Mapping Detroit’s Information Ecosystem


  • From left: Daniel Hewett, Garlin Gilchrist II, Amanda Noonan, Erica Raleigh, John Webb.

    From left: Daniel Hewett, Garlin Gilchrist II, Amanda Noonan, Erica Raleigh, John Webb.

  • Daniel Hewett, Rhode Island School of Design

    Daniel Hewett, Rhode Island School of Design

  • Garlin Gilchrist II, Civic Community Engagement, City of Detroit

    Garlin Gilchrist II, Civic Community Engagement, City of Detroit

  • Erica Raleigh, Data Driven Detroit

    Erica Raleigh, Data Driven Detroit

  • John Webb, Google Civic Innovation

    John Webb, Google Civic Innovation

  • Amanda Noonan, Internews Center for Innovation & Learning

    Amanda Noonan, Internews Center for Innovation & Learning


Garlin Gilchrist II
Deputy Technology Director for Civic Community Engagement, City of Detroit

Amanda Noonan
Director for Research, Design & Learning, Internews Center for Innovation & Learning

Erica Raleigh
Director, Data Driven Detroit

John Webb
User Experience Researcher, Google Civic Innovation


Daniel Hewett
Executive Director of Research and Partnered Projects, Rhode Island School of Design


William Hoffman
Head of Data-Driven Development, World Economic Forum

Presented in partnership with the World Economic Forum and Internews 

A robust and dynamic, inclusive information ecosystem is critical to the ability of any city to serve its citizens. But mounds of data alone are not enough; humans and their communities are still actors in the system. How can we design systems that are truly of benefit, trusted and usable? How can a human-centered design approach be applied to establish an information ecosystem that becomes an adaptive network of inclusion, identifying needs, solving problems and serving communities in a more trustworthy and innovative manner?


Hoffman: Hi everyone. Welcome. Just from a coordination perspective, this is the Mapping Information Ecosystems. So I know there’s a variety of options, but that’s what this session is about. Just to set the stage a little bit, my name is Bill Hoffman. I’m with the World Economic Forum, and in conjunction with Rhode Island School of Design and Internews, we’ve been collectively doing what you would call a global conversation. So we’ve had conversations similar to this in Dubai, Capetown, and Davos in a variety of ways, and today it’s Detroit.

The main point, from a kind of macro global perspective, is to really start to elevate and raise the importance of local context, local understandings, and really understanding how little we really understand despite all the data that’s out there, and so how can we recognize the questions we’re not asking, what are some of the assumption sets, what are the blind spots, and what are some of the processes and design tools we can use to elevate and give us those insights.

So with us today to lead us in the panel is Daniel Hewett. He’s the executive director of research at the Rhode Island School of Design, a fantastic guy, and I’ll let Daniel take it from here.

Hewett: Welcome. We’re glad to be here. I think this is a topic that we all think about a lot and it’s a great opportunity to share some thoughts, but also to get your thoughts. Today’s discussion will really focus on Detroit as a sort of grounding region to think about this issue of the importance of context. I think the reason I’m here or involved at all, is that we at the Rhode Island School of Design take this very seriously in the way that we train designers and artists. And specifically, we spend a lot of time teaching people to see, because once you’ve seen it, you can’t forget it. And if you can’t forget it, it becomes operative in your design.

So I just wanted to quickly illustrate that by describing an exercise that I do with some students every year. And so what I do is I send them out into Providence, Rhode Island, where we’re based, to look at parking lots. And on the surface, like data can be, the grid of a parking lot is exceedingly straightforward. It seems explanatory of what goes on in that space. But what I tell them to do is to go out and come back explaining what goes on in the parking lot. And they go off in pairs. There’s generally ten lots, and what I tell them to do is to take a cup of coffee with them, and they don’t know why. So they stop at Dunkin Donuts and get a cup of coffee. They get to the lot and as soon as they get there, what happens is there’s a person there who engages them and it’s the person who sits in those shacks that we all are familiar with at surface parking lots around the country. And it goes from there. And they’re there for the whole day and what they learn, of course, is that the lines on the pavement don’t explain that much and that the reason these lots work is because those people are so critical to not only that lot, but really many things having to do with that whole area. So the students come back very excited about this and their discovery, and they know the person’s name, they know something about their family generally. But one of the things that I think was really an interesting insight this last year was they came back and said that—I asked them how long the people had worked there and of the ten lots, the person with the least amount of tenure on the lot was 24 years. And as an indicator of the kind of value that contextual information can add to top down datasets, this is really important, because it’s not just a sort of a-ha. It changes the way you think about all the data from then on, because these people have such a pivotal role in that community. So just as an illustration, this is the kind of thing that we do a lot of, and it’s the kinds of things that these folks have done a lot of as well.

We want to focus on Detroit, and so why don’t we start with Garlin? Garlin Gilchrist is the deputy technology director for civic community engagement for the City of Detroit, works very closely with Beth and the mayor on many things that we’ve heard about already today. So Garlin, I’ll just pass it to you and give you an opportunity to say something about what you’ve been doing.

Gilchrist: So thank you and thank everyone for attending. Welcome back to Detroit for Techonomy 2015. So I am very excited to be here to talk about Detroit’s information ecosystem, and more specifically, to talk about the city’s evolving role in that ecosystem. So Detroit has been a place that has been data rich and information poor for a long time, but one of the reasons that that was true is because that data that existed was not very well organized for, you know, decades or even a generation. And we had this great organic movement of people and institutions that decided they wanted to do the work to organize that information several years ago and one of those people you’ll hear from is Erica Raleigh on this panel here, from Data Driven Detroit. It was a very important institution in sort of organizing that information.

