Interviewer: Thanks again for agreeing to take part in this interview. And you've already signed the form of consent. But as a reminder, this interview is being recorded. And the question would be whether you agree to that and also to it being transcribed by myself and my student research assistant with the help of a local large language model.

Interviewee 04: Yeah, that's fine.

Interviewer: Great.

Interviewee 04: I was just wondering, I don't know what that's good to ask right now, but what are you gonna use for transcription?

Interviewer: I haven't yet decided, to be honest.

Interviewee 04: Okay, cause.

Interviewer: So, yeah, have you got a recommendation?

Interviewee 04: I'd just go for WhisperX. I don't know, because that's completely, like, data protection-wise fine and free and usually really good. So it works for so many varieties. It's kind of I hope you don't mind I don't know.

Interviewer: Um yeah there's some unexpected setbacks. I mean it's fine. I'm happy for people to just speak in whatever language they prefer but and past interviews have been a bit of a mixture of german and english. So that's gonna be fun.

Interviewee 04: Okay yeah but I think it should also kind of like sometimes it works very well as well also with like translanguaging. So to say

Interviewer: Oh wow

Interviewee 04: Um yeah The only thing is really that it's not so good in like speaker recognition but I don't know whether the zoom meeting recording is. That would be nice.

Interviewer: Oh I mean, this is something actually I was going to look into this morning. Uh so I'll do that after this interview and there is a possibility of recording two um separate. And that I think I might have to do that

Interviewee 04: Okay

Interviewer: Because it's also detected issues. So uh yeah it's all very experimental

Interviewee 04: Okay yeah

Interviewer: Um but yeah maybe I have to use a second tool for speaker recognition.

Interviewee 04: Yeah I do like you know if there's anything that what where I can help there which is always very fascinating when it comes to speech recognition, so

Interviewer: Okay good to know I may we will come back to that because it's all very new to me. I did some experimenting with french um but not much so cool. And Yeah, as you know, this is about open science practices in linguistics. And so the first question is, would you say that you're a linguist? And if so, in which sub-disciplines do you situate your research?

Interviewee 04: Yes, I think I'm a linguist. And my sub-disciplines or my background would be sociophonetics, phonetics and phonology, PROJECT English in particular. Um generally also corpus linguistics and definitely computational linguistics. Yeah and I want to get a little bit a little bit more into like language assessment. That's at least kind of my way to go but I think that's my background here yeah.

Interviewer: Yeah, quite a few fields. That's really interesting. Yeah, we'll begin with your sort of personal take on open science. And so the first question would be simply, what do you associate with open science and what are open science practices to you?

Interviewee 04: So for me, it would be kind of it's basically about transparency in the research process. And accessibility as well. So I think generally, like Open Science describes that if you conduct a study, for instance, that you basically not only provide the kind of final written article, but also the data set, and the methodology in detail, like and how you went about kind of getting the findings there. Um yeah and maybe also that was also very often very important to me in terms of methodology like what particular systems for example to use for example for you know speech transcription for I don't know um pos-tagging or um stuff like that because that is usually I think very helpful. Um also to other you know scientists or linguists there. Um because you can definitely learn a lot from other people there. And I think, yeah, accessibility as well. So generally transparency about the whole study itself. But then accessibility is, of course, another big thing, especially if you're a PhD student or if you're at a university with not so many resources. It's, of course, always very nice if you've just got really like the outputs like online or you can just access them and there's no paywall. So that's, of course, very beneficial as well. Yeah.

Interviewer: Yeah, linguistics is traditionally seen as a humanities and in the humanities, some people prefer to speak of open research instead of open science because it's not a science, it's a humanity. And then there are others who prefer the term open scholarship, which is often meant in a broader sense to include open science slash open research and open education. So there's a lot of terms which I'm going to put in a chat in a second so you have an eye on them. My question would be, what are your thoughts on this? Which term do you think is most suitable for linguistics? Is this something you've thought about at all, actually?

