Interview with Occupy Data member working with Occupy Sandy

Interviewee: Member of OccupyData/Occupy Sandy, male, white, early 30’s.  Chose to remain anonymous.
Interviewer: Max Liboiron
4/18/13

Interviewer: Alright, let me start my backup copy. Alright, so if you could start by telling me your experience of Superstorm Sandy?

Man: Sure. So I actually had the flu the whole week. So I wasn’t sort of on the ground helping. My experiences mainly came after it working with data that was collected through canvassing and volunteer data and budget data and all that from four locations: Staten Island, New Jersey, Red Hook, and Rockaways.

Interviewer: Okay. And who’s data?

Man: Occupy Sandy.

Interviewer: Occupy Sandy, okay.

Man: For all four of those.

Interviewer: Okay. Would you say that you’re part of Occupy Sandy or just a data–?

Man: I think broadly now, yeah. Yeah, I mean, so working with the data, I’ve sort of worked with a number of people who were leading up especially some of the data collection. And we met around February. We we’re still kind of following up then. And that’s–I’ve been in close contact with them since then in terms of, like, working with their data, understanding their data, etc.

Interviewer: Okay. And what’s your job in regards to that data? What do you do?

Man: So this was just part of the Occupy Data NYC Yackathon. So we had–we look at all sorts of data sets generated both by Occupy and about Occupy and about Occupy issues. And I think we had heard that there was some interest in working with this data and sort of finding out what was there and how it could be used for the future. And that’s how we kind of came by it. So it was kind of mutual, both the request and also we were interested in trying to analyze and visualize it however possible.

Interviewer: Okay. And does a lot of the Sandy-related–where does a lot of the Sandy-related data come from?

Man: It’s all site-specific. Each site has generated its own data structures and content. And it contains a lot of canvassing data. So a lot of the sites did surveys of structures and people.

Interviewer: Like door to door canvassing?

Man: Yes. Yes, yes, yes. There’s a lot of canvassing data, which is what I sort of headed up with the data thing. There’s also data on volunteers, including their skill sets, their availability, that kind of stuff. There was data on some sites about the budget and about donations that they had received. And then they’re sort of counting for that.

Interviewer: And those are the Occupy budget again?

Man: This is Occupy Sandy.

Interviewer: The Sandy budget.

Man: Hm-mm. And there was also kind of like infrastructure or hub-related information about the number of items that they have of different types and where to send them and where to go and how deliveries are going to made and that sort of stuff. There might also be one or two other sets of data. But those are the major kind of pools that were generated. And almost–I think almost all sites had something in that regard except for the budget. We didn’t have budgets for some of the sites.

Interviewer: Okay. And is the data from all these different sites commensurate? Like can they–are they sets in that they can circulate like sets?

Man: Independently or together?

Interviewer: Together and/or independently.

Man: No. [laughter] The answer is no actually to both. And that’s what we were trying to work to do. So some sites had–actually a lot of sites still had, especially in Jersey, still had paper forms that they hadn’t even entered yet. So we did–we were helping with data entry for that. Rockaway was the same way. And in many cases, even for something like canvassing data, there would be two or three spreadsheets floating around that were not necessarily even the same format. So we’re working to kind of aggregate it. And like the main thing that we’ve gotten done so far is to look–since I was doing the canvassing stuff–to look at every single survey that was given and try to see what are the questions in common. And I can share this list with you. We have about a hundred and fifty questions, I think, that weren’t on every survey but kind of represent, like, a good swath of what was covered. There was some–we also prioritized them. So we think that these questions are–again, this was future-oriented– So we prioritized them in the sense of these are questions that we think that in future cases people should be asking because there’s they’re extremely important and deal with sort of like emergency safety kind of concerns. These are secondary questions that, you know, could be interesting to know that are not as important. And there are a few tertiary questions that we thought were very site-specific or very kind of idiosyncratic to different surveys. But yeah, we’ve been working to kind of pull this together into one…one data set which then we could itemize and release–

Interviewer: Okay.

Man: –professionally so others can work with it.

Interviewer: So when you say future cases, do you mean the next storm or do you mean other uses for the data?

Man: I think–of all other uses for the data—but when we were talking about prioritizing the questions, we had in mind I  think other coastal disaster kind of situations because things–we realized that things like electricity, water, mold would probably be part of any future hurricane or flood or anything like that. So we kind of imagined it in that sense. I didn’t–we had a brief talk about whether this would generalize to sort of all kind of like crisis informatic situations. But we didn’t think that, like in an earthquake or something, it wouldn’t work in quite the same sense. This was more for, sort of, coastal regions.

Interviewer: Okay. So would it be fair to say, if I paraphrase that, that what you guys perceive to be the biggest immediate problem are what you prioritized for, in the–

Man: Uh huh.

Interviewer: Okay. And what was some of the biggest problems?

Man: Things like ‘do you have running water?’ Is anyone injured? Is anyone sort of not able to leave the house and needs immediate attention? We included mold as well because it’s such a sort of a pervasive and long-term kind of thing, but mainly immediate health and safety issues or if somebody needs to be relocated, for example.

Interviewer: And why did you decide on those?

