 Okay. Well, hi, everyone. It's great to be here today. Many thanks to Dr. Piasecki and his team. Many thanks to Agnieszka for organizing the session and a big congratulations to the team on your successfully funded project. It's a great privilege to be here, too, with Drs. Morzi and Propharis. I will be providing some brief remarks and then moderating the session. I'm going to ask that you hold your questions. You can send them in throughout the session, and then we'll address questions at the end. I just wanted to just say that I have no conflicts of interest related to this activity, and any opinions or comments or recommendations that I expressed today are mine and do not reflect my employer, Marshfield Clinic Research Institute, Marshfield Clinic Health System. Like many people, you may have some unofficial disclosures. Zelda and Wally are very quiet so far today. They're peaceful, and it's a fairly rainy, cloudy day here in Wisconsin, which is the little red area that you see there. Typically, though, as we've all been working from home now for many, many months, and as soon as I start a presentation, of course, the mail or the post shows up, and the dogs go crazy. That's just my warning in case we have a visit from Wally and Zelda. As Jan said, I've been working in this space for over 20 years, and I think the research that's taking place in what we call the internet in its broadest terms, I think we're really seeing just growing complexity. A number of years ago, my colleague Michael Zimmer and I were approached to do an entry for the Stanford Encyclopedia of Philosophy, and so we just finished our third substantive revision just this year, 2021. But what I was thinking about when planning for this session is it's going back to the initial definition that we came up with back in 2011-2012 when we first put this entry together, and I think it still holds true. Conceptually and historically, internet research ethics is most related to computer and information ethics. It includes such ethical issues as participant knowledge and consent, data privacy, security, anonymity, confidentiality, integrity of data, IP issues, community disciplinary and professional standards or norms. And since early work, again looking back into the 1990s even, some questions have emerged. And I kind of jokingly said so many questions on this slide, but these are just some of the questions that we've identified over the years, some early on in our first iterations of IRE research. But as we glance at this list, it really is growing in complexity. Jan mentioned that when we were talking earlier about the new GDPR, and researchers are now facing so many different complexities as we work in these different spaces. And I would say that the early questions, those that had to do with issues of what is what is public, what is private, those questions have certainly not gone away. And I would say they certainly haven't been resolved to any great degree of confidence. And I would say to the contrary, many of those issues that we identified way back when have just grown either in intensity or in complexity. And I think it's fair to say that as we think about social media, we think about big data, machine learning, AI, those have stirred the pot of questions. And I think now we're at a point where we're seeing such complex considerations in internet based, internet, broadly speaking, research that we're pushing very, very heavily back on epistemic and normative constraints. And I think as we as our collective interdependency on social media has grown over the years, our relationship with trust, truth, representation has also dramatically changed. And, you know, I think it's funny, we go back all the time to this silly cartoon from the 1990s about anonymity, right? Anonymity on the internet. Nobody knows you're a dog. But think about where we are now. And we're in this space of what we've moved from that notion of anonymity to precise levels of identification to prospective modeling of behaviors, of prediction, of prediction. And it almost seems seamless, right? That these transitions have happened, you know, from that space of anonymity to where we are now. And yet the ethics of the transition, the really, the ethics are enormous. Excuse me. And a few years ago, I wrote a short response in public library of science. And we talked about, I was responding to an article that used a form of network, social network analysis called iterative vertex clustering and classification and IVCC. And the article was talking about using that model to identify specific populations in large data sets. And I went back and forth with the authors. And, you know, it was a really fruitful conversation where we talked about methods and ethics. And how do we front load ethics in these kinds of research methods? It also got me thinking about the true methodological power that we as researchers have at our disposal today. And going forward, I think that that power will increase. So as researchers working in on in between spaces on our internets today, we have we have unbelievable access to data. But I also want to remind us remind us of the risks that that we also take as as researchers in these spaces. And, and I think, you know, we only need to think over the past, you know, five to 10 years of some of the the famous or infamous cases of studies gone wrong to think about the risks that possibly could occur. So as as I read more about Jan and his team and this project, and as they engage on this very ambitious project, they are going to be facing a tremendous ethical, regulatory, disciplinary challenges. And, and so, you know, what what does a research team do? Where do you begin when you talk about front loading ethics? Well, in from a Western perspective, right, we would say, Well, we start at the Belmont report, we think about how we treat our participants in research, we think about risks and harms, and we think about the benefits of research and also the burdens of research. But I also want to suggest that we think about an old, an old document, another old document on being a scientist that really talks about the values that that are essential to our work as scientists and that we honor the trust that colleagues place in us, that we honor an obligation to do the best work possible and to embrace productive and honest work. And then this last one to uphold an obligation to act in ways that serve the public. And