Well, once Mayor Duggan came into office, once he appointed Beth Niblock the CIO of the City of Detroit, they really set a policy goal and standard that said that we’re going to make the City of Detroit, capital C city, the producer of so much important information that enables so many different tasks and institutions and organizations to function, we’re going to make the city a real player in that environment. So what that’s meant was establishing a policy for open data and transparency that meant that we were going to do all the work that we could possibly do to make as much city information, as much of the digital exhaust of city operations, available to the public at no cost, in a timely, consistent, reliable, and repeated manner. That would, we thought, fundamentally make that information landscape much better. What it did is it changed it. It said that now rather than organization and individuals having to rely on a combination of personal relationships or individual data jackhammers that they would take to a particular department’s information, now they could start to rely on the city to be the real source of this authoritative information and then they could use that information that we put out to add value on top of it to do the analyses that the city needs to make better policy decisions, to do the analyses that our community leaders and groups need to make to decisions for their own local neighborhoods and such like that.

So we’re really excited as a city to really be a partner to more organizations and more people in a bigger way that we think can really enable much richer conversations that start from a place of shared truth, much richer interactions between city government and organizations, as well as entities and individuals with one another that don’t involve city government, that just uses the data. We really think that this is one of the foundations of a great future of innovation and partnership for the City of Detroit.

Hewett: Terrific. Erica, your role?

Raleigh: We started about seven years ago, actually. It’s really fascinating, I was just reflecting back. I started in February of 2009. The original incubation grants for Data Driven Detroit were granted in October 2008 and we really didn’t get legs until the summer of 2009. But we spent a lot of time for many years doing exactly what Garlin described—you know, I love the personal jackhammer—you know, sitting in city departments with hard drives like, “Please, can we please have your data? We want to do good things with it and we want to give it back to you in a way that it’s going to be more useful to you and to the community at large.” And it’s just been, honestly, just a tectonic plate shifting in Detroit over the last two years, and especially in the last year, as the open data policy was put into place by the mayor, and as Beth arrived and Garlin got here and they’re actually putting data out into the world and I don’t need to spend hours looking for it, cleaning it, getting it ready, all of that, and we can spend more time doing the analyses that we want to be doing at Data Driven Detroit to make the data useful to any community, any resident, any organization, any municipal government, Detroit and beyond, who wants to use information to make better decisions for the community as a whole. So that’s what we’re here for and now we get to spend more time doing the good work we want to be doing.

Hewett: Amanda, you do this work on an international scale.

Noonan: Yes. I work for Internews. We’re an international—we used to describe ourselves as a media development organization supporting healthy local media, usually in difficult to access places where for whatever reason there was difficult access to information. But as the information technology space has expanded enormously, globally and nationally, so too have our activities to kind of take advantage of that opportunity. And I suppose central to Internews’ ethos is that information changes lives and it’s really critical for individuals to get the right information in the right form at the right time to make good decisions about their lives and their families moving forward. And that is a universal truth, if you like. It applies here, but it also applies anywhere where people live.

And I think actually the discussions this morning have surfaced that in many different ways. I lost count of the number of times people said we needed to get that information or people needed to see the information. We talk a lot about the technology and the communication, but we don’t talk about the actual information that needs to go alongside that so that people can engage and actually the information has some kind of impact, or the activity has some impact or the tool has some impact. So information is a really critical element.

And I work in the Center for Innovation and Learning at Internews, which is, if you like, the sort of innovation space for Internews, and we became very concerned and interested in mapping the information ecosystem—it’s the title of this panel—which was to look at the whole complex network that exists of information production, consumption, and movement. The way things are now, anybody within the ecosystem at any point could be any of those three actors, depending on their role. And what often happens is that we see it as very much a liner process that information is provided, news and information, for example, someone reads it, and then they act or don’t act. But actually, these days it’s a very different dynamic. It’s very interactive and also something where, if you like, the roles that people have, as I say, is sort of changing and moves and so we need to design tools that actually take account of the needs and challenges of the different people or the different organizations at different points in the ecosystem. It’s not just good enough to say we need to produce something to get to X. Actually, the challenges of the producers need to be considered in how you design something, the challenges of the technologist need to be considered. And so looking at how we might understand and map that, we started to use more human-centered research approaches to surfacing those needs and challenges. And so a lot of our work has involved building local capacity to really try and understand the context in which each information ecosystem exists, because that is really critical to the questions and the challenges that that community faces.

And I think that above all, our main concern is, as a researcher, I often have—I can see a need in a community: this community needs X—maybe in health or in education or for women—this is a way to get this information to that community, this is the information they need. But actually, when I go and spend time with the community and I observe the community and I hang out with the community, I begin to see that it’s not quite as straightforward as it may have seemed on first look. And so we begin to surface what the real challenges are of that community. And then with the community—this is really critical. It’s not something that’s done to a community. It’s something that’s done with a community. Together, we design the tools and supports that help them have sort of an impactful and, the word that was used a lot earlier today, a trusted solution. It’s really important that we understand where trust lies in the information ecosystem and how to leverage that as a way to communicate and foster impact.

Hewett: You raised something that has come up in our conversations, which is this line between data being an objective tool, a record of truth that’s a record of a system that’s latent, and a tool, a mechanism for change. John, your work has looked at what is the threshold that pushes people over the edge to become change agents or actors in the environment. Would you talk a little bit about that?