Man: That was really with the site representatives or volunteers. So that was sort of what issues–if you saw this in a canvass, they would say ‘well these are ones that we would say has to be handled immediately.’ So it was kind of based on their practice of response.

Interviewer: Okay, so what they saw in these places–they’d be like oh my gosh! No one has running water. We didn’t even think–

Man: Yeah. In terms of like case resolution, when they would see the canvass data, they would say, okay, this is a priority case that has to be handled as soon as possible. And then other issues could be handled, you know, a bit later. Like if you need to–if you didn’t have mold but you needed to replace drywall or something, like that would be a secondary case that would be handled later because it wasn’t sort of an immediate safety concern.

Interviewer: Okay. And how do you see some of these niche–or does the data even worry about long-term need as we shift from one type of crisis to another type of crisis?

Man: A few of the surveys did ask for some additional information like would you be interested in community organizing? Would you be interested in volunteering? Would you be interested in information about social and economic justice issues? The kind of community-building type questions–those are ones we either flagged either secondary or tertiary because they were kind of tangent to the immediate issue. There were only one or maybe two sites that were kind of asking that type of stuff. There were a few other issues that would come up at specific sites, and actually questions that we kind of redacted like what’s your immigration status, which we thought that that was….

Interviewer: One, illegal.

Man: Yeah, and very problematic. But there were issues that–I think some of those were an adjunct for things like what languages do you speak or what languages do you need to be communicated in, which was a practical question. So–

Interviewer: Or like what kind of vulnerable are you?

Man: Yeah. There were some–there were ranges of the canvasses. Some canvasses were very—more—what would you call it?– objective or quantitative in terms of check off these boxes if you have water damage, mold damage, whatever. Others were very open-ended. And one question that a lot of people in the group really liked was the question what do you perceive as your biggest need right now. So letting the participants self-define what they think is their highest priority and then, you know, go from there.

Interviewer: And do you see any trends with that data?

Man: [laughter] We’re not quite there yet. There’s kind of–there’s temporality issues. So only one site, as far as I can tell, had a very rigorous case and case tracking system involved where they would open up a case for each issue, that they would have multiple follow-ups on that case, and they were able to see all of this in the system. There may have been a second site that I’m still unsure about if they have that. Many of the sites canvassed once. Maybe then they’d go out a week later. But we had trouble kind of establishing the temporality of the data and whether this is a follow-up or whether this is a first survey, that kind of thing. So we’ve had real trouble seeing trends so far, at least historical—like, time trends. We did publish some very basic, like, aggregate numbers about, like, number of diapers given out, number of water bottles given out, that kind of stuff, which was very easy to get. But in terms of canvassing data, we had trouble even just reconciling all of the data formats so far.

Interviewer: When do you think that’s going to happen?

Man: So we’re meeting on Monday actually to kind of–we have something that we think is a unified data set but we’re not sure. [laughter] So we’re meeting on Monday to test it. And then if possible, start visualizing it. So I think this next–this upcoming week is actually kind of a big week for us. We have two events. So I’m hoping that out of sometime in the next month we would have that release. And that’s on our website, which is occupydatanyc.org.

Interviewer: Okay. And for the record, today is April eighteenth. For the transcriber and the people reading this later. Occupy data–what was it?

Man: NYC.

Interviewer: NYC.

Man: Dot org.

Interviewer: Dot org. Cool. Great. Okay. Can you talk a little bit about your experience canvassing?

Man: I didn’t get to canvass.

Interviewer: Oh, you didn’t canvass?

Man: No, no, no. This was all–I sort of received it after the fact. I could talk, I think, a little bit about people’s experience canvassing, but that’s all second-hand.

Interviewer: Sure. Then what filtered down to you would be interesting.

Man: Sure. Let me collect my thoughts for a second…[long pause]. I think people found canvassing to be somewhat physically stressful and also emotionally stressful because sort of witnessing that much trauma. But I also think that most people really enjoyed it because they got sort of face-to-face contact with people, and they actually felt like they were helping. My understanding is that several–I don’t know what area this was. I can’t remember. But in many areas, the Occupy Sandy canvassers were the first ones to have any contact with residents there. So they were able to kind of reassure residents that they weren’t alone, that you know someone was there to help them or try to help them a little bit. So having that first point of contact, I think, was also rewarding. And for many people, they said they loved canvassing. We loved talking to people and kind of having that experience, I guess.

Interviewer: Okay. How do you think or did social or mass media play into sort of your experience of Sandy or what you’re doing with data or how you might use data?

Man: Sure. Well that was my main experience of Sandy because I was sort of getting it from behind the scenes. I think that it has informed some of our kind of research interests in the data in terms of to what extent–was Occupy Sandy a first responder or what gaps did it fill? And how did community come together to fill those gaps where government and charitable organizations did not sort of fill those gaps? So I think that has been–to make visible that work has really been our driving interest in this. And you know, I think Sandy–Occupy Sandy has received really good press, I think, from what I’ve read, at least compared to sort of Occupy Wall Street a year before that. But I don’t know that–I can’t say I’ve seen a more detailed account of what that was. It was just sort of like ‘its great that people are coming together and helping.’ You know, Occupy has turned into something that’s working or more possible. Or that’s sort of the media impression I have of it. But we haven’t, again, seen specifics or seen, like, a direct comparison, I think, of Occupy Sandy compared to other groups, which is why we’re interested in that.

