 So, I want to start with an article by John Osbaugh called The Infinite House, which contains a critique of a process called the Five Wise, which is a method of retrospective and getting at root causes of failures and other events by repeatedly asking why a particular symptom occurred until you get to a new symptom and keep going back the chain about five or so times, which is they've decided it's a good number for getting out to whatever the root causes and taking an action that prevents that outcome in the future and gets you to a better outcome. So you stop. So where was I? I was at the Five Wise root causes. Great, we can have better outcomes once we get to these root causes, change things in a process, learn from them. Awesome. But one of the critiques that he offers of the Five Wise is that in a lot of ways it's actually too simple a model to understand the ways in which the complex systems that we work with can fail, right? A lot of the times that the systems we work with and the world we live in are much more intricate. They have a lot more entities. They don't really follow this neat linear pattern of cause and effect. And a lot of times things have multiple causes. There's really no good way to trace backwards to these multiple causes using this method. So a lot of times systems look more like this, right? These really deeply interconnected webs of many entities and rich interconnections and we need a way of getting at many of these paths so that we can get a better model of the world so that we can learn from and what he offers as an alternative because the infinite house, which is instead of asking why something happened, we ask how something happened and we do that sort of repeatedly and deeply and try to get richer information at every step. And he says in order to learn which would be the goal of any retrospective or post-op investigation, you want multiple and diverse perspectives and you get these by asking people for their own narratives, something that stood out to me is really important, right? You want people to share their stories and their subjective experience of a failure event, right? It's not something you hear very often. You ask probing questions to ask what was going on here, what were you thinking about at the time, what were the options you had available to you, what were the alternatives that were possible, who did you contact and really explore all those little branches that you could go down. And once you do that, once you construct that individual narrative, you can compose that with other narratives from people who experience those events and get an even richer model than we had before. And one of the reasons this stood out to me was it's not something that I was taught or that I would learn from by computing or software education. It's something that seems hugely important to the way that I actually work with software in the industry. And surprisingly enough, it was sort of conceptually something I was introduced to more in my experience with theater studies. So in theater studies we learned about Bertolt Brecht, who was a German playwright and director from the early 20th century. And he offered a similar critique of the Greek tragedy that he saw it as a similarly linear, far too simplistic model of storytelling centered around an individual hero who underwent some sort of tragic event and has to suffer their way through it until they get some insight into their suffering and was able to take an action to change that. And Brecht really wanted his theater to be an instructive theater, one that really expressed all the intricacies of the world that we lived in so that we can take better actions to get better outcomes. And the way he felt that he could do that was by creating techniques in what we call the epic theater, where audiences were constrained to see the events on stage through the lens of multiple characters. We de-centered the individual hero's narrative and saw that as a part of an ensemble, many individuals where you saw the events playing through through many individuals on stage. And the characters on stage articulated their decision-making processes as well. They talked about their alternatives, how they could have taken another decision and maybe have gotten to a separate outcome. And that struck me as really, really similar to what I saw in this work. The other thing that stood out to me was that focus and embrace on narrative as a way of understanding these complex systems and computational space. And that embrace of narrative is something that I know more from the conversations that I've been having with my partner about ethnic studies and she's been very patient teaching me a lot of that stuff. And really ethnic studies for those who aren't familiar is a field of study that seeks to understand and teach and share the stories and histories of racialized people and people of color, in particular through their own perspectives, using their own lens, using our own words to describe those experiences, which is in opposition to many traditional fields to do that from an outsider, sort of objective, alienating, usually Eurocentric white perspectives. And it's a very broad field. It's an interdisciplinary field. Among the work it produces is a lot of quantitative information and studies and research to explain these stories, but it can't live up to its mission without embracing narrative and really seeking out individual and collective narratives about the communities that they seek to understand. And so it really goes out of its way to find those in as many spaces and as many media as it can. And nonetheless, despite all the information that it produces and the rich models that it is able to produce about the world, it often gets criticized almost because of its embrace of narrative, its embrace of individual perspectives and perhaps even because of the kind of analysis and the results of that analysis that we get from it. And that's something that, at least from what I've seen, General Halsbach has not had to deal with as much, right? And that's something I think we see in the tech community as well in that when people share their experiences of marginalization and their experience in the tech community, they're often really the really common refrain is like, is there data to support what you're saying? How can you make such a general claim about this? Is there a study? Is there research? And of course there is. There's a wealth of research. There's a wealth of information. But even if there wasn't, that shouldn't invalidate the experience, especially when we are doing the work to get the deeper stories, to get the deeper explanations. We're doing that work to compare those stories with other people in our communities and to really understand those models. And there's a lot of reasons for this, but I think at least part of this reason, coming from a programmer's perspective is that I think programmers, software engineers really like to privilege quantitative tools as a means to understand the world. They really seek that as the primary, if not sometimes the only means for getting information about the world. And a lot of the times there's so much information that's lost when you're not exploring that. And one illustration I'd