Interviewee 04: So in detail, definitely no. I think I'd probably go for open research. Cause for like open scholarship. I don't have that many associations I don't know. That also entails maybe uh that you just really put yourself out there so that you do have something like a very detailed personal website. Uh that you basically describe what projects you're currently working on and stuff like that. Um open education I'd say is more like possibly in language learning domains. But I think for linguistics generally, I think it's more like open research. But then like your research is accessible, transparent, reviewed. Yeah. And open science maybe as an umbrella term.

Interviewer: Okay. Yeah, I mean, I will continue to say open science in the interview, but you imagine it as being very broad. I also see it as an umbrella term, and you can use whichever term you know you prefer. And now I'd like to speak about your own experiences. Do you take part in any open science practices? And if so, which ones and why?

Interviewee 04: Yeah, so basically when it comes to publishing, of course, it's always the best way forward. If you've got like, you know, fully accessible, for example, journal articles. So that's possible in some journals, but not so much possible in others. What I did more in the past, but I should kind of work on again, is just, for example, publishing more of my, I don't know, presentation slides. This is something that I did very, very openly in the past, but it just basically I don't know if it's laziness or if it's just, I don't know. So that's basically something there as well. Something else is definitely scripts. So Python scripts, that's something that I generally put out there, which is hopefully also helping somebody at some point. Yeah, but that's basically the case. I wouldn't say I'm the greatest when it comes to open science. Um but I just know like for myself it helped a lot um especially also in my phd thesis um I don't know. And then it's also very nice like in the community for for example I think there were some I don't know Praat scripts by PERSON. Like scholar in the U.S. And just in general I probably would have never had the chance to kind of you know meet this person. But then later in the conference um he was actually there and then he could already talk about something and you already know. Okay that person kind of had the super nice tutorial about uh I think it was. He's got a couple of tutorials about Praat and stuff like that. Or um PERSON, who's currently, I think, in CITY. Oh, can I actually say personal names here? Yeah.

Interviewer: Especially if you mention them in positive, I think. But otherwise, I will anonymize.

Interviewee 04: Yeah. But there are many examples, basically, you included as well, where you can kind of access that stuff in a very nice way. Yeah. I think I did some stuff there in the past, but I definitely had to catch up that way. And it's also, of course, I think something like a bigger question that we also maybe discussed before in CITY at the conference there, that some people are very reluctant and maybe even opposing to really opening publishing or publishing completely openly. In an open form their research. Sorry publishing their research in a very open way. Um it's probably just a very different perspective on research itself and I think especially in the humanities also when it comes to that at least my kind of take on it like educational sciences in COUNTRY Bildungswissenschaften uh there is sometimes great resistance as well. Um yeah

Interviewer: And why might that be? I mean, it's an applied science, isn't it?

Interviewee 04: Yeah. I don't know. I think possibly one thing that I could imagine or that I could see is that those people who work more like quantitatively or empirically I think they are more open science practices than those who maybe work more qualitatively. I mean, some other things are probably that people are afraid to put like everything out there because then you can't publish it maybe and you know particular application domains or journals. So that their research is stolen, maybe in some way or I don't know, because maybe it's also just not really common practice and people don't see the point. And it's, of course, also a lot more work if you really have to, like, put or make everything transparent, put all the scripts online, produce slides online, make sure that they're nice. There's, of course, and that's also something that I, myself, of course, you're a little bit afraid if, I don't know, super, like, statistics nerds kind of sees your code or your statistics and then says, ah, that's completely that's completely wrong. And then you're like, ah, right. So I don't know, maybe a fear of embarrassment as well could be a reason. But I think especially like if your research is more quantitative than qualitative, I think the quantitative people are probably a lot more open to open science practices than qualitative people. I don't know.

Interviewer: Yeah. And I mean, you definitely do practice various aspects of open science. And where did you learn about these practices? Or were you encouraged by specific people or systems?