And I’m trying to remember from social media my perception at the time. The only thing that comes to mind specifically is the sort of the needs hashtag that had come up that I think was effective at both highlighting needs and also getting people to respond to those needs. And I think I was surprised by the effectiveness of the hashtag in doing that because it’s kind of a–a hashtag can be a pretty blunt tool given that anybody can use it for any purpose that has nothing to do with the hashtag itself. And it’s also very limited in terms of how much data it can contain in a tweet, for example. But I was pretty impressed with how that was able to manage so much of the communication load in a very informal way.

Interviewer: Cool. Did any of that data end up with what you guys are working with?

Man: We have tweet data because we do regularly collect tweet data on a lot of things. We have tweet data on it, and I don’t think we’ve looked at that yet. So thanks for the reminder on that actually. [laughter]

Interviewer: We would also be happy to look at that data with you.

Man: Yeah. Let me see what we have on that. I’m trying to remember what point–at some point–I can’t remember. Maybe it was around December, we made a more concerted effort to try to collect tweets live. When we noticed an issue popping up. And I don’t know where this fell on that or how many we have, whether we have a random sample, whether we have everything, etc. So I can definitely get back to you on what we have about that.

Interviewer: Cool. I was recently at a conference. And these people who are not New Yorkers were looking at Sandy-related tweets in the first three days after the storm, and they couldn’t figure out—they were like ‘And contrary, you know, to expectations and popular belief, there just weren’t a lot of tweets coming out of these affected areas in the days after the storm.’ And I put up my hands and was like, there’s no electricity in those places after the storm!  And they’re like, oh. It’s like yes, local grounding to sort of figure out what’s going on with your data is very useful.

Man: Yeah.

Interviewer: They’re like, oh yes.

Man: And that’s something that in a way came up with the canvassing issue, because they had to use paper forms in many instances because there was no other thing available. And there was a lot of discussion of, would we like to see electronic devices used to capture this information digitally so that it wouldn’t have to then be entered. But you know, there was a lot of discussion about that may not be possible and paper seemed–paper was definitely the preferred method because, you know, it didn’t require any of that electricity like the cell or wireless or anything like that. So yeah, that was a really pressing issue. Also, we noticed that people were capturing, like, people’s phone numbers. And we were really curious about whether that was useful or not because, you know, they may not be able to contact them at all. Or some groups maybe have asked what’s your phone number and does your phone work as a follow up, yeah.

Interviewer: Right. I also heard a lot of people who talked about, when they had a clipboard, how it worked as a boundary object that people would gravitate towards them visually because they were– sort of the type of thing they were doing administratively was apparent.

Man: Yeah.

Interviewer: So that’s also something that doesn’t happen with like–with a hand-held like a phone because everyone’s got one and it could be a private thing–

Man: Sure.

Interviewer: –as opposed to a public thing.

Man: Sure.

Interviewer: Let’s see here. So a lot of people are making connections between the aftermath of the storm and inequity, particularly in terms of class is very strong, and then race and gender sort of pulling up after that. Sort of what do you think? Is the data talking about that or that side of the data?

Man: No, but I’m working to put it in census blocks, which have demographic information. So that would be a point of comparison. But we’re dealing with sort of, like very, very small geographical segments, like super tiny pieces of the city. So I actually don’t know how sort of rich that data will wind up being.

One thing I will say that we want to kind of follow up with and we’re going to talk about next week is we’ve been reading–we saw a couple of stories about banks that have been buying up properties that were damaged and then turning them at great profits because they’re–

Interviewer: Flipping them.

Man: Yeah, flipping them because they’re beach front properties, etc. And we wanted to talk about is there data out there on that even in the property record that we could obtain and kind of try to visualize as kind of a continuation of this because a lot of the data we have now is six months old. And not that it’s not interesting–it’s more than six months old. Not that it’s uninteresting, but we would like it to be a little bit more sort of future or present oriented at the very least. And certainly I think it can help inform other people what went on. We would like to look at some future-oriented things, and the home flipping thing seems like an ongoing issue and something that will continue for awhile. So we wanted to do something with that.

Interviewer: So are you–is people no longer gathering the raw data from the ground?

Man: Not that I’m aware of. My sense is that it’s much more informal at this point. So the canvassing happened mainly in, like, maybe a two or three week window in November. And that’s where all of that data comes from. And then maybe some of these sites with case files are still doing updates on the case files if they’re still unresolved ones. But my sense is that it’s become much more informal at this point.

Interviewer: Can you define what you guys are meaning when you say case?

Man: Sure. They were–my understanding of it is that it would be a residence. So a case would be– Actually, I’m not sure about that. It could be either geographic or residence and then issues at that residence or it could be a specific issue like this person doesn’t have water, could be the same as this person doesn’t have electricity. But those are two separate cases. I’m not exactly sure. I believe it’s residence-based.

Interviewer: Just like ‘these people and their problems.’

