Translating the generic principles of Open Science to Arts and Humanities research – A conversation with Erzsebet Toth-Czifra

Erzsebet Toth-Czifra is an open science officer at DARIAH-EU in Berlin, Germany. She has also worked as a content integration manager, external lecturer, and language teacher in Budapest, Hungary. 

She shares with Jo what Open Science means to her personally and professionally in this podcast.

To see all episodes, please go to our CONVERSATIONS page.

Which researcher – dead or alive – do you find inspiring? – Good question! I’d nominate here  Katalin Karikó, a Hungarian researcher whose contribution to the Pfizer COVID vaccine was crucial. But luckily, I don’t need to look very far to find inspiring, and living, scholars around me:  Laurent Romary (INRIA France) with whom one can discuss the craziest Open Science ideas, Jennifer Edmond (Trinity College Dublin, DARIAH)  who always reminds me of how much disciplinary cultures matter in what we are doing, Maciej Maryl (Polish Academy of Sciences) for his inspiring work and support for the Ukraine or his colleague, Marta Blaszczynska, an inspiring and badass young female scholar who is my partner in crime leading the DARIAH Research Data Management Working Group.

TRANSCRIPT

Jo: You are listening to Access 2 Perspectives conversation. My name is Dr. Jo Havemann and I’m here with Erzsebet Toth – Czifra

Is it correctly pronounced?

 Erzsebet:  Yes. 

Jo: Okay, welcome Erzsebet. I’m really happy that we finally got to record this episode and have our listeners be able to participate in some of our brainstorming conversations.

Erzsebet: Yes. Hi Jo. And I think I need to say hello to Jo’s dog as well, sitting in the Zoom setting. We all need more dogs in our lives. But thanks a lot for having me today in your podcast. I’m really looking forward to this conversation with you. 

Jo: Yeah. So sorry, listeners, you might hear some growling or whining from a little lap dog. Oh, there’s also the #Hidog open science hashtag, which is worthwhile to look up. So hashtag hidogopenscience. And then you will see some of the dogs in the open science community.

So we met a couple of years ago. So we’ve spent some time together in this open science community. And I tend to introduce open science as nothing special. It’s really nothing new in my understanding. And I think we probably also share that; it’s good scientific practice. And the difficulty now is the reason why we need to talk about it and where good scientific practice has been difficult to comply with the digital age and the variety and the high number of digital tools that have come into play. There’s also many pressure points in the academic system we will talk about, which also gave rise to what some refer to as the open science movement. And open science is like to me, it’s like organic food. It’s a label for something that should naturally be in place and be a practice. And now we talk about it in a digital era. My understanding, what’s your take on open science? What does open science mean to you? 

 Erzsebet: Yes, I very much like your metaphor of open science being like organic food. Another dear colleague, a Berlin colleague of mine and now Metro under linguist, also compared the other day open science to veganism in the sense that you can adopt, you can check what’s working for your mind, for your practices, for your life situation, step by step. For me, metaphors are especially dear because you may know that I’m coming from a background in linguistics where I’ve been studying how metaphors shape our cultural concepts, cultural discourses. And so I think I’m safe to say that for me, the whole revelation towards open access and open science started with corpus linguistics textbooks and some illegal photocopies of them. I think limitations in access to knowledge have always been something very tangible throughout my studies. So I remember episodes when we had international items on our reading lists and the lecturer said, okay, go to Google Books and try to read as much as is available from this book. Don’t worry about the Censored pages or other episodes where the institutional librarians sent emails around saying that, okay, good news we have free trial access to this and this scholarly database, Scopus or whatever. So please, everybody, just use it for your best and download as many things as you can. And later on during my PhD, so visiting libraries like University libraries in foreign countries to get access to the latest findings. The latest literature was very much a community practice. So you can imagine it like once we got the scholarship, for instance, to Berlin through BID, we sat together and we put together a list. So please, for me, bring home this and this and these records. And for me, please make a copy of this and this and these book chapters. So basically back then we had to cross physical borders to overcome virtual funds such as payroll. It’s pretty bizarre if you think about it. And then it was about so for me, having open access to publications has always been a kind of a no brainer. Nobody had to explain the significance of this to me. But also later it became clear that having open access to these traditional research outputs like book chapters like journals and the like is important, but it will not fully solve the problem. We need to go beyond papers. And it was also during my PhD, the first time we were at the stage of making international connections research wise for the first time in our lives. And in one of these conferences I heard about discourse analysis text analysis tool called MAXQDA. And I was amazed by the magic it did. And I found it super useful. And my colleagues were generous enough to share their work within this analysis tool. They share their markups, their text and notations, and I could bring it home. And it was quite a disappointment upon my return. Trying to open the content, trying to open the closed format of a closed proprietary system in my environment. And it was completely no option that my institution would buy a license for me to use this proprietary MAXQDA software. So this question of, can I run your data on my tool? Can I make sense of it in my own research environment and vice versa? It became also apparent I would say so in this sense. For me, open science is both something very global and something really individual. So on the macro level, I would say it’s about better connecting researchers and research projects from geographical regions, from disciplines, enabling scholars globally to access, to deposit, to analyze resources beyond borders and disciplines, regardless of their affiliation, regardless of their nationality, regardless of funds, institutional budget. And the beautiful thing is that on the level of individuals, on this micro level, it’s about the ability to look over your colleagues shoulders and to better understand and follow what they are doing, to follow the whole process step by step. And what is important is that as individuals, we have a lot of power to revisit and change our practices in this respect to make them more transparent, make them more accessible. You can think about blogging as a publication. You can think about writing up methodological notes as a publication, updating your GitHub repository as a publication. So there are always options which are eventually always a combination of something closed and something open. But so there are small steps that are at everybody’s disposal. Using social media, openly commenting on each other’s work, properly naming the variables and the statistics. 