like to show about this that I found really interesting was from Gerald Weinberg and his introduction to general systems thinking. And he's describing sort of the space in which we can understand different parts of the world. And he puts on this graph, he puts the level of randomness on the y-axis and the level of complexity on the x-axis. And he cars this into three spaces. And in the lower left corner, we've got a space that is really well organized, which is to say it's not very random at all. But it's also very simple. There's very few entities. There's very few interconnections between those entities. And conveniently, this is a space that we can understand through formal means, through analytical means, through mechanical means. And that's great because we get a lot of information about these kinds of things in this space. And in the top level, we get an area that is highly random, right? There's large levels of randomness, arbitrary levels of complexity, but sufficient randomness to be analyzed pretty thoroughly using statistical methods, which is also great. We get to understand a large part of that world. But there's this vast space in between, or in the third space that is sort of too structured and too organized to be understood with statistics, but yet it's too complex to be understood formally. So it asks a big question of what do you do there when you have neither of these tools available to you? And this is a space of ambiguity, right? This is a space of uncertainty. This is a space where we don't have algorithms, we have heuristics, right? This is a space where we have judgment, maybe intuition, but not so much script, right? And it's exactly the space that I think the Infinite House and this Embrace of Narrative, I think, tells us that if we are willing to leave these tools behind, we can explore a lot more of our complex systems by these means. And I think what software people try to do is that they try really hard to get everything to fit in this formal space or in this statistic space as much as possible, and maybe without realizing that they're using these other tools to pull them into those spaces when they have them available. And the other thing is that even when we recognize that our systems might be too complex for these tools, we might draw a hard line and say, in order to reduce the amount of things that I have to think about on a daily basis, I only care about the computer things, the things that run on the machine or whatever that may be, and everything else, the space that gets a lot fuzzy and mostly the people. We're just going to leave that off because it's too hard to think about where it's not worth the energy or it's not the thing that I care about. And I think we do that to a peril because I challenge really anyone to be able to draw that line distinctly. Technology is a tool which is to say it's an extension of human capability. People build, run, operate, and use our software, and we might very well miss points of leverage that we would get if we were willing to reach out onto the other side of that boundary. And so I am glad to have stumbled across Alsbaugh's paper because it does give us some tools for navigating that space, but one of the current requirements of being able to do that well is the idea that we have diverse perspectives to inform that exploration. And the reason you want that is because when you're doing that composition, no matter how deep you sort of do that individual narrative, when you do that composition into a collective narrative or shared understanding, if you're getting the same perspective every time, you haven't added any information to this space. Like if everybody is seeing the exact same thing, you're not seeing anything new when you're composing the narrative. So you want definitely a diversity of function at the very least, different people who interact with the system, ops, QA, I think we're comfortable with that in most places, but you also really want a diversity of history and experience and really because people's sort of value systems that they bring with them and other contextual information really affects like what you're inclined to see at a point where failure are some other events. And what's really meaningful about it here is that I think that there's a positive feedback loop that happens there where you create, where when you have a diverse environment with diverse perspectives, you create a better learning environment, when you create a better learning environment, you're sort of better able to attract and retain those diverse perspectives. My loop works the other way around as well where if you have a homogenous environment, you're sort of less able to see what it is that you might be doing that keeps those diverse perspectives out of your environment and that feeds back in the other way. I do, the last time I have, I want to quickly just sort of restate the question of why we would want to do this and I think a few people, and I've touched on a little bit earlier, you know, see sort of the business and organizational benefit to doing work like this, to be able to learn and get better outcomes in our organizations, but I really think that it's important to be able to have richer models of the world because of the hazard involved in that as well. And I want to explain this by way of analogy. I once took a motorcycle safety course and one of the first things they tell you in a motorcycle safety course is that most drivers can't see. They don't see. And if you're in their field of vision, it doesn't matter, you're directly outside, they're not trained, they're not attuned to seeing objects in their field of vision that look like you. And so it gets really, really difficult to want to share a road when you're exposed and vulnerable to the environment in a way that someone who's in a car is very shielded and sheltered from it. And that might especially be the case when someone's in that car and is willing to move fast and break things. And I think there is a big risk involved in that, that our inability to perceive people as vulnerable in our systems as part of our communities, as our users. We can take actions that have unintended consequences that are really rather big. We build platforms in which people experience waves of online harassment. We make changes and tools that inadvertently expose people's personal information to people that shouldn't have access to or can compromise their safety or even open it up to surveillance. And so it's important I think that we are able to use these tools to be able to proceed the world better. For many of us it requires really giving up our sense of expertise over experiences that we don't have, over consequences that we don't feel. And giving up our sense of expertise over the methods and tools by which we understand it. Because I want to believe at least that the technologists are here in an attempt to make the world a better place and really trying to make a delightful leisure time and better work environment or whatever the case may be. And I want to be a part of that goal as well and I think we need to do that not by driving past or around people, not by driving through people but really by driving with people to that goal. So that's all I have to say.