Webb: Yes. So I work on Google’s Civic Innovation team. I’m a user experience researcher. Basically, our mission is to organize the world’s election and civic-related information and make it accessible and useful for people. We also—you know, so use that on Google search and other products and so forth, but we also make that publicly available through the Voter Information Project and through the Google Civic Information API. And over the past year and a half or so myself and some other research colleagues have been looking into what motivates people in the United States to do things that are civic and what holds them back, even when they have something that they really care about: What are the barriers and why are they not taking some form of civic action? And so we published it actually a couple of months ago. There’s a blog post, but there’s a link to a full 46-page detailed report on this research and it’s called “Understanding America’s Interested Bystanders,” and speaks to a segment of the US adult population that we call interested bystanders that we estimate is about 50% of the population. These are people who are paying attention to the world around them, the issues and so forth, but they’re not taking action in any sort of regular way.

I’ll sort of pause there, but happy to talk more about the research and the motivations and some of the barriers and so forth.

Hewett: Yes, I think one of the things that’s most interesting about the work and where I think you want to go with it next is looking at the stratification within that of different demographic groups, different levels of commitment, what it takes to pull someone over the edge. So yes, I would love to come back to that.

One of the things that’s also come up in our conversations is the relationship between top down data sort of activism that the mayor’s doing and really using this to advance an agenda and what’s going on, on the ground. And we’ve talked about this as sort of feedback loops. Where is it that the mayor learns what’s going on, because the information is flowing effectively the other direction. And I’m interested in Erica, and from both of your points of view, what you’ve done to increase that flow. Certainly, the mayor comes into office and has a big agenda and I would think that the idea is that it’s moving in one direction. He wants to do a lot. What happens to get people communicating, trusting that information will reach back to the mayor? We heard of a few sort of technologies this morning, but I’m wondering what else you all have done?

Gilchrist: Yes, so one, the inclination to want to listen is really where that needs to start, and I think Mayor Duggan did a great job of coming in and wanting to listen to what was happening and really understand people’s experiences as a way to start designing the things that he wanted to see improve in the City of Detroit on behalf of Detroiters. We talked about, in some of the earlier sessions, both Beth and the mayor mentioned it, the Improve Detroit app experience that we have, which enables people to be able to report a set of city service requests, so things like if there’s a tree that fell and is blocking a street or a sidewalk, if there’s a pothole, if there’s someone that’s dumped illegally tires, or an abandoned vehicle or a broken streetlight, people can use one of three way to report those to the City of Detroit and then trust that those reports will be received, accepted, and acted upon. They can use a smartphone application, they can use a website, but they also can make an old school phone call, and that’s really important when we think about how the city can really make its ears bigger to listen better so that we can inhale or take in more information. It’s that we need to make sure that as many people, as wide a swath of Detroit as possible have access to the technological improvements that we’re making as far as service delivery. So with Improve Detroit, you can also call a phone number in the City of Detroit and we have some staff people who are trained to be able to receive that phone call and put that service request actually into the Improve Detroit application so that the operational benefits that we’ve had from changing that process so that people know that their services will be delivered, they can benefit from that, but they don’t have to have a smartphone to be able to do it.

So this making sure that whenever we’re thinking of a solution, we’re thinking all the way through how that solution, not only at design and how it works, but how it’s delivered and made accessible is really important. And having an inclination to listen is where you start to take that into your design process consideration when thinking about how are the real human beings, the real Detroiters going to interact with this fancy innovation that we’ve made to make sure that their lives are actually getting better. Because at the end of the day, you know, as a public servant, we are about making the lives of our residents and constituents better and if we’re successful, then we’ll be able to continue to have Detroit be a great place to live.

Hewett: Erica?

Raleigh: I think it’s really important to think about too the sort of—so you’ve got the city side and I think it’s absolutely essential that the city is being thoughtful about reaching every single citizen that resides here. In terms of our civic tech community, and also as a sort of public data utility housing data, not only—you know, bringing in data from the City of Detroit and other municipalities, but also bringing in data from many other places—we need to take it into, you know, actionable information so that people can have the impact they want to see in the world, and working with community to make that happen is so critical and it sometimes get left behind I think in civic tech land. So we are always trying to be as thoughtful as we can be about that. We’ve done some participatory research and analysis. We’ve also worked to provide an app so that people can combine all of the data that we have at D3 with community-collected data so they can ask the questions that they want in their own area and then analyze the information together with us to then go to the mayor or the police chief and say, “Hey, this is what’s happening in my community and this needs to change, this isn’t okay,” or, “This is the plan that we want for our neighborhood. This is what we want it to look like. Here’s how we can get there. Here’s how you can work with us as a community as a block group as a neighborhood organization,” so that people can participate in that process. I think correcting information asymmetry is just critical in all of this, and so we just want to be here, being thoughtful about how to provide that information and the ability to wield it wisely. And we’ve been working with Garlin and Beth, the CIO’s office overall, to put on these summer data workshops to improve data literacy and actually taking that information from our open data site, the city’s open data site, putting it together into how do you actually use it then to start making change in the community.

Hewett: I think that’s really interesting. I think one of the things that we talked about this morning is, you know, the pothole gets fixed and assumes that that’s the last bit of contact between the city and that particular resident is the fixed pothole. And I think one of the things that you’re saying that I find particularly interesting is that there’s an ability to understand the meta in this and recognize that there are people, those people who actually spoke up through their app are voicing a willingness to participate. And to John’s work, and I know to Amanda’s as well, that level of inclusion seems like a huge opening to engage that community at the next level. And I’m wondering if some of this work takes that further in terms of understanding who those people are and engaging them further based on their participation.