So these are small practices like from starting a conversation on social media or even just properly labeling your variables and statistics and small things like that, but that are at a disposal for each and every one of us. But the further we get from the desk of a scholar or desk of a research group reaching institutional scales, national scales, European and global scales, the more complicated it gets. And questions like equity and participation in research. These are big issues that are deeply embedded and even deeply affected by their actual political realities and current issues. And I think the big question that keeps all of our minds busy these days, of course, is how these political realities change participation, for instance, for Ukrainian scholars, but also how it affects Russian academics who stood up against the aggression. So how deeply vulnerable academia becomes during war time? And what are the sources of resilience in this respect? So the work that you’re also doing, Jo, to help the situation, but also initiatives. I think you’ve already mentioned the Science for Ukraine initiative in your podcast. It’s a community group of volunteers, researchers, students from academic institutions in Europe to collect and disseminate information about support opportunities for Ukrainian scholars. But there is also another initiative that I’m following and contributing to a little bit. It’s saving Ukrainian cultural heritage online. It’s not only arts and humanity scholars and digital humanists mainly, but also culture heritage professionals, librarians, archivists who are working together to archive digital collections, digital Museum archival collections from Ukrainian institutions while their country is under attack. So these are super important contributions that also have to do a lot. 

Jo: Sorry. I just wanted to ask about saving cultural heritage in Ukraine. Could you share some of the activities that are being proposed or what people can do online to help secure cultural heritage? 

 Erzsebet: Oh, yes. So basically, I see two main lines of this very important work. One, of course, is on site, and I will send you the link to the super important work that is done by my colleagues at the Luift Center of Urban Studies. Basically, before the war, they have been providing digitizations also, including, if I remember correctly, 3D digitizations that are modeling the city of different historical periods. And now they went to full capacity, archiving the ongoing disaster, I would say, and also in parallel, they are providing onsite shelter to the refugees within these institutions. So you can imagine the situation of one day you shift from modeling the middle aged city model of Luift to offering that for people fleeing from the Eastern parts of the country and doing the contemporary documentation of the war in parallel. So that’s a very important line of work. And this other initiative that I mentioned, the SUHO initiative, Saving Ukrainian Cultural Heritage Online, has to do mainly with web archiving. So they are making sure the longevity, the long term archiving of the already digitized cultural heritage materials that are already available on the web, but we don’t know how long they will remain openly available on the Web. That’s a global initiative that mainly originated from my colleagues in the US. So it means that, of course, language competence is always a big plus. But scholars from across the world and culture professionals alike can globally participate in the saving action. 

Jo: Yeah, that’s good to know. We will share these links also on the show notes or the affiliate blog post to this episode. 

 Erzsebet: Yes. And so beyond this, but not completely independent from this, I think there is this crucial question of whether open science will solve the structural inequalities in knowledge production. And so, somewhat paradoxically, as open science is getting more recognized and established and embedded in policy and funder mandates, it’s becoming really mainstream. And parallel to this, we just began to understand what can go wrong or potentially what went wrong already in its implementation. So a colleague of ours, I think he will also be sooner or later guests to your podcast, wonderful podcast series, Jo. Tony Ross Helloer and his team just published a great deal of very spot-on insights on how open science could become just the extension of such privileges. And this is something that is very visible in my everyday open science advocacy practices as well. For instance, when it comes to open access publication fees, of course, richer institutions are more in a position to pay such fees. An established tenured professor is more in a position to publish their book. Open access or research teams with solid external funding will have more time and effort to prepare data for open dissemination, by contrast to somebody who needs to take an extra job on top of their salaries to make ends meet. So it’s really just a question of privileges and capacity sometimes. And I think this drives our attention to the complexities. And it suggests that these simplistic blanket solutions will not work when it comes to making open science mainstream modus operandi within an outside academia. And the more I spend, the more time I spend with advocacy and healthy research teams, the more I believe that along the same line, epistemic cultures also really matter. So when you engage in conversations about open access and open research practices, it’s very interesting to see how, for instance, publishing means very different things to each and every one of these research teams and disciplinary communities. And these differences translate into two very different needs. And honestly, this is something I truly love in connecting scholars with the ideas of open science. But this, on the other hand, is also something truly challenging. So on the bottom line, I think we cannot make this message loud enough that open science will not truly open until it’s equally inclusive, with all the disciplines, all the knowledge areas, and all the geographical regions as well. 

Jo: Yes, I totally agree. And I also agree with Tony’s paper, which we also will link to in our show notes and blog posts, that there is this misconception of when I asked early career researchers if they already know about open access nowadays, since two or three years, they say, yes, of course, we’ve heard a lot. And then the next sentence will be that it is very expensive, we can’t afford this or the library pays for it. But that’s not the point. This is not what Open access is about. There are few stakeholders, a few of the publishers and journals that take advantage of it and try tremendous article processing fees. But that’s not what open access means. Primarily, as you explained, it’s about what the word says, making research outcomes publicly available, which they ought to be. And yes, there are fees involved, but the fees are certainly not in the thousands of euros or dollars. And then there’s also other barriers that Tony your colleague pointed out, which you touched upon, and you also know the direction towards it. What are the limitations of open science? This is also one of the major fears, yes, it’s nice to have everything open, but many researchers feel, probably rightly argued, there’s also responsibility by researchers to make sure that the information is curated well before it goes public. And that’s also a misconception that I often experience when I advocate for open science in my courses and conversations. Open science doesn’t necessarily mean that you put everything like that, that is, to publish the whole workflow one by one, but it asks for thoughtful consideration, planning, documentation, and that then should be publicly available. And yes there is sensitive data in all disciplines. Medicine is the most well known example with patients’ data, but there’s many other sensitive data in social Sciences with interviewees. And of course, you don’t have to make public any private information about research participants.

Would this be something to talk about for a few minutes? The limitations of open science and the misconceptions where people often have understandable fears to embrace open science because they fear they make themselves vulnerable, or they may also cause harm by making unsolicited or uncurated information publicly available? 