Man: Yes. Yes. And then follow up on that.

Interviewer: Okay. I was talking to another Occupy data person. They said we don’t have data sets; we only have cases.

Man: Yeah.

Interviewer: And that was the difference. And I think that’s what you were talking about.

Man: I think we’re talking about the same person.

Interviewer: Yeah, it’s not standardized enough to be…to be data sets at this point.

Man: Sure.

Interviewer: Okay. Can you–it seems like, when you were talking about the house flipping data, that you were making this point latently but I want to see if it’s explicit, a connection between data and activism or like data visualization and activism may be in particular. Can you talk about that relationship a bit more?

Man: Yeah. I think it’s–I think visualization is important for raising awareness of many of these issues, and because it makes things that are hard to see or often invisible patterns visible. And it makes them more concrete and able to understand. I think it brings awareness to a much wider audience with that. In some cases, I think it also has more actionable capacities to it. Actually, you know, it’s able to identify specific locations or very specific things sort of analytically that you can target to make an impact. So I think it can be used both on the general sort of consciousness-raising sense but also in the planning sense in terms of what…what to pursue. So if we find, for example, if there’s one particular area that house flipping is sort of rampant in, that could be an area that people might want to do a direct action in. But if it’s a much more distributed geographical issue, then maybe that requires a different approach. But at this point, we don’t really know what that is, but I think visualization can give, some, you know, an understanding to that.

Interviewer: How are you hoping to circulate the visualizations? Or is that even on–?

Man:  So it’ll be–yeah, it’ll be web-based, first of all. At minimum, it will be up on our website. We do distribute through social media. We distributed through some listservs. I was pleased to see that we got a mention in The [New York] Times a few weeks ago. There was an article about Bloomberg’s geek squad, which is about his sort of data miners within the Mayor’s Office. But then, later in the article, someone pointed out to me, it said ‘oh, and there’s a lot of civic hactivism going on and Occupy Data recently convened and started to work with Sandy data.’ So I think that there is some interest in the press about this. So I think we would want to approach them as well as kind of a mass media tool. But I think that since we’re primarily working with the individual Occupy Sandy offshoots that they can be a good broadcaster of their own data. In some ways we’re just kind of facilitating work for them. I don’t think–I don’t see our work as totally independent of them. And I think it’s–we’re just helping them with this one piece of it, and so helping with other pieces of their projects.

Interviewer: You just mentioned the word “hactivism.” Do you guys consider yourselves hactivists?

Man: We hold hackathons. Yeah. So yeah, and we’re actually having a yackathon next Saturday, the twenty-fifth, I believe, to in part discuss issues like ‘what is hactivism.’ Some people talk about ‘civic hacking,’ which seems to have a base, which some would say is apolitical. So we would like to discuss issues like that and see what, you know, people think about that. So I think the yackathon is both to present some of our work and talk about it, but also to discuss some of the bigger issues here about the questions that you raised.

Interviewer: Is it open to the public?

Man: Oh, absolutely. It’s at The New School. And it’s actually listed on our website. I want to say it’s around noon or two. I forget what the start time is.

Interviewer: Cool. I will try and be there.

Man: Great.

Interviewer: So we have a bunch of questions too about climate change.

Man: Okay.

Interviewer: I assume–maybe this is a horrible assumption that you’re on board with climate change; you’re not a denier.

Man: I’m not on board with it, but I’m aware of it. I do not support climate change,[laughter] but yes, I’m aware of it.

Interviewer: Excellent. We have a special sort of set of phrasing questions for people who are deniers but I assume that Occupy Sandy people in general aren’t, so.

Man: Agreed.

Interviewer: Okay. So how–so I’m interested, again, how the data relates to this. Does your data at all interface with climate change-nish like adaptation or mitigation or–like you’re talking about future…future-leaning data, right? And if you are on board with climate change, then you assume this will happen again, right? That’s implicit in some of what you were talking about.

Man: Yeah.

Interviewer: So is there anything that’s climate change-ish about what you’re working on?

Man: Other than the cause of the situation, obviously. I think–the way that I’m thinking most immediately about this is to look more at the data structures. So I think to think of this in sort of ethnographic way and think about how people came up with the data structures and what the data structures did for them and how it facilitated their needs and their work around these issues. So it’s kind of like a meta-analysis of the data and more about the data, again, the data architecture more than the data content itself. I don’t know what in the content would directly answer or directly address some of the climate change issues. And I don’t think that came up on any of the canvasses, which is also an interesting thing to consider for the future too, to ask people, or to engage that discussion with people around it. But I think—I’m–we’re all interested in not only the data itself but the process that we came to all of this with. And I think that the idea is not to have kind of like a centralized system for activism, certainly not. It’s more to, I think at best, to provide a kind of best practices guide in the same–I’ve talked about this in the same way that there are, like, street medic guides or other kind of guides that orient people to some of these issues. I think there could be serious work on some data guides that explain kind of how to set up these things or what kind of principles to consider when you’re, when you’re developing these kind of systems. So I think understanding how the systems were built and maybe talking more to the people who built them. And how those decisions got made if they were intentional at all would be really interesting, and I think revelatory for what we can do in the future and what might arise in the future, and how we can maybe learn from this and improve for the next time.