Interviewee 04: I think it was, of course, basically during my PhD. So I benefited a lot from open science practices by other people in my PhD domain. So especially PERSON and the people in CITY. So those people kind of working on PROJECT English, they were very open itself. They were also like you had to talk to them, of course, as well. But they were then very open and also gave the data set and stuff like that, which is, of course, nice. But I think it was just also when I kind of dive deeper into programming and like computational linguistics, there are just many things that are kind of completely accessible online which was very nice and that's where I benefited a lot and that's where I could also see okay this is actually a very great way forward because it's just very it's just very convenient. It helps the whole community yeah but it wasn't like that I had a particular person kind of pinching me towards like open science. It's just. I think it's just a very good way forward generally. Also in the current debate about like, you know, scepticism about research itself. I think open science might be a very good way forward there.

Interviewer: So it wasn't a requirement or anything like that for your PhD? You didn't get bonus points?

Interviewee 04: No, I mean, another thing, maybe just one example would be I published my PhD thesis online, first of all. Fingers crossed it's going to be published as a proper monograph in December this year. But basically, it just made sense. There was also another solution because you have to, in order to get the title, you do have different options usually. And another option would have been to print it out 80 times and give it to the local library stuff like that which is just completely weird. Yeah um and then at the end of the day like I just know. Okay my PhD thesis is out there and it's already being used. Because it's out there. You can find it easily via google as well if you just you know look for it um and it's just a lot more convenient and I don't know. Of course it's also like free so the publication at least with the university library was completely free. It took I think maybe one week if at all. Uh and you know it's easy as well.

Interviewer: Yeah and accessible um yeah um really interesting. We're gonna move away from your personal experiences and and associations with open science and try and think about the broader field of linguistics and of course you're welcome to think about the domains that you're most familiar with. Um as far as you can tell how widespread are open science practices in linguistics?

Interviewee 04: Well, I think there are some disciplines where it's a lot more common practice. I think computational linguistics in particular. If you think about Interspeech, well, if you think about like more the computational side, if you think about Interspeech, for example, if you've got the abstracts there, which are like little research papers after all, they're all publicly available. And you will find, I don't know, if you want to do another forced aligner, you will find all this stuff online. So that's easily accessible. It's also you've got whole repositories online. You've got a whole websites and blogs basically describing what's going on and how it can use these things. So that's definitely, I think strong there, but I'm not so sure about other domains. So I'm not so sure about language assessment, for example. So that's, even though you've got these journals, of course, you first of all need to have access to them. But I think that's not as open overall. Yeah also when it comes to definitely language assessment is a little bit secretive. There are reasons sometimes for it, but sometimes it's just very unfortunate, I think, that For example, there's not so much of a great study on, I don't know, the competences of English language students in COUNTRY overall. Nobody really knows what's going on there. Yeah, so just one example there. I think phonetics overall is okay. Sociolinguistics is okay as well, I'd say. I think the more computational it gets, the more open science it will be. I don't know. That's like my very uninformed, possibly not very thought through opinion on that. And the more qualitative it gets, it's probably less open science. That's my take on it. I don't know.

Interviewer: And thinking about the open science practices that are commonly practiced in these subfields, at least, you've mentioned open code, open data, open access, I think, as well, right? Are those the main practices that you think are commonly practiced, or are there any others?

Interviewee 04: I mean, open publications, first of all, as well. The other thing is maybe just research transfer as well. So I think very often, especially also linguists kind of stay in a little academic bubble in their university buildings and are not very open, like outgoing to really describe, okay, why are we doing this and why is this important? Of course, usually we work on very nitty gritty, detailed topics. But I think we've got a lot more potential in kind of making our research more accessible to the public as well. But also lay people can more like understand this generally what we are doing. That's maybe another thing.

Interviewer: And if we think especially of those maybe less computationally involved aspects of linguistics where open science practices are not so commonly widespread, what do you think could be done to raise awareness and to increase the uptake?