 Erzsebet: I think one of the biggest reservations against open science that I encounter in my daily job is to what extent is open science relevant to the communities, the specific disciplinary communities that I’m serving, like arts and humanity, or we are serving entirely like arts and humanities scholars. So the big question here is how much science is in open science? And if you look at the dominant impact of science engineering, techniques, mathematics fields on the big open science agendas and also on the practices, then I think the answer is obvious quite a lot. There is quite a lot of science in open science and working in arts and humanities feels we do not always feel addressed by these general keywords like reproducibility or the publication of new results or preregistration. But we are wondering instead, like, okay, where are my books from this story? What if I don’t generate my data but instead collect them from the library and I don’t have full control over it? How can I openly share it while the ownership over it is not mine, but it’s shared between the subject of the image and the producer of the image and the cultural institution who is taking custody of that photograph? Or other questions like what if statistical reproducibility does not make any sense in my research because I work with qualitative methods? Or what if I’m working in a language other than English or working in a small discipline and therefore my citation impact will never be comparable to that of a physician? Or how can I guarantee the equal representation of smaller cultures who will reward the efforts of upscaling, for instance, NLP tools to different languages under resource languages? So for me, of course, as I told you before, open science, open access and the whole idea and then the principles underlying became really game changers very early on. But I still remember very well these initial frustrations that I came across. And the more I learned about what we call the open research culture, the more confused I became about how my own research fits into this. And the more I learned about it, the more I wanted to change this. Luckily I was fortunate enough to meet people who were puzzled by the same frustrations. And I just realized that I’m not alone with these concerns and we could start a discussion. And the support and the encouragement I received from these communities from these conversations was strong enough to start seeking ways in which I could merge this home of my scholarship like humanities and open science, and start learning about and supporting open research practices that are organically growing out from arts and humanities. And you mentioned open access and the discussion about open access and publication fees. I find it super important to highlight that there are some innovations, there are some solid contributions to open research, open access that come from humanities, and what we call the diamond open access publication models is clearly one of them. It’s basically the idea and the model where researchers can openly access publications and authors can publish without publication fees. So in arts and humanities fields where there is not a huge amount of external grant money going around and people are publishing as part of their daily jobs, as part of their salaries. Paying article processing charges, or even super high book processing charges never really became a reality on a large scale. So diamond open access models where libraries team up or institutions team up to co-found open access publishing infrastructure directly is a response to that. Yes, but you were asking about limitations of open access and finding a good balance between the open end and between the closed. I think that in many cases, scholars from these arts and humanities fields are already practicing and even fostering, I would say, genuine open science without explicitly calling it. So there is this myth of the lonely author who is working alone. But if you think about it, knowledge is very rarely, if ever, produced by individual human individuals only. And in reality, we routinely build on each other’s work, we build on each other’s resources, we analyze tax, corporations created by others. We interlink collections of artifacts to establish hidden connections between them and reach them or use scholarly databases for discovery. And so this has been recognized by humanists too, along with the advent of digital technologies and research. So it’s a discussion from 2010 and humanists like the philologist Gregory Crane are really influential in this respect. So there is a discussion on that. I mentioned that there are tools available, there are solutions available that are not necessarily explicitly, open science branded, but are naturally emerging from doing collaborative research digitally, I would say. And these solutions usually reflect on limitations, infrastructure limitations, or legal ethical limitations. For instance, you must have heard about the initiative abbreviated like IIIF, the Interoperability Framework. This is both a guidelines set of communities’ collection of good practices, but also a collection of tools and solid infrastructure and know-how to enable arts and humanities colors to embed images to their publications or to their working environment directly from the cultural heritage institutions along clarified reuse conditions along clear licensing information along clear protocols. Also in terms of technicalities, so many of these are brilliant examples of how limitations and epistemic specificities can actually sparkle innovations if scholarly communities join forces and address those challenges in more depth. 

Jo: Yeah. So you mentioned quite a few legal aspects which many researchers feel like the past. But I can also assure your listeners it’s actually not so complicated as it might seem. And also, as you pointed out, these legal aspects to consider is also to secure ownership, to shift ownership when necessary for the image creators or painters or whoever owns the Copyright to digital or paper based work or a text or data set that as researchers are then allowed to model it and to repurpose it and then also our results to be able to repurpose. So there’s Creative Commons licenses which have been popularized, but there’s also more traditional license types which are still at play. And it’s really not a certificate. It’s just a matter of reading up what each license type implies. Well, it can be difficult. 

 Erzsebet: I would challenge this idea of Creative Commons as a solution. That is our solution, because creative, it’s a whole licensed family that’s solved all our problems in this respect. Also, because what I see in many cases supporting scholars in open and sharing their work is that in many cases, these legal challenges are not only legal challenges, but they are coming with a whole complex package of infrastructural ethical, technical, conceptual implications so it can get super complex. I mentioned that one question that I was wondering before starting this job at DARIAH that I’m going to talk about is hopefully there are so many diverse disciplines belonging under this umbrella term force and humanities, musicologists historians, literature scholars, fashion designers, like NLP specialists, what have you and whether we can talk about shared challenges at all when it comes to their data workflows. And one such shared challenge, what I previously mentioned is that in most cases this data workflow starts to originate from outside of the institutions. Outside of what would be a hypothetical research lab. Humanists are not generating their data, but they are collecting and cultural heritage institutions are a natural partner and natural starting point for that. So just telling you an example of such complexities, one of the first practical challenges I had to come across and starting this open humanities job was a case of publication, an archival resource guide. It was written on the topic of the first world war and first world war culture heritage. And what happened is that just before the publication, the research team received an email check up from the publisher asking them, okay, so before we send this material to the print, how about the license of these images? Even after very thorough, very repeated checks and research, they were not able to find the reused information to all the images and eventually they had to leave a significant portion of them out of publication. And I think this is clearly the worst that could happen to culture and cultural materials and research materials for that matter too, is that due to these reasons, they became invisible from the cultural discourses, from the scholarly discourses, and, you know, especially materials that are likely to be under Copyright, like early 20th century materials and 20th century materials in general. They are super exposed to that. There are topics, there are authors, there are heritages that nobody really cares or wants to touch and investigate, not because they are under Copyright in many cases, but because the Copyright holder is unknown. So this is really, I would say a distortion. And this is something super important to keep in mind that in this respect, arts and humanities research is super dependent on digitization and the open availability of these digitized collections. And it will affect their working conditions because we really cannot expect in the foreseeable future that each and every culture artifact will be digitized. And maintaining this watchful awareness towards this epistemic, marking this asymmetry or this inequalities in the availability of cultural materials is really something important because there is a great deal of topics, authors, materials that are really sunken behind this what I call this digital horizon we really need to remember of what is invisible from digital discovery systems, for instance. 

