A 3-Question Framework for Making Ethical Tech Decisions

Key Takeaways

[music] You're watching the new stack makers, a podcast for people who develop, deploy, and manage atscale software. For more information and articles about atscale technologies, please visit the newstack.io. Now, enjoy the show. Hello and welcome to another on the road episode of the new stack makers. I'm your host, Heather Joslyn, editor-inchief of The New Stack. And today we're going to talk about uh data and ethics and h and doing the right thing and why we should make it easier and not harder and all the challenges involved in that. And uh we're talking today with Meg Derty. Hi Meg. >> Hi. Nice to be here. >> Nice nice to thank you for joining us. And and uh we are um you you we're here in San Diego Jupiter Con. you gave the keynote address, one of the keynote addresses this morning and um Meg is a uh fellow at at the Harvard Medical School Center for Bioeththics and uh we're going to talk about what you talked about in in the um in in the keynote, but also the bigger issue of of data and sharing and ethics and and uh how to make that a um act how to make protecting it data and sharing it uh ethically an easy thing to So, um, you talked about, you know, how to make, um, the, uh, the right thing to do the easy thing to do. Um, it's often the hard thing to do. Um, but, uh, what are some of the challenges within an organization that can arise that make it harder to do the right thing? >> Yeah, absolutely. Um, my talk today was really about first defining these two things. Um [clears throat] that's the hardest thing when things start to fall apart or decisions get made and teams are confused. It's often because there could be a definition problem. So first and foremost is deciding what is what is right and what is easy and making sure that we understand those two things. And one of my my core messages in my speech earlier today was about >> deciding what is right often shows up in our core values in an organization or a team. And so having that scaffolding of for decision-making is really helpful. And so some um the organ organizations that I've partnered with and I have worked with, that's where things start to fall apart when that scaffolding for ethics doesn't quite exist or maybe it just as I said, it's more like a hallway poster and people don't really think it's used for decision-m. Um but really having those structures in place matter. >> Yeah. And that that's a good point too about um the values of an organization and you as you mentioned some it's a hallway poster the mission statement or the values are not um are not uh well known by everybody who works there um or they're seen as you know some some something that people put together early on that >> exactly >> may not necessarily apply in real life. Um so so aligning with value the values of the organization. Um you you talked in your speech about a a specific project you were involved in. Um can you can you talk about that a little bit? >> Yeah. Yeah. Um so the project that I was involved with we have a a group of 15,000 researchers who are using Jupyter notebook for genomic research and there are many ways that Jupiter gets configured and most importantly it's meant to be sharable and embrace principles of open science. Um we had built Jupyter notebook inside a secure environment um surrounded by uh federal policies and so sharing just wasn't uh you know it was technically possible but there were policy reasons why we could not share publicly most importantly patient privacy reasons >> and this was at NA >> and this was at the national institutes of health. Okay. Yes. And so um as as we work through that getting together the right policy stakeholders with the technologists to really say what how how can we make this happen or what needs to be true for this to happen >> and um in the story that I spoke about this morning is really about um building internal momentum for an idea >> and that's [clears throat] what me and a few colleagues that were with me on the consortium really just decided it's not just about a button that needs to exist by having this button that says publish on it. We are making a commitment to open science which we already are. So this is kind of an example where the technology and the policy and then the values come into play to really check in with ourselves of are we open science and the answer is yes we are but there's always the answer of to what extent and there are limits to yeah we my old boss um used to say uh we want to share widely and wisely >> and it's kind of a good kind of mantra to keep with you to remember what we're trying to accomplish. >> Yeah. >> Yeah. So, so there was you we often talk about when we talk about change or a new project within a a software engineering organization, we talk about stakehold stakeholders. You can't it can't be top down. It has to be >> involve everybody. Can you talk a little bit about you know so did you have like interview like meetings with the different stakeholders? Did you like how did it work in a practical sense? Yeah, I um so my my life before technology was um more on the political side of of of policym and so I spent a lot of time building coalitions and alliances and getting volunteer communities to work together and so I'm proud to say that many of those skills translate into engineering and software organizations and