion, it led to mandate messiness, it led to all sorts of different problems

and so one of the things that, when I was involved in talking to the people at the European Commission that coordinated Plan S, one of the things I really emphasised was the need to bring different funders together, to have a coordinated approach to implementation

and Plan S seems to have done that in a way that's never been achieved before. So although Plan S might have some challenges, I think it does have fundamentally some benefits and since the guidelines have been expanded in the last year or so to include provision for green open access, I think it's a more balanced policy than it was or looked as though it

was going to be 18 months ago, so I feel more positive about it, but I still think these crunch points with particularly some of the large commercial publishers need to happen in order for the direction to be fully shaped and the read and publish

deals that are beginning to develop in this whole area still need a lot of negotiation before we get to a more satisfactory position of correcting the power imbalance that has occurred in this whole industry for rather a long time and for the

academic community to reclaim, as I would see it, more control over communicating research results, which I think is important. We have another question and then I also invite you to look at the chat because there's a

quite extensive discussion on open access definitions and what preprints mean, but I think, so I think that is a little bit difficult to answer in a talk, but please feel free to look in the chat. But a question I can ask is from Steffi, and she asked how many publishers

allow the deposit of a preprint manuscript on a reputed preprint server before or while submission? And the second question or the follow up question for that is, is there a tool to check it at one glimpse or has everybody checked the publisher specific policies? So she comes probably from the author perspective.

Yeah, well, the Sherpa Romeo service that we set up in 2003 was specifically designed to address that question of being able to check easily whether publishers allow a work to be self archived, as we called it then, I think that terminology is still used now.

When we set up Sherpa, the whole colouring of open access wasn't solidified, but I think Sherpa is responsible for green open access. And its founders of the original Romeo project that included Stephen Harnad and Charles

Oppenheim, used that terminology and we adopted it when we implemented the Sherpa Romeo service. So that's what it's designed to do. It's designed to provide an easy one stop shop to check publishers copyright transfer agreements in relation to the ability to self archive.

So you can check the statistics there actually, because it does provide a view of how many publishers allow this and the majority do in some form or other. One of the things that's interesting about the data that we got there in relation to our

study was that this perception of the Inglefinger rule, as it's called, that is the rule that posting a preprint counts as prior publication and therefore a journal will not accept it a previously published work. So that's the Inglefinger rule. The perception that that was still prevalent amongst publishers was really quite high.

So even though it's a minority of publishers now, it's perceived to be a significant barrier still. And we were quite interested in that lag in understanding that actually the vast majority of publishers do allow some kind of self archiving. And that's why services like Sherpa Romeo were set up to design, to really inform that whole space.

So that's where I would recommend you go, not just because I set it up, but I think it's still genuinely useful. There are APIs to it as well, if you want to pull the information and use it for services that you're delivering yourself.

Thank you very much. Please look into the chat. Maybe you can give some answers there as well. Thank you for joining us. And I will hand over to Philipp.