Jo: Yeah, I totally agree. Also, I want to apologize. I didn’t want to come across simplifying the legal aspects. I come from a Bioscience background where things are very repetitive, and I would argue it’s maybe not as complex legally as it is in the social Sciences and arts and humanities and along images. You explained how the complexity kind of evolves and where it comes from. And Copyright ownership of data. It’s a constant struggle and questioning and shifting also in the Biosciences. Actually, with my statement earlier, I just wanted to take away some of that here to encourage early career researchers to look into the legal aspects of their work because it is important and it does matter. And it also empowers researchers to take ownership of the project and to assess ownership of each component that contributes to a research project. Would you agree with that? 

 Erzsebet: Absolutely. And you know, Jo, what I find particularly in this respect is that it said, of course, this is something said easier than done. I think there are crucial twists in research workflows if you are working with sharing in mind. You mentioned interviews. So if you work with human subjects, interview subjects when designing your consent form, you need to have an idea already at that point what you’re going to do with those interior transcripts, whether you want to share them somewhere, whether you need to anonymize them. There is GDPR, but also whether you share them in a repository, whether that repository will be harvested by bigger European discovery frameworks as well, like the European Open Science Club. So there are a whole range of tweaks and twists that you need to think ahead when working with sharing in mind, for instance, at DARIAH, for instance, give a checklist to people working with cultural heritage material before your first visit to the archive, digital or physical. Here are the things that you might want to ask from your archivist in order to be able to share your data five years later, ten years later, two years later, whatever. But this challenge is, as I already mentioned, rarely ever only legal and cultural ones. But there is this frequently voiced concern in open science discourses that open science seems at risk when it relies on closed and proprietary systems. And yet the open infrastructure or contributions to open infrastructures are often based on voluntary work, voluntary labor not properly rewarded. There is no difference in working with cultural heritage in this respect. Just last week I came across very sad news that Flickr and the Internet Archive have deleted a very significant collection of 5.2 million digitized books and book illustrations. Internet archive book images collections. This collection covers some 500 years of print culture, so of course it was used by lots of historians, artists, researchers, and many others who embedded these pictures into their collections. They reused them, they brought them to their own research environment and the like and reached them but added additional description and the information mentioned to them and so without any warning, it just disappeared. Millions of the searchable images just became wiped away. So infrastructure, public ownership and I would say, even possibly control over the infrastructure that is instrumental for our research workflows is another challenge in implementing open science. I would say.

Jo: We really need to build a system. And I think the understanding is there for many stakeholders, especially those of us who consider connected, globally connected, interoperable, sustainable research infrastructure.

It’s a diverse discussion to have. And what I also learned with Africarchive. And naturally infrastructure has been built in silos out of need for specific communities, and we’re not trying to plug them together. And it’s not as easy;  it’s just like our sockets. Like, there’s American sockets, there’s UK sockets, there’s Central European ones. So we need all kinds of adapters. Okay.

Jo: But in terms of sustainability and functionality, how can we preserve scholarly literature, scholarly knowledge, not only literature, but also artifacts and images and the cultural scholarly heritage. So using repositories of some sort and then for licensing or Copyright reasons, like you just said, they might just disappear again overnight. And then the question is who makes those decisions to deprive the researchers and the cultural custodians of the content overnight? 

Erzsebet: Yes, I think the problem regarding who makes these decisions, this is of course, in many cases, and especially coming back to the Internet Archive has to do with negotiations around two conflicting or seemingly conflicting interests, being copyright and protecting copyrights wants copyrights once royalties, whatever associated benefits versus opening up the content for the public. And these might be conflicting. Sometimes there might be. There are certainly conflicting interests mainly. I don’t want to put too much emphasis on this somewhat commonsensical blaming publishers, but in many cases this is indeed the publishers who are pushing for compliance with Copyright minutes and the like and trying to save their own interests. But so in many cases, I think it takes us back to the issue of designing infrastructure, designing resources with sharing in mind. That is, if one knows in advance how exactly they want to share that content, how exactly they want to make it available for a smaller community or a broader one. There are lots of legal, ethical, social, technical elements that have to be sorted out in advance. And Jo, I will come back to your socket metaphor a little bit later. So I will try to keep this in mind, not only because you may know how dear metaphors are to me, but also because DARIAH has to do a lot with socketing, and I will tell you about it in a second. But I think in order to talk about infrastructure and why we need infrastructures, how to optimally design infrastructures in the context of my own daily job at DARIAH, I need to come back also to this basic recognition of open science that we are not producing only papers anymore. So the question is how to sustain these new types of digital scholarly objects, like the software, the annotation layers, whatever we mean by the term data, and all the other resources. So how can we sustain them? Because libraries, even digital libraries, are increasingly challenged by preserving them. And how we can connect them in a publicly maintained environment and in meaningful ways. That’s I think the big question that not only DARIAH, but also other research infrastructure is aiming to answer and aiming to tackle. And just to tell you a bit of history about it, I think it was around the 2010s when institutions on the national level recognized this kind of challenge increasingly like research institutions, and they realize that they need to make collective investments into such services, such research infrastructures that these institutions individually even nationally, could not maintain or could not have maintained alone. And this was a good timing because at the same time, this need for long term planning and making infrastructural investments for research was also recognized by the European Commission. And this gave rise to what we call ERICS, European Research Infrastructure Consortium, which is a special form of establishment form of organization that is organized along disciplines. So all main disciplines have one ERIC   in Europe. And so they connect European research organizations and scholars and scholarly networks on a disciplinary basis to make such collective investments into research infrastructure together. So my organization DARIAH became such an ERIC in 2014. Yes. The rest is history. 

Jo: That’s great to know that there are organizations and people taking charge of that and making sure that there’s an efficient, also financially efficient approach to connecting those silos with each other, to not only preserve scholarly knowledge and cultural heritage, but also make it accessible and reusable, I think, for what you described, like how humanities research items as a collective term are being reused to investigate, to study cultural legacies, 

 Erzsebet: society, culture, art, 

Jo: art, music makes sense.  I feel in the natural Sciences, we’re slowly but truly losing grip. We’re producing data. And who has the time to actually look through all this knowledge production that we’re currently producing? And also what it’s worth unless it’s properly contextualized. Are there some other issues also in the social Sciences and communities? 

Erzsebet: Yeah, absolutely. I think context is a key element here, and also a key organizing principle and added value along connecting these different institutional silos or infrastructural components. Absolutely.