so a lot of it is really um yeah like you said sometimes it's top down but often times in these sort of big massive >> digital transformation projects, it can be bottom up because folks who are on the ground doing the work, they're going to see exactly where the problems are. So, what what my approach often was was finding other people who wanted to make some things happen. And like I said, really just >> finding getting it done, finding out how to get it done, you know, each brick wall and just saying, how are we going to move through that? >> Yeah. Did you find do you find that um the technologists are are concerned about ethics or do you feel I mean we seems definitely hear from developers who are concerned about about the ethics of what they're doing or was it something that was there an education process around it for them? Yeah, I think um it's definitely we're all human >> and we all have we all make ethical decisions every moment and so part of why I'm passionate about bridging these gaps is to really remind people even if they're tech in the role social role of technologists um they have some ethics and so bringing those to the surface requires I think a lot of uh trust building in conversations some people don't you know they if somebody is used to engineering problems they have a if this then that yes no very kind of this structural natural way of producing things. >> Yeah. >> But talking about ethics and values can feel overwhelming because it can feel messy. Um, so part of what I'm hoping to kind of what I wanted to bring today and what I wanted to bring here and into the future is like let's just structure ethics and values conversations in a way that doesn't really scare people away and lowers some of the barriers um to having a conversation. >> Yeah. I think sometimes people feel when when when and I've been part of ethics code writing and other organizations I've been involved in and and I think people what you want to avoid is like people feel like it's like a list of got thing gotcha things or like like a checklist of yeah you know if you don't do these x number of things you're a bad person or something. >> Exactly. And I one one opportunity I took last year was to complete a fellowship in bioeththics. Um >> which brings me here today. >> But going through a year-long fellowship, the the outcomes were really about ethical analysis and having confidence to look at a really thorny problem and knowing how to methodically walk through that problem to reach some type of consensus. One thing that I really um really resonate for me working in healthcare technology is this duty to resolve. We have a duty [clears throat] to resolve problems and often in software engineering we can have kind of problems that are just floating there but we just keep doing the work. We just hope that it doesn't show up, right? [laughter] And so so we keep doing the work but really um having deliberative spaces to bring people in to just say it's okay. >> Yeah. it's they're non-punitive spaces to be open to saying the wrong thing or maybe not not really knowing what to ask. Um so those types of things really help I think just kind of lower the temp I like to lower the temperature on the conversation. >> I like to talk about ethics but I'm also a very positive person. So [laughter] people think ethics and they think okay I'm not talking to her she sounds like legal and that those are two different things. Yeah. >> Yeah. Well, like the concept where you're talking about is so what they I guess like psychological safety I guess they say in engineering organizations where you you can say you can ask the question that might be dumb quote unquote or or and no one will >> you know um there'll be no no consequences for that. >> Exactly. Exactly. Yeah. >> Um you you've put together like sort of a three question framework for dealing with uh navigating those e ethical tensions in in one's work in the workplace. Um can you explain that framework? >> Yeah, absolutely. Um so the first two questions come from um an amazing colleague and creative director. Um and the questions are who is it good and why why is it good and who is it for? >> And so uh why is it good and who is it for really just helps frame your problem set. Who are we solving for? What are we solving for? Mhm. >> But I think the third question that I've really been sitting with for a while, especially now we've got this these advances in technology happening so quickly to us. Often times it feels like it's happening to us. We're not shaping it. Yeah. >> The third question is why is it bad? >> And we work in agile frameworks and methodologies and there are ceremonies that we call you know premortem. So let's have a let's have a meeting where we talk about how all the things could go wrong in this project ahead of time so we can at least get it out. we know the risks and we can sort of manage them. >> I think the same is necessary for ethics. It's okay to be working in these new emerging fields. >> I think it's also equally important to pause and say where could this go wrong? >> Yeah. >> And and just sit with that discomfort. That doesn't mean you have to stop. But I I want to encourage people to do that with open eyes >> and open understanding of what what could go wrong and then and then manage against it. >> Yeah. >> Yeah. >> Yeah. Yeah, >> it definitely avoids the sort of um helps to avoid