Raleigh: Such a great point. It’s very meta. I don’t know that we’re there yet.

Gilchrist: I mean, I would hope that—there are many ways that you can count and measure civic participation. But one of those things that it’s founded upon is actually belief in that system, belief in the government that you’re participating alongside or participating with. And so a lot of the work that we’ve been doing in the city is to reestablish or establish that trust, re-baseline people’s belief that the city is listening. And I think once you understand that a system is going to be responsive to you, I think that increases your incentives for engaging with that system. So I do think that these things matter. I would love to hear if there’s been research done to back that up.

Hewett: Yes, John, does that square with your experience, that feedback does in fact lead to participation?

Webb: Yes, but it’s a challenge and it’s super hard to measure, you know, the effectiveness of different civic interventions. I mean, you can—you know, voting is obviously something that can be measured, but, you know, in terms of whether people are attending more public meetings or volunteering more and so forth as a result of the things that the city is doing, it’s really hard to measure. I think one of the big challenges for engaging people locally is sort of an awareness of what is even going on in their community and with kind of the shrinking of the local media newspaper landscape, it’s harder and harder for people to even know what’s going on in their communities. They can certainly report service related issues, you know, potholes and streetlamps are out. But why should they actually vote in a local election or what are even these offices that they’re voting for, what do these people do? Who are these people? What’s on my ballot? And without sort of local reporting and that knowledge and awareness—I think people are interested, but, you know, they’re getting information from other sources these days, oftentimes social media, Facebook, and bypassing what little local reporting that we have at the community level. And so anyways, that’s more of a challenge, but something I wanted to call out.

Noonan: I was going to say, I think also—well, I think this calls out the notion of actually trying to understand the information ecosystem before you even start to design your system for communicating with communities. Different communities will communicate in different ways. And so really understanding the nuances and the dynamics of those systems and then working with the communities to design those trusted pathways and networks, really with them onboard and actually, hopefully, if you can, actually building capacity within those communities to kind of keep that ball rolling and keep that information flowing. Because also, it’s great that people are providing information about a pothole or a light or whatever, but actually that, as you rightly said, is creating a pathway which could then be used for so much more information and communication. So building that capacity to be able to give people the sort of networks and the tools to do that would be very useful.

The other point that sort of comes to mind when I’m hearing these conversations, when you have started sort of from the outside in as opposed to the inside out, if you like, a more inclusive approach, is that it is a little bit exclusive in that there are always going to be outliers. There are always going to be members of those communities who are not included or engaged in any of these processes. So ways to try and communicate with those individuals, even if it’s not by a direct connection, there may be indirect connections that we can start to leverage and design for. So I think, again, it’s that mapping piece that really understanding how people really communicate, what information they get from different types of sources, how they trust those sources, who or what is influential within those communities, and then using those nodes as design—

Hewett:  I think that’s fascinating. And I think that’s exactly the kind of work that I think we—I can say that my team needs to work on. Because formalizing what you described—the grandfather network, is that what you called it, or the grandmother network?

Gilchrist: Yes.

Hewett: Garlin described the fact that his grandfather is not wired up and so he relies on him to make the moves into the data community for him and then report back. And while that’s a lovely illustration, it’s probably happening at a scale that if well understood could actually be used, to Amanda’s point, in a way that it is really rewarding for your grandfather and others like him.

Noonan: Because also the grandfathers—sorry, I don’t mean to interrupt—the grandfathers may have a very different perspective that is being entirely missed from the current view that we’re getting from the data that we’re receiving from the community and the data that we’re giving to the community and how it’s analyzed. And it think that’s a really critical point that actually, as you said, you know, the questions that the community themselves want to ask, the data that they themselves want to collect is a really critical element in this ecosystem and it’s something that can easily be overlooked because we’ve got a core system that works. You know, I don’t mean to say it in the pejorative sense, but it’s like we need to kind of like really benefit from what’s already being done.

Hewett: I think one of the other interesting things for me about that is that grandfathers have a way of saying things that an app can’t communicate and may open up insight into what improvements to that app might do. I don’t know if—

Gilchrist: That’s totally right. And I think one of the things that we as a city are going to move into more in this year—so I like to think that the last year we really have been doing a lot of infrastructure building, as far as sort of how we are interacting with this technology community here in the City of Detroit, but also just how we’re managing data and information internally, what processes we have around that. And we’re going to continue to do that work, because there’s plenty of it to do. But in this next year we’re going to really be thinking harder about what do those community engagement aspects look like that not only talk about training as far as what are the resources that we are making available and how can they be utilized, but also flipping that on its head and saying, you know, here is what we have, now let’s have a community conversation about whether or not that reflects and describes the experience that you have for the person who lives here. And that is a way to potentially reach some of those folks that can give feedback in a much more direct, much more succinct, much more actionable way but do not have access to some of the channels that we may have made available before. So I’m really excited about that, because, again, I think it’s about how our ears can be larger, how we can hear more from more people, from more types of people. It’s super important.

Hewett: And more qualitatively, absolutely. Yes, that’s lovely. I was wondering if you all would reflect for a minute about what makes Detroit unique in this. It’d be easy to slip into an idea that we’re talking about a sort of prototypical city and that you guys have made great headway, Chicago’s making great headway, but that it’s easy once you know the formula. But certainly, you’ve had to adjust whatever you already knew or have learned to this particular environment. And what I’m thinking about in particular, just to get us started, is geography and the incredible scale of this city and how spread out it is and how some of that may relate to data access. I’m just wondering how you see Detroit as special and how you’ve responded.