That’s a very important notion for DARIAH as well. And just to tell you an example, one of the latest and very important long awaited services that we have co developed with other organizations and sister research infrastructure is something called the Social Sciences and Humanities Open Cloud. And it’s Open Cloud not only because of its commitment to open science, open humanities, social Sciences and cloud based web services, but also because it’s going to be a very important part of social Sciences and humanities resources within what we call the European Open science cloud, the EOSC. And so it was a question of how to meaningfully bring a whole different research domain of social Sciences and humanities which are internally very complex and very diverse to this big European open science cloud. At DARIAH, we really didn’t want to, don’t get me wrong, it’s not the job of DARIAH to reproduce all the facilities, all the data repositories, all the culture heritage, digitized collections, all the text analysis tools, all the image interoperability frameworks, whatever researchers are working with on the European level, but instead DARIAH’s role is to develop a consolidated view of what is available not only next door on campus, but also on a little bit broader level, what is available across the DARIAH member countries. It’s currently 20 countries and a couple of hundreds of institutions. So lots of lots of treasures and tools, services, training materials, data repositories that might be relevant beyond the institutional, beyond the national scope. And the question was how to meaningfully bring these diverse resources together. And we realized also learning from the mistake of earlier attempts to provide the list. These are the tools you use for text analysis. These are the tools you use for annotation. These are the tools you use for data visualization. That is probably not the best solution to proceed. So instead we decided to go for a kind of marketplace approach. So the output became a discovery service, especially for social science and humanities. This is the social science and humanities open marketplace, where, yes, scholars can look for tools and services and data and publications and training materials according to their research questions, according to their disciplines. But what really glues them together is that we try to embed those instances and connect them in the context of research workflows, publishing procedures, research procedures, step by step. So if I want to do, for instance, comparative analysis of parliamentary corpora, then here is step one. Here are the resources that are first to consult in the data collection phase. Here are the analysis tools. Here you can validate your tagging bla bla bla. In the context of workflows, the connections, the internal assessments, the potentials of these resources, the reusability potentials, I think are becoming much more clear. And as part of this work, they also did the developers of the marketplace a bit of a work on, for instance, extracting references to research tools from publications or data like what we see. And this might be the case for art sciences as well, but maybe not that much as it is the case in the humanities that providing proper machine readable, recognizable citations for research tools, software, data sets, it’s not yet a community practice. So the developers of this marketplace did a little bit of a work on automatically extracting and cross referencing tools and papers to see not only what is available, but also how those resources are used in actual research. 

Jo: Thanks for this overview. It’s really fascinating and also promising. And also there’s a call for us to foster more holistic culture for the researchers’ awareness to think about archiving, to think about creation of the data. And I think various stakeholders can also assist the researchers to do that better than what has historically grown as a practice.

So explicitly, this is not to blame anyone or ourselves and how we do our research as PhD students or whatever. But yeah, it’s a matter of being informed and being aware of how to best contextualize like to add as much context to any data points via the manuscript or an actual data set interview like it needs to be contextualized, irrespective of the discipline or the research team, and then to decide what infrastructure, what accessible or available infrastructure to utilize for long term storage, and also to provide accessibility to the knowledge produced. These are questions that I’m also learning by the day. At the moment you think, oh no, I have a pretty good overview and then another thing comes around. A whole new level of complexity. But it’s also fascinating and I feel it’s a little bit like being a researcher. You learn more and then you realize there’s still more to explore and to know and to be aware of. But we look at it pragmatically. And if we want to support the researchers, how can service providers like myself, trainers and also institutions like DARIAH, what are you doing in making the researchers’ lives easy and also efficient for data or research knowledge, accessibility and storage? 

 Erzsebet: Okay, that’s a very good question. Beautiful one. And I can resonate with your experience, Jo, because when I started acquainting myself with the practices of open access, open science, open research culture, it seemed so easy, even sometimes in the context of my own research. And the more experience I gained providing support, the more I think the opposite, actually. And I can only repeat myself saying that it’s epistemic. At DARIAH, I’m super lucky because if you’re a linguist, you’re working in a certain field, a certain subfield of that subfield of that subfield, and by the end of your career you might become competent in that subfield or even field. But at DARIAH, since we are serving all the arts and humanities disciplines, so I’m looking for richness here, I have the chance to work together with very diverse communities who are coming with very diverse needs, workflows that are completely digital and large scale and computational versus almost fully analog workflows, dissertations, writing projects and anything and everything in between in terms of scale. So there’s a huge diversity here and I’m learning a lot. And in terms of how best to serve scholars, it’s super important to stay in the ground of research realities. I think if you imagine, DARIAH is a

European organization who provides all kinds of research and teaching support for the arts and humanities disciplines and connects several hundreds of scholars and research facilities and tools in 20 European countries and cooperating partners now. I think facilitating open access to resources has always been in the DNA of DARIAH to make scholars’ lives a little bit easier. So in this photo, if you think of open science as a kind of engine of scholarly innovation, then the synergies with DARIAH are quite obvious. Open science had been implicitly there from the beginning. But at some point, especially upon watching this epistemic differences around open access open science and how much the whole paradigm had been dominated by heart Sciences, life Sciences, biomedical Sciences, I think at some point it became important for DARIAH as well to go explicit and to build a no corner agenda for arts and humanities scholars that is firmly grounded in disciplinary realities. And I became the one who, as DARIAH’s open science officer, it was part of my job to support the implementation of such a strategy. So what do we do in practical terms with explicit open science alongside the implicit ones, connecting services, connecting networks, connecting scholars with similar interests? So we found it very important to strengthen the discourse around community practices in open humanities and talk about disciplinary challenges, talk about shared needs, talk about success stories, disaster stories, people who publish their books open access and still survived, people who shared their data and they made a difference in their disciplines in this respect. People who implemented open peer review within their own disciplinary context conferences. So it’s important to foster the discourse around these issues and see how we can bridge open science and humanities disciplinary reality. So we have this blog called DARIAH open, it is a dedicated discourse for that. And what is also, I think, super important is that if the idea is to stay really grounded in disciplinary realities, then it’s also obvious that this is a kind of work that we cannot do alone. And luckily we are not alone in this. What I’m saying is that it’s super important to have diverse communities around DARIAH, not beyond the National DARIAH to have communities who contribute to and enrich and validate our work and spread the word about it. And one of these very precious communities whom we have at DARIAH is around the Open Methods platform. It’s a platform that brings together open access content, all kinds of content about digital humanities tools and methods specifically. But it can be a video tutorial, it can be a blog post, it can be a preprint. Of course, it can be a peer reviewed article as well. It can be a podcast, for that matter, anything, the media doesn’t make any difference. We want to include content that is also out of the site for this formal, established research assessment. And we provide a kind of little discussion around digital humanities tools and methods in multiple languages, because multilingualism is an important value in this research. So not everything that counts, not all the knowledge that counts has been written in English, right? 