Interviewer: Do you have the names of some of the people who are on the ground making this, because I would love to go and ask them those questions?

Man: I have them somewhere.

Interviewer: Okay.

Man: I think you’ve talked to one who seems–who I think is probably the most data-savvy and also on the ground, fully in both senses.

Interviewer: Is that Nate?

Man: No.

Interviewer: Okay. I haven’t talked to anyone else yet.

Man: Okay. Okay.

Interviewer: I have things set up but I haven’t spoken to them.

Man: Okay.

Interviewer: Okay. I want to go back to this data structure issue.

Man: Yeah.

Interviewer: Can we talk a little bit more about what you’re thinking or discovering or sort of–yeah, about these data structures?

Man: We started off very generally just asking everyone–and we had almost every–all the sites represented. There was–maybe New Jersey wasn’t but we have their canvassing. And we started off and just said ‘what kind of data do you have?’ And we just threw everything up on a board and then actually quite quickly it clustered into those five or six categories I talked about: canvassing data, volunteers, budget, distribution, and maybe some other kind of hub information. So it seems to me that that’s some evidence that whenever this happens in the future, there’s going to be these kind of data structures. Given that, we can start to think about how they should relate to each other, what kinds of data have other people found useful that you might also find useful, are there particular tools we can recommend to build some of these things. I think it would be great if the data was inter-operable, meaning the sites could communicate with each other and their data. I think that would be good for both sending volunteers around, for sending materials around. But that…that strikes me as quite a ways off, that there’s a lot more kind of people-oriented discussions that have to happen first before we start even thinking about data inter-operability. But I think that even if that last part didn’t happen, the earlier steps would get us so much further than we were coming into this where we had multiple spreadsheets all over, very inconsistent formats even within a site. And I think we’re also kind of responding to a problem within sites where people say ‘our data’s just a mess. Like we don’t know what to do with our data. It’s very hard. We’re working very hard to manage this and to make good use of it, and to respond to these issues as soon as possible. But we’re doing a good job with it, but we think we could do better,’ because it’s just a huge undertaking and a huge amount of work and huge complexities in it.

Interviewer: Do you think–this is, a I don’t know, a provocative question. Do you think there’s such a thing or a genre of data or data structures called, like, disaster data, that it would be different from other–

Man: I’m starting to think that. And I’m actually–a colleague was talking to me about what’s called “crisis informatics,” which is in part looking at data and community relations around crises. And we started to think about how people in our profession could maybe help with that a little bit more and help give what we know about data and what we know about data management really, to sort of lend to that. So yeah, I think that there are–I think that’s what we saw when we looked at all the canvasses, that there really were kind of reoccurring questions or reoccurring categories of questions, and that those would be useful again in the future. Again, I can see it shifting a bit if you’re talking about kind of a coastal issue or an inland issue. But beyond that, I think that most of the things we saw would be useful in the future. There would be a kind of disaster information structure, yeah.

Interviewer: Do you think–maybe another–just based on what you’ve already said, another characterization of disaster data might be this binge and purge of getting data, compared–and then also this intense localization that needs to maybe not be local but is anyway because of how data gets gathered.

Man: Yeah, I think that there–I think it ultimately falls to a few people on the site who are willing to sort of mine over it and handle it in whatever way they can. And we saw some people were–some people were running reports and making maps even of sites that needed immediate attention. Other people were color-coding things. Everybody has their own kind of process for managing it, but it gets managed no matter what. I think it would–I think that it would be–I think that in general, in the public, there’s not quite an awareness that data plays in getting people and things where they need to be in general. So I think that may be an unforeseen thing that is part of any of this. And if it becomes more of a foreseen thing, then people like me and other people who work with data–I think we would be able to sort of help out as much as possible with that at a local level. And I think that that–that might be all that there needs to be. I think some people definitely came into–with bigger ideas about kind of a centralized system for doing this or a software even or a code-based kind of thing for doing that. And that runs into a lot of issues because you have different information literacies in these groups. Some people, you know, are comfortable with some level of database work but others, you know, don’t feel comfortable beyond something like Microsoft Excel. So coming up with a standard system is extremely hard then because it’s likely you’re not going to be able to reach all of those things, which is why I think that guidelines or even just an awareness of this problem, which would then lead you to seek out people who can help with the problem, would be a more practical approach, frankly.

Interviewer: Hmm. And so that’s why you’re advocating this sort of like best practice guide as opposed to a kit, a software kit or something like that.

Man: Yeah, yeah. I think–and again I think I’m modeling that on some of the success of things like the street medic guides. It doesn’t–you know, often you need to receive physical training to work with that. But you can also read the guide, for example, and get a base level of information from it. And I think that that’s powerful. And it certainly raises your awareness of the issues behind it because I think most of the guides that I’ve seen stress more things like principles or background kind of values or decision-making processes–and then yes, you’ll have a specific list of ‘check for these things or do these things,’ but really if you learn the background principles, you know, you’re probably okay in terms of making decisions. I’d like to see something similar in the data realm. It’s just extremely hard because data’s also so new that we, even we don’t know everything that’s going on with it. And it’s constantly evolving, so it’s hard to freeze it at the moment and kind of try to come up with practices. But that would be the ideal, I think, in my view, at least for the moment.

