Authors:
Felipe Massa — Loyola University, New Orleans College of Business
Siobhan O’Mahony — Boston University, Questrom School of Business

Interviewers:
Shelly Qi — INSEAD, Department of Organizational Behavior
Shuping Wu — INSEAD, Department of Organizational Behavior

Article link: https://journals.sagepub.com/doi/full/10.1177/00018392211008880

---

1. Our first question is about the context. How did you become interested in Anonymous at the time and how did you get access to this interesting context? You described surfing the websites, collecting real-time digital data, and doing field observations. Can you share with us more about the amount of effort you put into the data collection? Also, we wonder if you can share more about how you get Siobhan involved in this project? 

Felipe: My first interest in Anonymous started at around 2009-2010. It got sparked by the fact that I didn’t understand what they were, who they were or what they were doing and how they were able to organize in the first place. I don’t know if you all remember from back then but at the time Anonymous had started attacking the Church of Scientology—they were very interested in disrupting what they considered a cult, and they were using these very unusual tactics for a collective acting like a social movement. They were using these hacker attacks against what at the time was not a government target or a large corporation but a religious organization. I just found it fascinating that this group of people with no clear leadership hierarchy were organizing themselves in the traditional way that we think about organizing at least from a social movements’ perspective. 

They were able to create a lot of damage and really get a lot of attention in the media so that’s where I started to get interested. I started lurking in the different forums where Anonymous was organized, and it was fascinating to see how they all would go through the process of introducing new ideas into this forum. They would talk about these ideas of going through several rounds of attacking Scientology and then come back and debrief and learn from it. That process was just fascinating to watch as an observer and to see that things were working even if there was nobody in charge. That was sort of the mystery that got me interested in doing this, and I think Siobhan got into my story a little bit and was interested in this kind of stuff from an open organizing perspective. I was lucky enough to be able to get her on board on my dissertation committee. Siobhan was one of the few experts that were out there who really had done this in the open-source community and understood how they organized. I was lucky to be able to find somebody to help me understand all this stuff.

Siobhan: This has been a very exciting project, and I think the collaboration with Felipe has been amazing. One of the struggles we had was about which story to tell. We had many different papers in many different revisions, trying to find the right frontend for the interesting novel phenomenon. We wrote a paper about coordination because that was interesting. We had over 70 different raids on unsuspecting targets. You can’t have an impact on a target without coordination, so we thought that was interesting. Then we had a paper on secrecy because they were anonymous and not identifiable. We had another version about technical innovation because these people have been trying different ways to create impacts on suspecting targets. We had a lot of trouble getting a receptive audience under these different theoretical lenses and we just kept trying. I have always avoided the term self-organizing because I feel like it’s masking the most interesting things–what does it mean to self-organize. After reading some new research on this, we began to say, “Ok, I think we can get into self-organization now.” It was hard because we know what those people we studied said and did, but we don’t know who they are behind that. We also struggled with presenting the work and being honest about it. It’s valid behavioral data. In the current paper that was published, all the stuff about being hidden or being in the dark is not really at the forefront of the story anymore.

I fell in love with each version of this paper, and we had a good time writing together. You know this is such a fascinating phenomenon that sometimes I think when the phenomenon is so fascinating, new and interesting, there are so many avenues you can pursue. The hardest part is to figure out what an audience will be receptive to at that moment. And then what is the right conversation to join because there are many conversations you can join when you’re studying a phenomenon that is itself maybe even a new conversation. It’s a complex system and there are many pieces of it. If you try to write a paper that explains the whole complex system, you’re going to be really frustrated. We worked on this paper for 7 years, reframing, editing and analyzing the data over and over after the data was collected. I don’t like to think about it, but we were told by another Journal that they didn’t see how this is relevant and I was kind of stunned by that. But then I thought that we have work to do. Fortunately, sometimes the world catches up with you, and suddenly after the capital riot, network activism sounded more important. All of these became the norm in any kind of activist organizing now. What happened while we were collecting data and writing the phenomenon became the dominant way of organizing and engaging in collective action and is now on these online platforms. The relevancy increased and the way we frame things became more obvious as we went along. Now it is interesting to see communities like Ukraine take some of these tactics and fight for their freedom and democracy. The paper on technical innovations that nobody liked three or four years ago is now going to be full of it. I’ve had this experience a lot in my career. Sometimes you are out in front of something novel, and it’s very hard to but you must be prepared to explain not just whether or not but why it is important to others who don’t share your enthusiasm. I think I have gotten better at that over the years but that’s really a tricky skill.