Jo: Those are our favorite topic and also painfully underserved, under communicated in the scholarly community as a whole. But luckily it’s good to see that there is now an increasing number of initiatives popping up, and it was very much uplifting to see that DARIAH is doing so much work or doing quite some effort in fostering multilingualism, naturally, because it’s serving several countries in Europe each coming with their own language and cultural history.   Erzsebet: Absolutely. I will tell you about it in a bit. Yes. But coming back to this open methods platform/community, imagine a 29 digital humanities experts performing from all around the globe, from Argentina, from Eastern Europe, US, Western Europe, who are coming from twelve countries and speaking together 20 languages and are able to read and curate content in 20 languages. It’s pretty incredible and something really precious, I would say. I really enjoy working with them. But in addition to the open methods team, later we also realized speaking of data sharing, as I think also in humanities, the discourse on open access became much more established than other elements of the open research culture. There was some discussion on open peer review, but the discussions around data sharing and open and FAIR data came a little bit later. And I honestly think that one of the reasons for that is because these are extremely complex phenomena. So even like ten times more complex than publication of papers. And back then I also realized that if there is a mandate for FAIR data, for open data, then DARIAH has to put in infrastructure and procedures, yes, but also support and training and accumulated knowledge in place to help ease the compliance for scholars who we are serving. And so in 2020, we started a working group for research data management in the humanities because constantly we saw that there are lots of discussions on the e-level about research data management and support and how to do that right. And how to write a data management plan and things like this. But our impression was that the generic discipline agnostic discussions around these topics will not necessarily serve value humanists, not because they’re so special or not, because humanities is the white child of disciplines, there is no such thing, but certain specificity that we already discussed, like being exposed to cultural heritage data and the like has to be taken into account. So we wanted to bring together experience and consolidate experience from different disciplines. So we have disciplinary champions in this working group whom we know, who we trust, who are doing an epic good work as scholars. They are not open science advocates, they are real researchers, like ten years or early career or in different situations. But what is common in them is that they are doing a really good job with working together, sharing their resources, working with sharing in mind. And later we also see that in addition to them, in addition to these disciplinary champions of representatives, there is a new type of professional role emerging across Europe who I would call data support professionals. They are sometimes called data stewards or open science officers, or here in Germany, in many cases they are called subject librarians because they are indeed librarians by training. And what is common in them is that they’re having brand new professional roles of supporting scholars in research data management. And we realized at training events organized by DARIAH around open humanities and open data and FAIR data is that there is a need for such professionals who are working in domain specific humanities specific contexts to meet each other and to exchange experience because everything is really new in this professional rules, the institutions mandates that are changing a lot establishing these support institutions. So we use this working group also for them to meet and network and exchange and consolidate good practices. And I’m Super excited that we are going to have our first writing sprint in June this year in Warsaw, and it’s going to be the first time that we meet members of the group face to face. 

Jo: That’s so nice. So you mentioned the event is going to be in Warsaw and DARIAH is serving 20. How many countries? 

 Erzsebet: Yeah, 20 countries and a couple of cooperating partners who are not full members. But in terms of members, we have 20 countries and one observer member which is Switzerland. 

Jo: So what’s your experience in terms of how open science can bring Europe together? Also because what’s visible in databases is mostly English literature, in terms of manuscripts and also the high capacity, I would suggest research institutions produce just more numbers because they have more resources to produce stuff, I mean scholarly output.

So how is open science helping to bring about visibility also to Eastern and Southern Europe. Eastern and Southern European countries are the smaller countries which are actually bigger in size in Scandinavia. How does open science serve cultural diversity in terms of scholarly culture and cultural heritage and perhaps also then towards multilingualism? But these are basically two or three questions. The first question would be, before we come to talk about multilingualism; is Eastern Europe now, for you also being Hungarian by cultural identity, do you see Eastern Europe now better represented in the scholarly system than before the boost and the move towards open science, or are we still struggling to see diversity? 

 Erzsebet: That’s a very complex question. And I would also add a very spot on one. Yes, I think we can come back to this basic principle that I keep repeating. That is one of the biggest challenges in implementing open science. I’m talking about practices, but I’m talking about policies as well, because DARIAH is also involved, like DARIAH  is representing arts and humanities in policy conversations, European policy conversations about the future of research funding, the future of open science, the future of FAIR data. And what is really difficult is to avoid or resist these simplistic blanket solutions and still find a kind of a functional common ground that is more or less working or flexible enough to make it work in this very diverse disciplinary environment, but also very diverse geographical environment. And we can see this at DARIAH very clearly when we have a look at the diversity within the national DARIAH and for DARIAH, it’s structurally important. We will never become an organized, strong, top down organized institution because we are aware that arts and humanities research itself is very much embedded in local skills, in local languages, in local contexts. And therefore we need to keep, if we do want to keep this diversity, which comes with a kind of sort of autonomy . But when it comes to open science, part of my job at DARIAH is to provide open science training, online training, offline training around topics that are requested by the DARIAH member institutions. And I have to confess that these are very different conversations about open science. There is, on the one hand, the decree of awareness. And I come back to Tony’s paper on open science which if done wrong can fuel or conservate existing power structures in academia. I see this in many cases, researchers coming from Germany, coming from Denmark, coming from Switzerland. For them, open science, open access, in many cases, is a no brainer because they already have funder mandates who come with strong open access publication imperatives. But even more importantly to that, they come with institutional resources to make those open access circumstance mandates happen. So they have the policies, they have the knowledge, they have the resources in many cases. But still, we shouldn’t think that diversity is only limited to Eastern, Western or South Northern angles, because what I routinely observe in my daily job is that, for instance, what works for France, the top strategy that works for France will never ever work in Germany because one country is strongly centralized, the other one is the opposite. It’s a Federation of regions, out and on regions. So it also has its impact on how infrastructures are connected, how standards are negotiated.