Raleigh: Okay. So Detroit in terms of land area is not actually that large in relation to other cities across the US, especially as you move further west. There are other cities that grew like we grew. We’re actually kind of an oddball in terms of the older industrial cities, but really we’re not that special in that way. I think one of the interesting things is you kind of have that mix, that big, sprawled out—I think, was it Mayor Duggan earlier that was talking about the single family homes and just because of our unique development history in terms of the timing and the middleclass that was grown here, we do have a lot of single family homes, which most older industrial cities don’t have that sort of composition of things. And then the regional aspect here is a bit—it’s similar across the country, I think. That’s the thing is that we’re sort of the bell weather for what is coming at a lot of other cities and it’s just at a scale that is almost unimaginable for most places. You know, Cleveland has similar percentages of the troubling aspects of Detroit. It’s just, it’s so much smaller that it’s not that much to try to fix it. So when you look at something like blight that is actually caused by this housing assembly line that’s been created by overproducing housing across the entire region for decades, it makes sense that as we grew out, as we developed new housing all along the fringe, we’re actually literally sucking population out of the City of Detroit. It makes rational economic sense for every individual to move further and further away from the core. Jobs move out, new housing goes up there. It just makes sense to move out. And we’ve left behind this very unique place with a lot of green areas or vacant lots that are now, in most cases, very overgrown, are like reverting to forest, or a lot of blighted and burned out buildings that then for the population that remains behind, that is a severely impacting factor on quality of life for the people who remain here. And we’ve also created a structural element to all of this where it’s almost impossible for Detroit to pull itself back out with the current population and the current composition and number of businesses that we have left in Detroit. It’s structurally impossible for the taxes that are generated here to actually cover the amount of services that we need, both for the land area, the population, and the commercial entities here. So in that sense I guess we’re a little bit unique and we’ve just gone way further out into that system for a lot of reasons, but there are cities coming right up behind us.

Gilchrist: Yes, I think, well, as a Detroit kid, one thing that we all have in common is that we have a lot of pride and we’re generally optimistic—

Raleigh: Sorry. [LAUGHS]

Gilchrist: No, no, no, but in the face of many challenges, one thing that I do think has been my experience with Detroiters is that we have—I’ve seen Detroiters have a tremendous, almost insatiable hunger for, once a tool or opportunity is put in their hands and they feel empowered to understand what that tool can do for them, that they will take it and run with it. And I think one of the examples, again, to sort of go back to what we’ve been talking about today so I feel like you may be familiar, this Improve Detroit application, right? So now I think Mayor Duggan gave the stat, north of 60% of service requests are coming through that. They represent every zip code. Every census track, every corner of this city, people are using this. It’s something that does not only serve a certain part of the city. And I think what that means is that if we build things in a thoughtful way, then people will use them. So what’s special to me is I feel like people’s hunger for these sort of things—I think people want to believe in Detroit. Like, fine, so people who are either coming here for some reason, people who have been here and remained here for whatever reason, they have an affinity toward the city that I think motivates them to want to use what’s available to them to make their life and their community better in a way that—yes, I’m biased. I think it’s unique. And so I think it’s our job to continue to feed that, feed that with high quality information, feed that with high quality tools, feed that with high quality opportunities for interaction, both with city government and with other citizens and residents and businesses, and I really think that that, we’re going to be able to create some models that other cities can learn from that are going to be coming up against the same challenges that Detroit’s sort of the canary in the coal mine for.     

Hewett: The people who elsewhere are trying to apply these things and are looking at them are not always in as favorable of a position as you are to have a mayor and CIO who get it. Not only that, but very often it’s not a municipal situation. It’s a corporate situation with large datasets of telephone information, telephone data, and their primary goal may not be development. And so one of the things that I know, Amanda, you’ve talked about, is this principle of go slow in order to go fast. And that’s something that those other communities need to see in order to appreciate. It’s something that you all I think do quite intuitively here. You know if you get out ahead of that feedback loop, you’ll run into problems. How would you advise—and Amanda, maybe you can start just by setting the stage for what it’s like out there where there isn’t the case. But then I’d be interested in how you, as Detroiters and people close to this issue, would advise others to pay attention to go slow portion of this.

Noonan: Yes, I think sometimes the—well, criticism may be too strong a word, but the concerns that individuals have about a process which seemingly is deeply vested in a lot of research and engagement is that it’s slow and that really we need to get to prototypes and we need to start iterating and we need to evolve and we need to innovate, blah, blah, blah. And that’s all good. I completely understand that. But I think the time that is spent at the beginning of a process to understand the real questions you’re trying to answer or the real needs you’re trying to meet within a community are so well spent because they in the end may determine a completely different set of outcomes from the ones you imagined when you were sitting in your seat thinking, “We need to design a service to do X or Y.”