Felipe: I agree this was a great learning process for me. Specifically, the idea of thinking about moving the field rather than being moved by the field and looking for how to start a conversation rather than joining one, and figuring out clever ways to feel like you’re joining a conversation while you’re trying to start a new one. I learned a lot from Siobhan on how to frame these things.

Siobhan: I think we have this data—they say that it’s quite nasty, so it’s hard to interpret, you know, but I do think it’s important to look at the dark side of phenomena and not be afraid of it. If we don’t do that, there is a lot of things you won’t be able to explain. So I think Felipe in particular was extremely brave. He could talk more about access issues but you know that was a hard thing. Some people were uncomfortable with the language they were using but this is real. That was in our data, so some of those things may be hard to read, but they are important for capturing what people do on the edge. We really didn’t shy away from showing it which I’m proud of now but that gave us a lot of work in the process to get the paper accepted.

2. I think that is related to our second question. Could you please share how you overcome those challenges in different stages of conducting research?

Felipe: I can start with data collection. Collecting data in this context was a huge learning curve for me. I think we are used to collecting data where the phenomenon is fairly static—there are meetings that go on, and there are people that are talking to each other through traditional means that don’t change. In our case, the participants in Anonymous were essentially moving all the time. Where they talk to the forum changes regularly; the way they talk to each other changes; the architecture of how they organize themselves is also fluid and hard to keep track of. So that was a challenge to figure out what is the place—where they’re actually coming together to make decisions and do these things that they were purposely surreptitious about. I think this is where the ethnographic methods are helpful and where you’re embedding yourself so that you are able to keep track of the changes in the community and follow the development of the community.

Getting approval from the IRB was a big challenge mainly because the people in the collective were doing illegal things. The institutional review board isn’t often keen on people studying that. So I spent a lot of time going back and forth with the institutional review board at my university at the time and trying to figure out what was the safest and best way to do this, which is where you were talking earlier about lurking and doing things without revealing myself to the people in the community. That’s sort of where I settled at the IRB would be a legally and still methodologically good approach to take, where I wouldn’t be intervening with anything that people were doing. We were all figuring out this on the go. Trying to create this online ethnography where the context changes so rapidly and where the subjects observed were anonymous. So there are a lot of inventions on the methodological side. 

Siobhan: As Felipe was talking about, the setting was changing. This is a very dynamic context. I think Felipe is being very modest here–he got into the context at a great stage and much of what we were able to observe cannot be observed anymore. Much of the things that we were able to observe have been pulled back into the darkroom, so some of those processes that we were watching unfold, that temporariness is gone and we can’t go back and get them. I think the other tricky thing is that IRC is the data source. The anonymous community really likes that because IRC leaves no lasting trace. So there was no IRC archive file, just chats that could disappear. So we talked about taking notes and how to capture that through taking screenshots or some other form. Then we archived the mass of them with massive amounts of conversations. 

Felipe: Most of the conversations were completely irrelevant to the organizing process so there was a lot of noise that we had to parse through to figure out—are they talking about something relevant, are they talking about a raid, or is this just people going back and forth talking about memes or things like that. I think that’s an ethnographer’s problem generally, but in this context with data being generated is huge. There were often multiple chats going on at the same time in different countries overnight at the end. I could have used a lab of people all gathering data at the same time to really capture all of it. So we really had to be focused and clearly define what was of interest and that was a challenge. 

Siobhan: We decided to focus on raids, specifically their form of attack, and I think the reason we made the choice is because it wasn’t just an exercise of disgruntled people giving voice but this is how they create an impact, and then we’ll be focused on that. It really helped us create the framework to understand all the pieces that go into raids.

Felipe: We move towards tangible things that we could actually look at and actually pick up on.  

Siobhan: Yes, we could write about culture, about language, but what we really want to know is how they create impact and that helps us a lot.

Felipe: Yes, without any of the scaffolding that people are used to. We have to figure out what the scaffolding was for them and how they were able to build this thing. That’s what the paper ended up being about and it was the mystery of how they actually go about doing this and creating an impact when they should have fallen apart. 