Interviewer: And beyond the sort of categorization of distribution, all that sort of stuff that you mentioned before, what might some of these principles be? Or which ones would you like to be in there?

Man: I think a simple one would be something like making sure your data is cut up in ways that’s filterable in the ways that you want. So if you know that you need to find out which houses have mold, you need a question that’s ‘does your house have mold, yes/no?’ Some of the surveys we saw were very free response kind of things or in the comments, they would provide actually some of the more emergency contact information, which someone may or may not see. So it’s not really on the canvassers to kind of, you know, make up those structures; it’s really on whoever’s making the form or whoever is sort of training or coordinating the canvassers to be thinking about that. So just thinking that, you know, what are the ultimate actions you’re going to need to take on this data and making sure that the questions or the way that you ask the questions or code the questions gets put into that format, I think, is crucial.

Interviewer: It’s interesting that didn’t occur to people.

Man: It did to many but, again, there were some sites that were just, ‘do you have any damage?’ And then you’d get a free response of ‘water damage, mold, whatever.’ And I don’t even know–I couldn’t even say what the paper canvassing said. The person entering data seemed–my guess would be, just looking at it and understanding how people do data entry, is people would kind of roughly standardize it. So they’d see ‘water damage’. And then if you were entering the forms for that day, you would always write “water damage” or always write “mold damage.” But somebody else might say “no hot water” or “some mold.” Right? So–and you can see kind of–I was able to see some patterns or clusters of that kind of stuff, which to me suggests that the people entering the data were thinking in the direction of how do I make this clean and standard, etc, but there wasn’t enough coordination even in the spreadsheet format with columns, for example, to let that happen enough. So I think people are roughly aware of actually the–some of the issues but may not know, from a technical standpoint, the best way to solve them, or may not be empowered to do that, because they may just be handed a spreadsheet that’s already set up in a certain way and they have to just work within that.

Interviewer: Right. Yeah, there is a bunch of literature about how when someone’s given a standard system and they’re confronted with a situation that is not standard, how they basically gerry-rig the form with, like, little things on the side or they’re not–or skipping certain questions or something like that to make it conform and make it more useful, but then how that gets taken back by the standardized system is super difficult, right?

Man: Yeah.

Interviewer: Yeah, yeah. Interesting. Okay. Let’s see here. How do you think your experience of this storm might be different from other people’s experience of the storm?

Man: [long pause]. Well again, I was not–I and my property were not directly affected by it, which I think is the most important difference to note. But also, I think–I had talked to some people who volunteered, not with Occupy Sandy but maybe with Red Cross or some other organization, and were extremely frustrated because they…they felt that all of the response was massively disorganized and were very frustrated that they had, say, food rotting on their shelves when other locations needed it. And why can’t–you know, why can’t they just know that it needs to go over here? And seeing some of the data from just one slice of the response and knowing that even if the slices–you know, there was Red Cross, FEMA, Occupy Sandy, etc. Even if they had their own data internally consistent, it still wouldn’t play together. [laughs] I’m like, that’s–I’m surprised it worked as well as it did. Like I’m actually…I’m actually, you know, amazed at how well the response went and how nimble it sort of was considering, you know, considering the situation. So I think I have a little bit of a different, more of a top-level view of it. And I’m a little bit more forgiving of, sort of, on the ground frustrations because seeing the top-level architecture of it, which again was quite good under the circumstances, but still very difficult to work with and very hard to communicate, very lack of communication channels, I think, between these things. I’m actually surprised how well it went. And I’m much more forgiving of some of the on the ground frustrations.

Interviewer: What do you think accounted for some of the nimbleness?

Man: I think the local organization, frankly. The fact that I think people had sort of just face to face communication and were able to make decisions sort of on the spot and didn’t have to kind of contact a hierarchy or anything like that, were able to just see a problem and respond to it or see a–not a specific problem, like a particular house, but just a general problem about how are we going to get water for these people, and were able to kind of come up with a plan on the spot and do something. So I actually think the local organization is extremely important, in my view.

Interviewer: When you say local organization, you mean?

Man: Each site being able to sort of run itself.

Interviewer: Okay. So like Occupy Sandy as opposed to FEMA or–?

Man: Yeah. Not that I know much about how FEMA works, but I assume that there is much more of a hierarchy there and much more of a protocol and a bureaucracy there. I did hear also about a lot of–from a policy standpoint, for a minute, things like–so we had Hurricane Sandy, and then we had the nor’easter. So I heard stories about, like, Red Cross coming in for a day, setting up shop and then having to withdraw because the nor’easter was coming. And I assume their policies prevent them from being out in the field under certain conditions. And I think that was very disheartening for the residents and also, you know, created this space also where sort of Occupy Sandy or activists had to step in to kind of make up for that deficit. And I think we weren’t bounded by some of those policies. So I think sort of the autonomism of it, letting individuals choose whether they wish to go out and canvass, you know, under these conditions, I think, has a lot of benefit to it if people are–obviously if people are willing to do it, whereas some of the structured ones, for legal reasons, wouldn’t allow them to do so.