I’ll give you a little example, which is not about Detroit, but is about Pakistan. We work globally—we do work in the US, but we also work globally, and we undertook a project in the federally administered tribal areas of Pakistan, who for the first time two years ago were able to vote in national elections in Pakistan. So it was about getting civics information to people who hadn’t previously been able to vote. And our team on the ground in Islamabad have been there for many years and had thought that actually using mobile technology would be a very good to get information to these communities, as they thought most people in those communities had mobile phones and had access to networks. So I was asked to run a pilot study using an IVR, which is one of these interactive voice recognition systems—you know, dial 1 and you’ll get information about this candidate, dial 2 about how to vote, that kind of thing—and take it to the communities and test it, user test it to see if it was useful. And I had some few questions of our team in the ground. One was does everybody have mobile phone? Does everybody trust what they hear from mobile phones? Do they use them for that kind of information? Do they need civics information or are they getting it from somewhere else? It was lots of questions which we didn’t really know the answers to, and so instead we decided to take a very different approach, which was we recruited individuals from each of the sort of main areas of FATA, we bought them to Islamabad, and we trained them for two weeks in how to understand and observe their community: qualitative research, survey research, observation, ethnographic research, photography, shadowing, the whole gamut of things. We gave them a whole set of questions to go back to their communities and observe and survey how people were using technology, how people were getting information, and what their information needs were, particularly around elections.

Long story short, we did daily synthesis, so the questions that we were asking on day one were not the same questions we were asking on day three. We were reacting to the findings that we were having in the field as we went through the process. And we discovered for an example, in North Waziristan, which is the most kind of Taliban-controlled part of FATA, that people were walking around with smartphones. Our researcher let us know on the phone at night that—he came out in the area told us that people were walking around with smartphones. And I said, “How come when there’s no phone connectivity there? The Taliban have taken all the phone towers down.” And he said, “They use them, their relatives send them to them as gifts from the Middle East where they’re working as migrant workers. And so they’re kind of cool things to have.” And I said, “Yes, that’s okay, but why are they walking around with them?” “Don’t know.” So the next time he went back, he looked again, he asked people questions, and he came back with, “Oh, well what happens is when they go to Peshawar, they’re able to connect to a network, so then they get their texts and their messages and their emails and they can call out and things. They go to Peshawar regularly because they have to go there to get their anti-depressants.” And I said, “That’s really interesting, but why are they using the phones when they’re in North Waziristan. It still doesn’t answer that question.” Anyway, long story short, it transpired that what they were doing was they were using the phones to transfer files between one another using Bluetooth. So they were taking audio files and video files that they had collected and then they were able to communicate into a trusted—this is very critical—person who they looked in the eye as they made the transfer. And so they were able to transfer lots of information very quickly and very securely between individuals in that group and that’s how they communicated within the FATA, where there was no telephone connectivity.

Hewett: And how long did it take you to get—once you had a hunch that there was something operative in that data community, how long did it take you to get to that piece of information?

Noonan: Probably eight or nine days as an iterative process, every day, going back and asking more questions. And then what happened was we were asking questions at the same time obviously about the elections and things and we discovered that people already had very good information about the elections. The candidates had been very effective at getting information out and about. We also discovered that people were already using IVR and they didn’t trust it because it was used to spam them all the time—not unlike here. So we knew the IVR was not—a) our original design questions were all wrong. IVR was not a good way of communicating. The information that we thought people needed, they didn’t need. Not everybody had access to mobile as we had assumed and those who did have mobile didn’t always have connectivity.

But what we did discover in the process was that Bluetooth was being used in a way for which it had never been intended when the smartphone was designed. So that’s that serendipitous thing. That’s that thing about technology being an agnostic thing and giving it to a community and they will innovate, and being able to sit and observe that process in action, for technologists and people who are designing tools and supports for those communities, is vital because that’s where we see where the true innovation is.

Hewett: And it requires a great deal of discipline on the part of—you’re an individual with this very deeply inside—as a hunch, you know—

Noonan: And also, we were working with—I should’ve said that the researchers were actually talking with us and synthesizing together every day, but also going back to the community and testing what they were observing. So it’s an iterative, ongoing, inclusive process. It’s not us on the outside saying, “Oh, this is what’s happening.”

And then the final thing was that, actually, what we did surface during that process was that there was a great problem with mental health in North Waziristan. People there are depressed. They have a really terrible life. They have the Taliban blasting out religious messages and songs all day, the drones are going over 24-7. There’s real mental issues there. And they’re also using these smartphones to communicate information. So perhaps what they need is information that’s designed in a way that can be packaged for Bluetooth communication so that they can understand how to get help for mental illness and whatever.

So there was a whole new scheme of things that we actually were able to surface during that process, and what I would say about it is that not only was it inclusive and iterative—it was slower, for sure, to get to our first prototype, which was the mental illness thing, but it was actually probably more effective, I would argue. But the other thing is there’s a legacy there. We created a team of people who are now embedded in those communities who, beyond that project, have the capacity to observe and feedback what the kind of impacts are of any solutions or any activities you then put into that community. And so that kind of context engine is a very valuable tool in understanding how any intervention in any community, be it Detroit or be it the FATA or be it New Orleans or wherever you’re operating, is having and then being able to change and respond to that.

Hewett: What I was thinking is that you all are somewhat flexible in the way you work. You’re out in a remote place and you’re communicating very effectively, and at a civic level, where there’s governance and policies and meetings and all those things, it requires a great deal of discipline to protect the space for that kind of feedback to make a difference in an ongoing way in what’s being done. And I’m just wondering if you wouldn’t mind giving us just a sort of view behind closed doors of how you all do that work to make sure that you protect that so that you do remain flexible and reflective about how that’s going on so it doesn’t get systematized in a way that’s counterproductive.

Gilchrist: So I mean the tone for that gets set at the top, as far as leadership believing that that type of process is ultimately how you identify at the right solution. And so one of the things that the administration has been doing is making a real commitment to evaluating process in the city through a lens of lean and like six sigma type of process optimization, cutting out waste and fat. And given that that’s a fairly well known approach to process improvement, and therefore it’s well understood that this is a process that has phases—you have to observe, then you do data collection, then you do data analysis, then you make a hypothesis, then you test that hypothesis, you iterate on it, etcetera. So one of the ways that we’ve approach that in our process improvement work, which applies not only to process improvement, but also applies to how we’re trying to run the government broadly, is thinking about things from that perspective of like the way to get to the right answer is to have a methodology and to follow it.