Siobhan: We presented this version and got some really helpful feedback. At some point, someone said to us that it’s not surprising that this thing grew and became out of control, but what is surprising is how control was re-established. We came back from that talk and discussed it. And we have to really find out what is really surprising to people.

Felipe: So that goes back to the title of the paper Order from Chaos. I think what surprises people in a room filled with smart people is a good way to figure out where you should take your paper.

3-1. How did you find out about the role of operators in restoring order from chaos from the massive amount of data you had at the time. How did the architectural configuration emerge? Can you share more about how you build your current model?

Felipe: Since the organization grows in an uncontrolled way that is in some way self-destructive and the community essentially falls apart about halfway through our paper. Then what is the interesting part is that a lot of these veterans of this group, which will become network operators, decide that they need to build systems in some way to help the community not just survive but thrive in this chaos without taking away the magic of what made the community work. So they transfer from being in an open forum where everybody could just post whatever they wanted to actually creating pathways for contribution. That’s where the network operators in the channel operators who are the people in the individual channels and then you have the network operators guiding these networks that’s where they start to be more visible and more present is in the creation of this these pathways that allow people to click through and select where they want to participate and the channel operators engage in kicking people rooms and control communication in that way. So you have this architecture of participation or participation architecture we ended up calling it after going through a few terms. Collective of people actually thinking about how to create scaffolding for contributions was the most interesting part for me—the channel operators were kicking people out of forums so they could decide who should be there and who was spamming, and the network operators were creating rooms guiding where people went and how they should participate, which is like a lab setting up paths for mice and then letting them go.

Siobhan: We didn’t start writing about control until quite late and again I really think I can’t emphasize enough how important it is to talk about your work with others. At some point talking about the room they really can’t control what people do or what people say that they can control where they go. We viewed the IRC Network as a critical communication channel that was critical for conducting raids. Additionally, most raids do not unfold cleanly, so there is a lot of messing around. There is a lot of false starts and iterations—that’s not working, let’s try this. But then we started appreciating the control aspect, and we had all these categories that were really tied to data very tightly. Once we started sorting by types of control, the theoretical importance I think became more clear cuz we just realized that it was an overarching category that we could put these under different kinds of control and then we could follow how that change.

Felipe: This architecture idea popped up right like a building where you can open some doors and close others and give access to some who prove themselves and restrict access to those who are novices. So it’s interesting to see that architecture emerges over time through trial and error and through the actions of people who are in the organization but are not engaging in any kind of formal leadership or building that kind of organization that you would typically see in a social movement or an organization. So that formalization wasn’t there. 

3-2. We understand that these operators do not have authority based on hierarchical position. I wonder though, whether these operators have role-based power?

Felipe: It was interesting that the control wasn’t over individuals or specific people. They didn’t try to establish hierarchical control over individuals but control in a way that I felt was more subtle through the architecture and through the design of cyberspace, which puts constraints on where you could go and where you couldn’t. So yes, they were engaging in control but not in a way that we traditionally think about in any other context. What I like to compare it to is when people are designing video games and they’re designing stories and narratives for people to follow. The feeling was that they were creating these distinct experiences that people could follow through and restricting some experiences to those who would gain access and have already been there.

4. Is this architectural control model generalizable to other settings of self-organization beyond just open source communities?

Felipe: Yeah, we were trying to join a conversation and ask questions that everybody cared about. A lot of it was looking for organizations, business organizations, that had attempted forms of control that were unusual in that pulled towards this direction. We found it in places where people were building video games and doing those kinds of things, which I thought was interesting because it is the culture where Anonymous comes from. It is this idea of creating worlds, right? And I think if we don’t find it in our organizations, we’re starting to find it more when people are gamifying motivation in organizations—they’re creating a much more video game-like existence inside the corporate world, which is sort of what we see in the Anonymous, in capital riots, in the Umbrella Revolution in Hong Kong, and in all different examples. We see a lot of different components popping up both in activism and inside organizations more recently, so we regard this kind of architectural control important for us to study. And you know, it is an open question of what constitutes control really, which is why I like to call it an architecture, because that implies suggestions about where you should go and some walls, but it also implies pathways of possibility for people participating in different ways of organizing.