Jo: I never saw it on that level. Thanks for bringing that up.

 Erzsebet: But speaking probably, yeah, the variable that I’m the most sensitive towards, you are right as a Hungarian scholar is the Eastern – Western one. So we are having very different conversations. And what I really love about the composition of DARIAH is that it’s relatively balanced in terms of membership. So as I mentioned, I really love working together. For instance, my Polish colleagues, Martha Blasmitska, who is my co chair at my Research Data Management working group, our research data management working group, and Machin Maria, and one of the shared barriers that we repeatedly overcome. And this will take us to the need to overcome and this will take us to the issue of multilingualism as we need to be aware of certain disparities when it comes to digitization, when it comes to resources, to digitize, resources to upscale languages. And there is a strong bias towards what is digitally openly available. And we need to keep in mind that institutions across Europe have varying capacities to invest in debt. So we are talking about serious disparities when it comes to open access to data, when it comes to upscaling tools, for instance, NLP tools to, I would say lesser resourced languages. So it’s part of our collective mission at DARIAH, but also beyond DARIAH when it comes to digital infrastructures to raise awareness of such disparities. And even if we won’t have capacity to digitize everything, at least on the metadata level, I would say we should put all these digitally invisible resources to a map of digital steel. Just pin them. Even if we cannot bring the content itself fully digital, fully available for computational analysis, we need to just pin them down, like making awareness to others that there is stuff there sitting in culture heritage institutions in this respect, there is a huge bias towards English language and global topics and the like. 

Jo: Or what’s supposedly global but often very much Western, Northern Americans and Western Europeans. 

 Erzsebet: Yes, absolutely. You know, it takes us to the issue of research reports. But before discussing multilingualism in the context of publications, I also want to highlight the problem manifesting itself probably even more strongly on the level of content and procedures that are beyond the research paper. So just to give you an example, there is a very exciting initiative in which DARIAH is collaborating with Princeton University, and it’s about training humanities scholars in the development of natural language processing tools to optimize them for lesser resource languages. So to compensate for this kind of disparities in terms of digitization that I mentioned to you, because most of these text analysis tools are, by the way, making wonderful and really wonderful contributions from humanists to the broader scientific public. They are trained on data. They are trained on training corpora available in English, right? Or French. Spanish in many cases. But how about Hungarian? Yes. How about certain Chinese dialects? How about classic Arabic? How about Yoruba? How about African languages? So we have this collaboration with Princeton or my colleagues, especially Toma Tashawatch, one of the directors of DARIAH, and they are doing incredibly valuable but super difficult work in upscaling and optimizing these NLP tools for particular lesser resource languages or language varieties. And this is such a merciless task, I would say, because if you think about it, people, the researchers who are working in this collaboration, I think most of them in their institutions are rewarded for writing books, writing research papers, teaching, attracting research grants, but by no means working on or upscaling NLP pipelines. This is completely out of the site for research evaluation, I would say. 

 Jo: Yeah, it’s lovely to hear about this initiative. And with Africarchive, we have a similar one, similar but yet different. We’re also digitizing with an organization called Masakana, which is a pan African data science multilingual NLP organization. The thing is, as you pointed out, most there’s around 2500 recognized African languages on the continent, whatever is considered recognized and then dialect. Well, kind of. And our project is to translate English articles with my African scholars into African languages and not the ones that are already pretty well covered by machine translation, like Google translation, but those that we’re actually building a corpus of scientific terms across disciplines. So we’ve categorized the disciplines that we want to cover a certain level of understanding for undergraduate scholars also. Not to make it too nerdy, basically not too nerdy science, but the research output that’s comprehensible by normal people, I think that is regional and local languages. And these are also still big languages, but the coverage in digital form is pretty much nonexistent. And many African languages are also historically on the aura, so there’s not much written archives on it or studies on the languages. So basically the idea is to enrich the African languages with these new scientific terms and thereby foster the use of African languages also for science communication. So that’s the idea. So we’re addressing several layers of decolonization or colonial heritage if you want. And it’s also called decolonized science. And it’s much harder. We thought it would be difficult, but it’s actually more difficult. So first we were hopeful that we would just translate the articles as they are, but it turned out it’s too tricky, too many details, too many kinds of specific terms. So we’re now creating lay summaries. And when I say we, that’s actually science. So we’re working between Africarchive, Makasana and Science Link, which is a South African science literacy organization or science journalist, and then also SE communications, they are specialized translations into African languages. So it’s a beautiful project. And I feel that we should also talk off record more about maybe how we can bring initiatives together because there might be some room for interactions and cooperation maybe in the future or as they are ongoing, exciting. Okay. So basically  as we raise awareness about the need for cultural specificity multilingualism, the importance of language when it comes to research. Like my argument, at some point it struck me when I realized whatever discipline you’re studying, if we are being forced to only use English and even for our native English speakers, technical English, that’s often assumed that many scholars nowadays have to use technical English. That sounds really scientific, but no, because that’s when we strip off the cultural context of the main subject. I feel everything like knowledge comes with culture and cultural experience. And if we only use technical language, then we might look at facts, but they are of no use really, unless you contextualize them also culturally and especially also where I come from, Biosciences in ecology; what’s the point? I’m talking about ecosystems and species and all of that. If you don’t use terms that are culturally coined also. And that makes sense for the very local ecosystem. What I’m trying to say here, I think it’s important that researchers are allowed to conduct research in their own language. And yes, we also need lingua franca. We actually do have multilingual franca currently and there’s only talks for about an hour. Well, maybe I’m in a very big bubble here. And another World region talks about Spanish, but there’s different realities in different places of the Earth. But we assume that the lingua franca being English, and that might be the major bulk of the research output, but it’s certainly not the only back of the research. So there’s more backs in Mandarin and Spanish, Portuguese, and a lot of the local research across Africa is being produced and disseminated in African languages and it’s never being picked up by any Western database. Sorry, I just wanted to finish this. All we see in Google scholar or whatever aggregate we look at is what we know about the world. It’s not. We know much more. You beautifully pointed that out previously.

Because we had several conversations about multilingualism in the past. Shall we create a vision for how we see a multilingual scholarly system very soon, like two, five years from now? What would a perfect, multilingually inclusive, globally connected scholarly communication system look like? 