Now, that’s not always clean. It doesn’t always happen like that. There are instances where we shortcut or short circuit that process. So we’re trying to be as thoughtful as possible, because at the end of the day, also, there are times when we need to be responsive and to be responsive immediately. Things happen that the government has no choice but to respond to and we have to be thoughtful about that when we need to deliver a service in that way, and be okay with the fact that that may not be the optimal way to deliver it. Like a concrete example is, during my first week that I was working last year, we had a once in 94 year rainstorm where like all the freeways flooded and trees were everywhere, all over the city. It was absolutely bananas, and there really wasn’t a great way for residents to be able to tell the government, “Hey, there’s a tree on my house. I need help.” So we sort of scrambled, and that day—that was like the first time I’d been in a cabinet meeting, literally, and it’s like, “Oh, hey, I’m the new guy and I know how to like put a form on the Internet. Let’s try having people be able to do that.” So we put out this press release with the link to this and blasted it all over social media and people immediately started using this to report trees in their driveways, trees on their house, trees on their sidewalk.

It wasn’t the most efficient solution, but it was a way to respond to a crisis. So I use that example to show that there’s a spectrum of problems that we face, and the ideal scenario is, yes, to look at this in a methodical, evidence-based approach, and we try to do that as often as we can because we believe that that’s where we’ll get the most robust and sustainable solutions. But sometimes

Hewett: I’m just curious what hearing this level of engagement and sort of fine tuning going on as the process is rolling makes you think about the figures and the data you collected.

Webb: Sure. Yes, so for the interested bystander research we conducted over a hundred interviews with people across the country, getting buy-in from stakeholders at Google, an engineering, data-driven company, you know, where we—a lot of people think we can get all our answers by looking at what people are searching for, but we can’t answer why or what they’re doing and so forth. I think this is a similar challenge across the civic tech ecosystem. We have all these pools of data coming available and we’re building these great solutions. It’s just we’ve jumped ahead of the game where we don’t actually know what people’s needs are and what problem we’re trying to solve, and it’s not always that you have the luxury to carve out this time to do such an ambitious research project like we did. But if you can start proving yourself and the worth of this type of sort of human-centered design and research to your team or the folks that you’re working with, with a smaller project—it doesn’t have to be quite so ambitious—then you can perhaps get the support to do something like we did, which literally probably took six to eight months to do. As sort of the user research champion on my team at Google, which has a pretty robust UX team, it still—you know, I still have to sort of seek out my supporters and build a coalition of folks, and then the goal is to bring them into the field as well and really make it an experience in something where they can start to understand the people that we’re trying to design things for.

Hewett: Well I want to thank you all very much. And I want to open the floor to questions from the audience, people who have questions specifically on something we’ve talked about or something else.

Audience 1: [INDISCERNIBLE 0:02:07.8] because we find that this is a huge issue with moving people off the civic tech paradigm, if you will. And can you just speak for a second about the curriculum [INDISCERNIBLE 0:02:25.9]?

Raleigh: Sure. So we actually, we partnered with the U of M Graham Sustainability Institute, who actually found us funding to do this three years running. So this was our third and last year under the project that we had together. And we actually started in the first year with just sort of really basic like, “We’re going to try something and we’re gonna see if that works,” and it was really like, “What are data? What different types of data are there? How do you categorize them? How can you use them in different ways? How can you visualize them? What are the things that you could create from them?” I think we spent like four hours—we did three sessions, four hours each, and it was the same curriculum but a different group of people moving through that summer. And then we learned from that summer—so we’d do pre- and post- surveys every year. We learned that people were like, “This was great, like the last hour was awesome, but this beginning crap, we don’t need that.” So the next year we retooled, and it was actually the year of the big flood all over southeast Michigan, and because Graham Sustainability Institute has this environmental focus, we thought we would bring—the app that I mentioned where people can customize the survey questions that they would like to ask, we actually put that in the hands of nonprofit, city, and philanthropic leaders and sort of mid-level managers in the curriculum that summer and sent them out into the field and they were collecting data on areas that were still flooded, areas that had debris from the flood piled up or whatever and then they used that to actually take action, and we did three of those.

This summer was more about this the open data portals and using free online open source software to actually map layers of data together, kind of taking people through—it was a much more—it was probably the more sophisticated, the most sophisticated workshop that we did so far. And we do lots of other ones as well and we usually—the presentations can actually be found online and we’re working on a full post about how—basically our process around it and then how can you actually take the packets of information that we created for each workshop and use them yourselves.

Humphrey: Andrew Humphrey, WDIV Detroit and Tech Time with Andrew Humphrey also. What examples would each of you have where your work in the City of Detroit has had some sort of an effect or impact on some of the poorest, most disenfranchised people within the city, not only where they may have received a benefit, but where they also had a say and were empowered?