Anonymous is all voluntary. Nobody’s getting paid to be there to engage in these raids, so you have to make it fun and interesting for people so that they can keep going and continue in the process and continue contributing and feel like they’re part of something. So in a way you’re creating this architecture not just as an architecture of control but as an architecture of possibility. In the future, our organizations may become more open like this. If you think about it, we are approaching in some ways like a market—I have ideas for picking where you’re going to go and what you want to do for this organization, like in a shopping mall. The shopping mall might have restrictions on which stores you can go into and there are some VIP spots; but in general, you are finding the place through this architecture that is designed in some way by these operators you’re getting some people that are novices, that are more leisure-focused game players, that are not obsessed over this, and then there are the people who want to spend 40 hours a week contributing to Anonymous and you’re giving her opportunities for both of those types of contributions. Now that’s a sort of rainbow version of it. But there is a version where you’re saying well, we only want the experts to make it through the gauntlet of getting to the point where they have access to these high-stakes raids on important targets, so you can frame it both negatively and positively and that’s something that the operators and Anonymous would do—they would say like oh this is a great opportunity for you to do this as people gain skills and become better hackers. What I like to think about is how do I translate this to a work organization in a way that is not this post-apocalyptic terrible version of the future that nobody wants, and where it’s actually where work becomes as pleasant as a video game. 

5. What are major organizational theories that can be linked to it?

Felipe: When we looked at the phenomenon, we immediately saw a social movement and collective action theories to explain what was going on. So that is one place, and you know there’s a lot of work going on in sort of applying social movement theories to organizations that have become more mission-driven and more open-boundary. So I think that is to me one of the stronger linkages. But as you noticed, we didn’t have a very expensive theoretical review section in the paper, and that was purposeful. I’ll be honest it was hard to create links with that. So like Siobhan said we went through the literature on coordination, like Bechky’s stuff on coordinating in times of chaos and why people drop their tools. Here people were coming up with all new things, so we had to look at those more micro phenomena. We looked at social movements and the literature on intentional communities. We were really pulling widely from these different kinds of literature to see if we could constitute something that made sense, so network activism and self-organization became linked to all of those as a sort of umbrella term to help us organize all these different rules that we were having. The ideas of networked activism and control are all old concepts but were by themselves insufficient to explain the phenomena. So we really had to pull from different ones to reconstruct how we think about control.

6-1. Could you please share some future directions of this stream of research?

Felipe: The temptation is always to go back to the front ends that you throw away. I would love to do something on the tactics and technical innovations of the powerless, especially the process of technical innovation is interesting here. These collectives don’t have much power and have to rely on digital tools that are distributed and rely on creating their own tools, in order to have impact on these gigantic systems and to use the power that they have because they have the advantage of knowing more about computers, which gives them power. It’s interesting to look at bricolage in a way adaptability as you’re leveraging these digital environments that the environments themselves are malleable. This is what I think is cool, and what these digital worlds take away from us is where we are usually adapting to context and not adapting the context to us. In this case, you can say this world isn’t working, let me create a new one, literally from these digital components. You can create algorithms and automation to make the world run on its own so you don’t have any managerial responsibilities over it. You just watch this thing. I think about Bitcoin and the fact that it was created by somebody who we don’t know who he/she was and now it’s out in the world creating. And there is an automated ghost in the machine that really pulls it all together, so I find that’s the fascinating and I hope more people get into the idea of studying these architectures and this form of management that you create something and you let it go, rather than this Theory X and Theory Y type of management where you’re choosing between those X and Y what’s the form. This has been in science fiction for a long time—the idea of the computer guiding behavior, but this is for the light version of that as we have these digital tools in the world and things that we can create—we can launch these things and see how they work. I think everybody is hesitant to go in that direction cuz it’s a little bit crazy to think about it, but we are all creating algorithms that guide our consumption of media and guide our behaviors when we engage with all kinds of things. I always go back to thinking about how much work people put into their reputation and social media, their characters and video games, and the work that we put into our reputation, and at work it’s much harder to motivate us. So what can we take from these architectural approaches to guide people’s behavior? What can we take that’s good from it and what should we be leaving behind? These are my guiding questions right now.

6-2. Do you have some suggestions on getting data?