Interviewee 04: I mean, possibly one way forward would be, even though there will be resistance, is to just kind of make people aware that the impact and the outreach will be higher if you publish and like or integrate more open science practices. Because especially like I don't know I'm currently at a more like INSTITUTION university and there aren't just many journals which I cannot possibly like properly access and if there's a very specific publication about something that we do not have. This for example which is probably also the situation for I don't know many scholars in like third world countries. But um it's it's not gonna like you can't access it so it's it's not helpful for others. You're not going to be cited. Um and I don't know. It's a little bit like kind of publishing a book and then just putting it in a drawer and it's there and you can maybe put it on your cv. But um yeah I think in open science it would be I don't know. I think it can be helpful to more people if you integrate more open science practices which also doesn't mean that you do not have to be completely transparent about everything but it's just nicer also to possibly make new contacts. I'd say yeah

Interviewer: So like thinking about the benefits rather than just the. It's more work.

Interviewee 04: Yeah sorry yeah Yeah, I also think it's not really a good example in this context, but I sometimes have colleagues and they, for example, say when it comes to, it's not really related, but let's say working with large language models. And they tell their students that they should, of course, not be allowed to put anything of their publications into these large language models. And then I'm thinking like, yeah, well, okay. But if you think about the big tech companies, they're probably going to do it anyways. So it's like, I don't know, sometimes maybe also a very old-fashioned approach to applications and stuff. I don't know. Yeah. Or, I don't know, resistance where it's not really necessary in a way because there might be other things going on. I don't know.

Interviewer: Do you see a sort of generation divide within linguistics? Because you spoke about quantitative and qualitative. Is there, in your opinion, also sort of younger and older linguists?

Interviewee 04: Like my personal very subjective perception would be that I think younger scholars are more in the open science domain than older scholars. But I think possibly that has to do with many different things that have happened. Also, if you think about like how academic publishing kind of changed even in the last like decades, so to say, that's probably just the result of that, which does not mean that like every like older, so to say, scholar in linguistics is not open science. It's very individual as well. But I think there is a general trend that like the younger, the more open people, or the more open sciences practices are there.

Interviewer: And do you think there are any specificities of linguistics that need to be taken into consideration when we try and implement open science practices? Or would you say, no, these are practices that apply to all sciences, including linguistics?

Interviewee 04: I mean, in general, I think it applies to linguistics. Pretty much all sciences in general. The only thing is of course if you for example I don't know conduct a sociolinguistic interview also you have to be very careful when it comes to anonymization. Yeah um which doesn't mean that you cannot publish stuff openly but it's of course a little bit of a trade-off between like being completely like open but then of course um making sure that you have data protection in place. That's maybe one specific thing. But then again, there are other ways. If you think about the PROJECT project, that's kind of a project with people in CITY, COUNTRY. I think they had a way around this by, for example, they focused especially on phonetic stuff. And that basically could share, because it was like back in the day, it's European Union, then UK, then US, Canada. They basically shared all measurements. So not all of the recordings, so to say, but all transcripts and especially the measurements of, for example, different vowels, like in big data sets and stuff like that. And that is still something that we can publish openly and that we can use. Yeah.

Interviewer: Yeah, yeah, yeah. Awareness of that. Very interesting. And yeah, what would you need to be able to do? More open science or open research, open education?

Interviewee 04: Possibly a bit more time. That would be one thing. Yeah. Another thing is, of course, if you do, if you've got like openly accessible publications for some publishing houses, they can be expensive. That's, of course, something to think about so that there is funding. But at the end of the day, it's not really difficult to do open science practices because you've got so many platforms we can actually do this. You've got github you've got osf. There are many different ways and there are some very good examples. We're just doing this and it doesn't necessarily mean that it has to be expensive that it has to be difficult. It's just also a little bit of self-discipline maybe quality control yeah and maybe losing the fear of that. Somebody could call you out on something which is difficult. Yeah um but there are not. I think there are not many arguments against open science practices I think but you know

Interviewer: I mean, you and I are both convinced that it's a good thing. And my fear sometimes is that advocates of open science tend to speak among themselves quite a lot. And so I feel like I'm only ever preaching to the choir. Do you have any suggestions? I mean, do you feel that that's just the case? Maybe that's not the feeling you have. And if so, yeah, what could be done?