the um sort of a herd mentality of you know we all agree it's a good idea. So >> yeah or or post talk and say oh we we had no idea this could happen you know it's like [laughter] well >> I'm shocked to discover gambling in this institution. >> Exactly. Exactly. Yeah. >> Yeah. [snorts] >> Um you also you t in the keynote you talked about um I mean we're at an open source conference basically. I mean that the word open isn't always a friendly word for for everybody. So can you talk about that a little bit and um >> how you how you design systems for people who are not in the room. >> Absolutely. I found my way to um Jupiter and open-source software through um two particular open uh ideas and projects. One was um open government data and civic technology and the other is open internet and the freedom of of internet around the world >> and in those two experiences >> through the Mosilla Foundation and the Open Gov Foundation I really started to see the limits of openness >> and there's a whole other podcast episode in this [laughter] but the the main takeaway is that um openness often feels like free open of course why wouldn't we want these these things out there but they're [clears throat] not neutral. When data gets into the wrong hands, when open data is um uh overexposed, >> then we we we um perpetuate real harm. And often, if you look at many different examples, there are communities, marginalized communities who have been the most impacted by these open data. And in my talk today, I kind of framed it as, you know, open data can mean extracted data as can open science can feel like extractive, right? And that's where one one area where I'm proud to work with a bunch of of folks where there's that closed feedback loop. So when we take when we take data when data we have get enter into a data donation relationship that data is shared back with participants the benefits of a new drug from a clinical trial actually becomes free for the patient who gave the data to support the clinical trial. So I [clears throat] think there are things we can manage with openness. Um but there are limitations and it's not neutral. Yeah. >> Yeah. >> Um, one one uh more thing. How if I'm say I'm a developer uh or an a software engineer in a technologist of some kind, how do I educate myself about these issues of ethics and software ethics and data? >> Yeah, absolutely. If you're at a research institution, find your humanities, medical humanities, bioeththics, philosophy departments. Get curious. Um, one thing that I have found, um, I I'm I'm not an academic, but I found myself in academic spaces over the past few years. People want to share. >> They want to [clears throat] share their knowledge. They want to they want you to know about their work. Ask questions. And >> you can also find your colleagues. Um, I I one of the reasons why I gave this talk today, I knew that there would be a few people in the audience who would come up afterwards and say, "This resonates that I'm so glad that you said these things." And sure enough, those people did. And those are those are the people who matter for my talk today because I don't want people to feel like there's no one else thinking about this and that we can connect with each other. Um, but I should take this moment to say that's um part of why I started this new space u called both and neither where I'm going to bring people together to have these types of conversations that I hope will kind of resonate throughout um the scientific spaces and um yeah re focused on research software engineers because those are those are my people um but expanding beyond that. >> Okay. >> Yeah. >> And when you say a new space what what what exactly what does that mean? >> Yeah. So, I have um the a new creative studio called Both and Neither, and we're going to be running community workshops and hosting events and socializing these types of frameworks, easy to use, easy to understand, implementing ethics in your technology work. >> Excellent. >> Yeah. >> And um any anything else on the on the that that's the the big thing that's on the on the >> Yeah, the big the thing that's on the agenda is this. And I'm also um stepping into uh another role um where I'm going to be full-time at the nation's first oncology focused venture studio. And I'm really excited the way that cancer research and technology are coming together in this way. um we have a chance to cure cancer and and I can't wait to bring these thoughts there because there's no place on earth that has more complex ethical decision making at the bedside at the moment of care than when you're faced with a cancer diagnosis. >> It's very true. >> Yeah. >> Well, um it's been a pleasure to talk to you and thank you for joining me today. >> Yeah, absolutely. Thank you. >> Thank you. Thank you. And uh this has been uh we want to thank uh Meg Dowardy from uh the Harvard Medical School Center for at Bioeththics. And uh this has been uh we also want to thank all of you for joining us for this conversation. And this has been Heather Joslin for the new stack makers on the road at Jupiter Con in San Diego. We'll see you next time. If you like this video, please give us a thumbs up. And if you'd like to see more videos like this, you can always subscribe to our YouTube channel. We're on all the major social media platforms. You can always find us at the newstack.io. We hope to see you soon.