Interviewer: Yeah. I remember the photograph of the door that said “FEMA closed due to weather” because it, like, circulated like crazy. Like it was everywhere. You didn’t even have to criticize it; it was just self-evident. Did you join Occupyness because of Sandy and Data or were you already part of that network?

Man: Yeah, I was already part of it.

Interviewer: Okay.

Man: Yeah.

Interviewer: Okay. And then, the last main question is when do you think the sort of crisis is over? Like when do you think recovery has happened or is happening?

Man: I think it’s deeply connected to the economic issues. So my answer is actually not never but quite a long time, until those deeper issues get resolved, because I think–you know, I talked to other…I talked to other people who have families in the area, for example, who’s families have stayed during the storm. And you know, I think an initial question is why did they stay? Why didn’t they leave? And they said, well you know, they’re too worried about looting after a storm and this is all they have. So they don’t want to leave it. And I think that underscores some of the structural inequality issues that lead to that mindset and create that situation. And those, you know, are exacerbated by the storm and don’t go away when the storm goes away. And even if the immediate kind of property damage issues are resolved, which in some cases I think they’re still not resolved, those deeper issues remain. And it’s still an extremely vulnerable area. So you know, I think it’s quite a long time before the aftermath of that is over. I think my big concern is that we’re now in a time when this is the first of what will be many occurring incidences. And I think this is obviously part of why you’re doing your research too. And that concerns me a lot because I don’t see it getting better any time soon, frankly.

Interviewer: So what you said maps on really well onto what a lot of Sandy and other activist organizations are saying, is that the crisis is ongoing and the storm is a punctuation mark on a longer crisis that happened before the storm and continues to happen.

Man: Yeah.

Interviewer: And so what’s interesting to me is that it sounds like you have fairly specifically storm data as opposed to crisis [data]–

Man: That’s right.

Interviewer: So like how are you thinking–or do you even think about how to reconcile those two things?

Man: I love the–I actually–I love the kind of canvassing idea. I think it’s a really powerful tool because it puts people on the ground. It gets them talking to people in the neighborhoods. It brings out issues that maybe we didn’t see or didn’t know about or didn’t have documentation of before and then can broadcast. And so I would actually be curious to see if that canvassing model can be generalized for several other purposes, both in these areas and in other areas. And I mean frankly, at the moment, the only one who’s ever–the only time–pretty much the only time this ever happens is with census data, maybe with some polling where someone comes and asks you what you think. But that’s in a very, very narrow, very narrow sense and very insensitive to different situations and geographic locations because it’s completely standardized across the board. So I would be interested—as in generally like with the data guide, can we talk about the value of canvassing or–I don’t know if that’s the right word but the value of canvassing or outreach, if you want to call it that, or data collection. And use it for a number of other issues in other areas as well.

Interviewer: Do you think there’s any plans for that right now? Or is that something you just thought of?

Man: I–it’s not something I just thought of. It’s something that I thought of about why would we need a guide. And I thought the guide would be bigger than just sort of this crisis management or storm disaster recovery. But I think it’s a way of thinking about doing activism that kind of we have to put out there. So I think it would have to be a much bigger process and a much bigger conversation with lots of groups before it really caught on.

Interviewer: Hmm. I like the idea of data canvassing and collection as a new form of activism. It’s not something that I’ve heard sort of out there very much.

Man: Yeah. I mean it strikes me as, you know, in some sense as governmental but not in a creepy sense or a bad sense of governmental [laughter]. Because, you know, a lot of the government data is used for planning or is used for action. And I think that’s the intent of things like census, right? But it doesn’t address important and specific issues enough. And I think it’s not–I don’t think we always have to throw out the entire mechanism. I think there are good parts to the mechanism. We should keep them and use them for ways that we want. So I can see real uses for having that kind of model. And then, you know, rather than having a constuct–again, rather than have a constructed notion of what people need or what the situation is, going and asking them what’s happening and getting out that–their self-understanding of the issue, I think, would be valuable in itself.

Interviewer: It sounds like that would be the difference between the creepy government mentality–like top-down versus bottom-up data creation.

Man: That’s part of it.

Interviewer: Yeah. Have you read “Seeing Like a State,” by Scott–

Man: No, I haven’t actually.

Interviewer: It’s about that sort of thing where one of the sort of tools of state-craft is to get data, but get it in a way that’s legible to the accounting systems of government. So like taxes, real estate, deaths. And so it’s like a systematic simplifying of thing–of complexity–so that it calculates, into the things that the government is already interested in. And it sounds like what you’re saying is, like, yeah that, no. We want messier–we want to deal with the messier sort of stuff and figure out where the mess is. Alright, well that’s all of the questions that I have for you. Is there something that I didn’t ask that I should have asked?

Man: Not that I can think of.

Interviewer: Great. Do you have any questions for me?

Man: Well, I do want to know more about what the lab’s doing, where it’s going. I’m very curious about it.

Interviewer: So we are–I mean–

Man: How did you start? How long have you been working? That kind of stuff, yeah.