Interviewee 04: It's difficult. I don't know. Yeah. It's possibly, possibly those people who are kind of more convinced of open science would speak together and they're kind of one bubble, so to say, and the others are kind of not in the bubble. It's difficult. I think like one way forward would be that we do have publication formats where you do have like open access publications, which are overall not too difficult financially because there are definitely those who are not in the bubble, so to say. And that's something, the discussions that we are having right now is, of course, if you need, like, funding for open access research, it's kind of the reaction, of course, that why should we publish, or why should we pay publishing houses for publishing our research there? That doesn't, that's, what's going on, right? So, like, other, like, especially when it comes to publishing if there are ways or if there are some publishing houses where you can have open access but it doesn't involve a great amount of cost that might be a good way forward in convincing people that hey you could just maybe publish it there and then you still have got your I don't know your monograph your research article or whatever. Um it's easily accessible but it's still in this kind of prestigious publication domain. Yeah that would be maybe something. Yeah but it's of course. It's, of course, very difficult as well to really convince people of this. I mean, hopefully, at some point, the open science community is gonna be so big that many people are just adapting to it, because that's common practice right now. But I think we're also a long way from that yet. Yeah.

Interviewer: Yeah, that's interesting. Yeah we've reached the end of my questions but maybe if there's anything else you wanted to add or to say about open science in linguistics or in the humanities for that matter more broadly

Interviewee 04: Yeah I mean one thing that is of course tricky is I think your overall position. So if you're a phd student um it's of course very great to do open science practices there. But then I think the fear is even greater that I know some statistics nerd or some great professor will criticize your research openly and then because it's open science if it's all accessible everyone can exactly see what is going on. Yeah um so that's of course something. Um and I think it would be nice at least that's something that I'm kind of planning for a conference is to make a little bit of a distinction to say okay we do have for example a talk here um but this is by a phd student and this basically means is that you can still of course you know criticize it but make it. Maybe you know it's usually like um so you can criticize it but maybe in a constructive way because like I had a lot of criticism during my um yeah like development in linguistics but that helped a great deal. Like the worst like the worst thing is that even uh either you've got people really trying to destroy somebody right which is not great um or you've got people just ignoring it so that kind of there are many people in the room thinking okay I don't know whether the research methodology is really the best one but then just not saying anything is also not so great um. So that there is like a constructive kind of criticism there and especially that PhD students for example are kind of in a more safe space there that they can kind of test things out they can. You know that there is more understanding for that. So that's something that would be great because at the end of the day if you do open science you really you put yourself out there and you are completely transparent and then of course it's always possible that somebody could criticize you. But at the end of the day even if you do not do open science that might also be the case and that's also usually the case you know. So

Interviewer: It's true.

Interviewee 04: Yeah I don't know. But at least like the open science people could say hey I'm being completely transparent here I don't know what you do. So tell me just what would be a better way forward you know.

Interviewer: Um Sometimes I feel, okay, you have this alternative idea. Try it out. Here's my data and here's the code. You disagree with whatever parameter I used in line something. Try another one. See if the results are completely different. I'm curious.

Interviewee 04: Yeah, definitely. Which is probably the best reaction in these cases.

Interviewer: Yeah, I mean, not that I've ever said it that way. It's in my dreams, you know? Yeah. Yeah. Great. Nice. I think I'll stop the recording here, if that's okay.

Interviewee 04: All right.

Interviewer: Yep. Thank you ever so much.