 Erzsebet: That’s a beautiful question, Jo. Thanks for that. What is clear to me is that it’s necessarily a decentralized one and a multi layered one, by which I mean that as you mentioned, there are serious infrastructural prerequisites of that. Like there is this important study by my Polish colleague of mine, Emmanuel Kolchitsky, who shows how much Eastern European countries and other countries as well, how much native locally relevant local publications in local languages are not indexed in international databases like Web of Science or scopus. So there are serious infrastructural prerequisites to that. And infrastructures like the Ariane system, infrastructures like Oprah’s are doing very important work in this respect to expand the inclusiveness of the scholarly information management systems and discovery environments. So this is one thing, but that’s about infrastructure where metadata standards are talking to each other and information management systems are talking to each other in multiple languages. And that’s of course investment. But money well spent, I would say. But there is also this in addition to the infrastructure layer, there is also this social layer. So we still have a lot of work to do with the Anecdotal reviewer too, who says they make comments like, ‘I suggest they should be revised by a native English Speaker because the language register of the publication is not formal enough.’

Jo:  Discussion topic on its own. 

 Erzsebet: Yes

Jo: Or maybe not. Let’s not discuss that. 

 Erzsebet: It’s not even getting to this road. That’s the social layer. Also, I hear a lot from colleagues, from ex-course mates I’m working with, that if they work with their locally relevant topics, local data and local languages within Europe; so I’m not talking about big cross cultural investigations here. Data from the Czech republic, data from Poland, from Hungary. In many cases they feel the need to justify in publications why this is relevant by contrast to working with the American or English equivalent. So that’s also the social cultural layer  of something to be changed for the better and in the middle or on top of the social and infrastructural components to fix. I would also add the dimension of recognition and research reports. So as long as we are rewarded along with bibliometrics, citation counts and the like, I think it remains almost inevitable that research will be skewed or even distorted towards the big topics, towards the big languages, towards the big publishers. So of course, for a Hungarian scholar to publish something of good quality in English, that’s a lot of added effort and knowledge and value. But how about multilingual publications? How about the effort of translating back, bringing knowledge back to your local communities who raised you as a scholar? So I think there is a whole lot of work to do on the level of policies, on the level of research assessment as well, because currently multilingualism seems to be quite an overlooked dimension. 

Jo: Also liberating funds, I think, which would come through the policy change. There’s a whole profession, translators and interpreters, which are needed. And it’s basically like science journalism. We need translators who have a certain understanding, or at least interest in scientific topics of any kind, to make sure that they capture the research aspects. Also not just translating words, but actually having an understanding of what’s being written about, and also from the scientists to make an effort to keep their work and output comprehensible. There’s nothing fancy about using technical terms or overly technical lingo. It just makes everybody’s life more difficult in scientific writing courses.

Yeah, that’s really a beautiful vision to have. Also a lot of work still. But I think we’re in a good way. We can really do this to provide for culturally sensitive and inclusive, multilingual scholarly communication that is totally possible. And we know where to tighten the screws, where to add some wood, where to plant new trees. I started the woods paraphrasing, so I wanted to come back to an ecologically sustainable image as well. But yeah, it’s totally doable. There’s a lot of work underway already to make scholarly communication properly, globally inclusive. And we’ll keep pushing; so some final remarks just to have a roundup for this episode. And let me just say, thanks also so much for spending the time and sharing all your wisdom with us and some insights into the highly important, highly needed and also highly rewarding, I assume work that you do. 

 Erzsebet: It is rewarding for sure. And in many cases, there are times where I just realized there is this common basic idea that we are operating in our own bubbles and the like. I think this is a super important future priority for open science, to support people, for advocates, to break out of their own bubbles from time to time, to really reach the realities, to reach the actual practitioners, the actual scholars who are having a busy life, who are publish or perish, who are having families, who are maneuvering their way to tenure and institutional realities and funding and the next career move and staying in academia, leaving academia. So this really happens. What I’m trying to say is that I feel incredibly lucky for having the chance to work on these things, trying to be useful for them. And this necessarily implies leaving the comfort zone bubbles for them as well. For us as well. But I’m also super grateful for all the efforts that we mentioned already, all the fellow colleagues, all the fellow dreamers who are indeed meaningfully fighting for a more inclusive and multilingual scholarly ecosystem, I would say. And they are doing the work, doing the right things, and doing things right. I think so. I’m super grateful for forming intellectual colleague communities with them. And of course, including you as well, Jo. Thank you for sharing your insights from time to time. 

Jo: Yeah, we learn and grow with each other. And this is also what I love about this podcast format and the format is to have a conversation, to create visions and to dream a little bit and then see what is possible. And yeah, just to keep moving and keep building a better world. I think we have responsibilities to leave the world a really better place. Previous generations, not saying they messed up, but they were kind of okay, but now we’re not okay in this world.

But also I believe that research and science and interdisciplinary exchange and conversations like these can really allow us to utilize research output to solve most, if not all of the issues that we’re currently experiencing. And the world will always struggle. There’s also the fight for survival. And nature has always been and will always be; but just to sustain a livelihood for I don’t know, I’m a biologist, so I’m looking at it from that way. But also now in terms of war, Ukraine, Russia being isolated, there’s many other countries that are conflict ridden. I think as humans, we know better and we can do better and we will do better. So, okay, I was trying to end on a positive note. Oh, yeah, hopeful. Let’s be hopeful. And we know we’re doing the right thing. And it’s great to see others are on the same and very similar journeys. And as long as we keep talking to each other, I think there is also good reason for hope and progress.

 Erzsebet: Yes, it has to do a lot. I think on a small scale, it’s easy to imagine this whole open access, open science thing is opening up these bigger, smaller black boxes like opening windows and research processes, opening windows on what’s going on, opening windows on power relations as well. And I think that the podcast format is a really good and pleasant opportunity to open windows on not necessarily black boxes, but boxes that one would not necessarily have at hand. So thanks a lot for having me and boxing with me a little bit.

Jo: Great pleasure. Yeah. See you soon again. Let’s also see in the real world because we live in Berlin and spring is coming. So let’s go for it.

 Erzsebet: Hopefully.

Jo:  Thank you  Erzsebet.

TRANSCRIPT