Felipe: I think right now it’s not necessarily about access to a dataset or access to anything. It’s about choosing where you want to spend your time. There’s so much going on. I’ll keep it to these communities that you can enter as an ethnographer. Right now I’m collecting data on crypto communities. I’m collecting data on these communities that are more creative communities, organizing things online in part because of the pandemic. There are so many options for looking at how architecture takes shape now because we are in a transitory period where we imagine a remote work for us. We imagine a workforce that might be hybrid—that part of us is going to be remote, and part of us is not. I think a lot of organizations are looking for help in figuring this out and would be willing to share data from the pandemic on remote work and on what it looks like to apply an architectural approach to helping remote workers remain productive and interested in their work despite the lack of managers.

7. Could you please give some advice to PhD students who are interested in conducting ethnography research? How many projects did you carry out during your PhD career? As you shared with us previously, the Anonymous project took you a lot of time and dedication. How did you balance between this project and your other interests?

Felipe: I get all those fun things because that’s how I can stay motivated. I think we should realize that faculty have gone through what we went through as PhD students. We all have been there, and I still feel the ghost of that feeling of not knowing exactly what it will be received and how it will be received and how to frame things. The discomfort is normal, and it’s something that if you’re not going through as a PhD student, I would think it would be weird. So I think that’s the first thing when I talk to PhD students in general. This feeling of both inadequacy and not knowing what direction things are going is normal and okay, and part of the process. I wished more people had told me that along the way and that it’s going to be okay, cuz to me the best thing about what I was doing is—I was so interested in this Anonymous thing from a phenomenological perspective. I felt it was new,  interesting, weird and something that I would have probably gotten into myself even if I wasn’t doing a PhD. This might be my own flavor of what motivates me—I need to be in a context that is intellectually stimulating and interesting to me, which is why you see me in architecture, wine, hacker and things like that. They may seem disconnected and weird, but the connecting thread is that those are the things that keep me going and keep me interested and motivated. I personally respect people that do all kinds of research inside the traditional work of organizations, but that is not what interests me. I think finding what is that the thing that really drives you both in terms of context and method is important, and building your toolkit along the way will build your confidence.

Another thing that was helpful to me is that I always thought the work I was doing was important from a social perspective and an impact perspective. That made me more motivated. If anything, my advice is to look for the things that keep you going. That way there would always be work, and sleep was negotiable at the time. I was in a lot of different projects: I was working on this while working on the Org Science architecture project, and collecting wine data and doing all that stuff. But that was vacation, right? I was hanging out with my friends, drinking wine; I was paying attention to these online communities and talking to my advisor about architecture, which were really fun conversations. I would love to spend time in a museum learning about architecture, and that’s what I got to do. So it did take over my life, but it didn’t feel like a bad life at the time. Interesting contexts are not just interesting because they really put into relief the phenomena. You can really see all these things, these weird things, in work organizations, and they’re really there—like the symbolic value of wine and the architecture of materiality, and the tangibility of architecture is way more on display than what you see in organizations. The participation architecture in these hacker communities is not just something that is there to motivate people like algorithms in a HR context, it is necessary for their existence. So these are things that to me help put certain phenomena to relief, and then I use some toolkits of theories that I got exposed to—mostly macro but not always to explain them. My learning curve is always steep and I think that’s the nature of ethnography. In this process, finding somebody to emulate that knows the right flavor of bananas and that you want to be is important.

Siobhan: I would love to see us build on each other’s work and terminology as people go down the path. I know that the road to studying really novel phenomena can be super hard for scholars—you have to really believe in it and be passionate about it. The feedback people give you is important and it may hurt and you have to wrangle with it and figure out where to go next in a way that honors your passion and honors the context you are studying. We did get feedback that in our case does not line up with the data. In that case, we say thank you. I think you nailed the problem but I think we’re gonna try a different path to solve it. I would encourage people to deeply engage with the literature, find a theoretical puzzle that a lot of people care about, and then take your novel analysis in your novel setting. Try to not reinvent the wheel, not create a whole bunch of new terminology, but contribute back to that age-old question, in our case, about control. And don’t give up what you figure out. Would you add something, Felipe?

Felipe: Not all of us are lucky enough to have Siobhan in the dissertation committee, and I think that’s part of it, too—find somebody that believes in what you’re doing. Even if it’s not on your dissertation committee, find somebody that shares your enthusiasm for the puzzle 🙂