Interviewer: So we started partially as a call from the Institute for Public Knowledge for funding about Sandy. And it was basically a couple of people who were really invested in climate change. And basically one guy’s investigating climate change and me because I invested in research activism as a model for doing research. He was like hey, this is an opportunity to do something –to make those things come together really seamlessly. And so we recruited a bunch of our, basically, friends, people that we knew. Almost all of them are sociology graduate students. And then there’s me, which is not. And then, like, now we have a couple of other people as well. And they came together and applied for this grant, and through the course of doing things realized that through either living in coops or being part of Occupy or other sort of things, that we all knew process. We all knew how to do distributive decision-making. We all knew how to do facilitation in a non-hierarchical way. And we we’re like this is clearly a chance to do something different as far as how research goes. So we decided to become a mutual aid research collector. We did a little bit of reorganizing to make sure that the people that were in the group–that was their value system–

Man: Sure.

Interviewer: –which involved some slightly unpleasant maneuvering at the beginning. But now it’s amazing. Like I love going to work on Mondays being like, hey guys, what do we think about distributive authorship today? And that’s what we talk about. And so right now we’re still basically in data collection. A good two-thirds of the people involved are still doing coursework. Right, they’re at that stage in their graduate work. And then I’m like a post-doc, so I can basically do what I want whenever I want with whatever money. And so we’re sort of doing a lot of maintenance work, laying a lot of ground work, getting a lot of structure in there while we’re gathering data. And then starting in the summer, we’re just going to–we’re going to do the analysis sort of part of it. We’ve already done some analysis and actually got an environmental justice award for it—for doing sort of what you’re talking about, making certain things legible that aren’t legible because as far as we can tell, we’re one of the only groups doing not community-based research but cross-stakeholder research, which is actually pretty rare for activists to do, right? Most activists–like the model for activist research is to do community-based research and basically advocate on behalf of a very specific set of people. We can’t do that if we’re talking to the Mayor and to Coney Island residents, right? So it’s really interesting that way. We hold–have held, and continue to hold what we call “salons.” I don’t really like that name because it’s very French and frou frou, but I don’t know. We’re–like people like you and other people we get together and are like, hey, what are your guys’–what’s going on?

Man: Okay.

Interviewer: And the last one we did was extremely academic. And we want to move away from that. Like it turned out a lot of activists were also graduate students. And it turned out all of us except for one had a university affiliation. We were like, uh, okay, let’s try this again. We’re looking at publishing Op Ed pieces, laypeople pieces as well as academic papers and books. We’re looking at putting together symposiums. We’re very invested in mentorships. We have undergraduates that we work with now who have internships that are actually just full-blown memberships, right? So they’re learning all these techniques. So we’re also trying to change, like, the material things that are going on in the world with our research. We’re going to do a white paper for the government because it looks like the government is saying ‘we did a really good job. We just have to do a little bit more a little bit faster next time.’ We’re like ‘no, structurally we think there were things wrong and there’s what they are’. And here’s the fundamental difference between the grassroots and the government response and how they don’t have to be grassroots and government responses if–because they–I mean they will be, but you can blur those lines a bit for better effect. Yeah, so that’s sort of what we’re up to. We have a very open–oh right, so we’re looking at sort of changing the materiality of structures on the ground for New York City, but also in academia, right, changing what it looks like to do research with actually robust mentorships/internships where they get paid as much as we get paid and distribute–like applying for grants as a group and then distributing those according to needs. And kind of like a socialist grant–

Man: That’s great.

Interviewer: –making or something like that. So it’s going really well. And it’s really exciting. And we are hoping to produce papers about how we operate as a group as well as a sort of best practices guide for activist research.

Man: Yeah, because you’re putting forth a new model of a collective really, of how this works.

Interviewer: Yeah.

Man: It’s really interesting.

Interviewer: Because the only other way that I’ve seen is community-based research, participatory action research, where it’s like the messiah and his or her community, and the messiah is a very facilitated model, non-hierarchical but it’s still that model. And I haven’t seen anything being like no, community-based is not the only form. And that’s exciting. So yeah, that’s what we’re up to.

Man: That’s great.

Interviewer: And we’re going to be up to it for at least a couple of more years because that’s what we have funding for, so.

Man: Okay, great.

Interviewer: Yeah. So I’ll keep you–let me turn this off now.

Man: Sure.

End of recording

Download Transcript here.

Occupy Sandy Open Data here.

This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License. This data may be shared, copied, distributed and used free of charge provided you attribute the source of the data to Superstorm Research Lab within the work you produce.If you alter, transform, or build upon this work, you may distribute the resulting work only under the same or similar license to this one (Creative Commons, CopyLeft, or Open Source), meaning the work must remain free, open, and the data must be attributed to SRL. www.superstormresearchlab.org. An example of this attribution is available below. Members of SRL are co-owners of this Creative Commons license. Any of the above conditions can be waived if you get permission from SRL.

Example Attribution: “Interviews [for your project] was collected by the Superstorm Research Lab, a mutual aid research collective. For more information on how this data was obtained, see www.superstormresearchlab.org.”