I think it's it's really in that last point where I was thinking about the web immunization project as really critical. And recently I was I was talking with a library system about distinctions of of miss and disinformation. And so as this team is starting this this project, looking at this at misinformation, you know, what what are they going to uncover? And what I what I hope what I really hope from this team and I have great expectations for this project is I'm hopeful that that empirical research that empirical research, looking at these forms of rhetoric, looking at empirical data around misinformation, disinformation, I'm really hopeful that that's going to contribute to a healthier discourse. I hope our our citizenry is is better informed as a result of the work that's going to be done. Excuse me. Okay, so just a few more comments. I think we do know a lot more about the ethics of conducting internet research or research on Twitter than we did just just a few years ago. I think we've really come a long way, particularly from some of the work that that Nick Propharis has done. I think we're really starting to understand those those those very complex tensions that exist in those spaces of public and private. And what do participants expect? There is a substantial literature across disciplines, but a trend that I've I've seen and I've heard in many conference presentations. And I think it's a little confounding for ethicists and for for certain for some researchers that that there's this presumed public nature of Twitter that that's kind of become this default position. And I think we still need to tease that out. Regulatorily, I think that that probably is the correct framing of of Twitter data. However, what about the ethics piece, right? I'm also concerned about the ways in which that public nature enables us to, you know, within, you know, an hour to, you know, get to develop an API, operate, you know, get this API to to to to grab all sorts of data for us, right? It can enable we can have all sorts of research projects, you know, acting very quickly. And so I said this a number of years ago about these spaces that the shifting research landscape is complex, that the fact that data are coming from these myriad of sources, some of them intentional and some of them unintentional. I'm concerned to around about the concept of research bystanders or collateral subjects as sometimes called in the streams of data that that get scooped up. One, once one's connections in a social media landscape do matter, right, they do matter, even if those those connections are distant or impersonal. Human subjects research broadly understood is fun, it is or should be fundamentally different in the age of data science. However, from a US regulatory perspective, the regulations really haven't kept up, they still haven't kept up with with the the age of data science, despite the rule revision in 2018. Methods such as IBCC rely on continuous data streams and and continuous analytics, and many of these data mining and analytics studies would be considered secondary analyses. And the degree to which a researcher has access to identifiable data or the ability to ascertain information about individuals through, for example, re identification techniques are used then as as determinants right of the level of risk and benefit in the current US regulatory model. So I just want to wrap up with a few words from a regulatory perspective. In the US we did have a rule change in 2018. And I think a lot of researchers were hopeful that the revised common rule would address some of the technological changes taking place to some little little degree they did. I think most of us didn't quite see the the changes that we expected. But when we talk about these kinds of large data sets, we kind of have two different paths here from a US regulatory perspective. We we either fall completely outside of the common rule definition of human subjects research. And that our definition is a living individual about whom an investigator, whether professional or student conducting research obtains information through intervention or interaction with an individual and uses studies or analyzes the information. I'm leaving out the biospecimen piece for now, or obtains uses studies, analyzes or generates identifiable private information. So so the path would be okay, does this does this data set, for example, comprise human subjects research? If if not, there's no there's no ethical oversight, right? It's it's outside of the purview of the Ethics Committee or the IRB. If we agree that it does fit that definition of human subjects research, the next step would be okay, what kind of research is it? Typically, the types of research we're talking about today would qualify for what's called an exempt determination status. And and once that determination status is is is exempt, that means IRB oversight effectively ends. So again, we're back to a point of there's no IRB or Ethics Committee oversight. And and are we comfortable with that from from a research perspective? Typically, what we would see, we would see that research in this vein would fall into an exemption for criteria, which is about secondary use of identifiable private information. And this brings us right back to those questions that we started with, right? Well, okay, what is identifiable private information on Twitter? Is there such a thing? Or are we going to go with that that perspective that it is de facto public? So I leave us with that to think about these regulatory issues that use use words, like when identifiable materials are publicly available. So we're back again, squarely to the public space is Twitter, Brit Large, a public, a public space. And so I want to leave us with those couple of lingering questions about public and private. I'm I think we still have it's it's great. We all this is job security, I think because we all have so much more to to learn about internet research ethics, writ large, and the ethics and the regulations, I don't ever feel that that they're going to be completely in sync. So that puts the onus back on us as researchers operating in an unregulated space to do the right thing. I think it's pretty clear ethics and regulations, you know, aren't always in the same sandbox. So with that, I'm going to turn it over to my colleague, Dr. Morse, and we're going to talk through some machine learning and some ethics.