Gilchrist: I can give a concrete example of that. So the most on the Detroit, the city’s open data portal, the most downloaded, most utilized, most traffic dataset we have is on reported crime incidents and so we’ve seen a lot of—that’s also been not only most downloaded but also the thing that people have created their most custom views and custom slices of the data on, so like, “Show me all the robberies that were on a street with this name, show me all these crimes that are in area,” these types of things. That’s been utilized most heavily by block clubs all up and down and all over the city and so we’ve seen very concrete instances where communities in places—like neighborhoods like Bagley in the northwestern part of the City of Detroit have used that data to actually work with their neighborhood police officers very closely to respond to what they perceived as sort of upticks in certain types of crime, but they’ve actually used that to work with police officers directly about how they want to see their neighborhood policed. And so I use that neighborhood as an example because it’s not one that you’d think of as, oh, this is one of the cool like rich kid neighborhoods. It’s a neighborhood where, you know, like my mother and my grandparents, they lived and grew up in, in a part of the city that isn’t often talked about a lot. But it’s an example of people that, again, all over the city utilizing the resource that we have to be able to make local decisions, and this one was about public safety, which is something very near and dear to many Detroiters.

Raleigh: I can give a reasonable example. We’ve been in partnership with Community Development Advocates of Detroit for about five years, six years running now, and that is a participatory community planning process that’s been done in multiple neighborhoods across the City of Detroit. Our work is actually—I’m trying to remember, it might have been you earlier talking about how like if we actually bring data to a community and we say, “This is what we think it is like, from the administrative data sources that we can find. Is it right?” And then we can actually feed back into our data systems, taking that sort of data exhaust that we’ve processed from many other situations and databases all over the place, we can actually correct that with local knowledge and then it’s a level setter for a conversation around future desired outcomes for that block, census track, overall neighborhood, district, whatever it is. So in those community conversations, I feel like it’s probably one of the better examples of how we’ve been able to reach a larger cross section of Detroiters than in most of our other work with sort of philanthropy and solid, larger nonprofit organizations, but actually working directly with residents, with CDAD.

Hewett: Okay. Other questions?

Tom: My name is Leslie Tom. I’m the chief sustainability officer with the Charles H. Wright Museum of African American History and a Detroit Revitalization fellow. And I was just wondering, in your discussions for mapping Detroit’s information systems, what is your viewpoints of security and privacy?

Gilchrist: So from a policy perspective, the data that the City of Detroit is making available via our open data portal into the channels is data that was already subject to be released by the city if we were compelled to do so via a Freedom of Information Act request. So we’re not releasing anything that has personally identifiable information, information related to people’s income taxes or their healthcare. But we are releasing things that are public record. So when you decide to get an alterations permit for your home, that’s a matter of public record. But there are some things that are kind of surprising that are public that I didn’t even know going into this process. So one of the coolest parts about the portal is we have a way for people to suggest a dataset. They can say, “I want to see this and I don’t see it, so you need to make it happen.” And it’s actually been a case for me really learning about what actually is subject to that standard and what isn’t. And so one of the interesting examples that we haven’t gotten yet, but it’s on the website, you can see someone requested it, is about parking. And so apparently if I get a parking ticket, it’s actually public information that I got a parking ticket. My license plate, where the ticket was, all these things, that’s like actually public information that if someone today filed a Freedom of Information Act request, they could get.

So the way that we’re trying to approach this is something where it’s ostensibly public, that makes it fair game for us to release. Now, I understand there are some considerations around like there’s public and then there’s like public, right? There’s like for real, right out in front of people and right out in front of you. But we ultimately believe that our default should be transparency. We ultimately believe that we should not make it hard for people to get public information from a public institution, and so we’re going to start from that place as often as we can.

Hewett: Does that answer your question? Because I’m curious, does it make—are you going to make it easier for me to find out where you got that parking ticket?

Gilchrist: Ultimately that would be the intention, yes. [LAUGHS]

Audience 2: [INDISCERNIBLE 0:10:34.6 ] Just curious, across your different work, if you’ve considered the role of private sector data, or maybe in the case of Data Driven Detroit, community data—maybe that’s what you meant by data exhaust—in terms of using it for different kinds of social impact or effective governance?

Raleigh: We do already use some of that. It’s interesting—so Leslie and I have met before, I’ll say that. We have a fellow in the program as well. The question is a great one and it’s something that we struggle with internally at D3 all the time, and especially as it relates to sort of proprietary data that we either purchase or we obtain for the specific use of one analysis, or multiple analyses, but we’re not able to release at sort of a record level. So we’re always thinking about how can we aggregate or analyze or integrate that information, those data into something that becomes actionable information, so we’re getting the results out, again, this actionable intelligence out into the community’s hands in an equitable fashion, but it doesn’t necessarily mean that we need to be releasing or violating proprietary data agreements that we’ve entered into in order to make that data useful. Does that answer the question? Because I think specifically too about utility shutoffs, right? We work with DTE. We have these data. We’ve been able to help inform the strategy for the City of Detroit and the Land Bank Authority in terms of blight interventions across community and that is one essential element in that huge database that we’ve compiled for the work. But shutoff information and vacancy information down to that point level is just, it’s not something that they’re comfortable with and probably shouldn’t be, so—but we can make the data useful.

Hewett: Well we’re right at 3:15, so I want to be conscious of the time. So to close, first of all, I want to thank our panelists and ask you if you would to talk to us, or just tell us as we close, what will you go back to, to work on tomorrow specifically?

Gilchrist: I’m working on making the city’s ability to put things on maps a lot easier.

Hewett: How specifically?

Gilchrist: So specifically everything from very clean tax parcel information, taxable status, tax value, to working with some of our partners in the fire department on fire data and incidents, to actually data on how we’re going to be planning road and construction projects for 2016.

Noonan: I’m working on a report on data literacy and how important that is as an issue of inclusion and an issue of ethics, actually, and something that at this point has been really overlooked and really needs a lot more attention.

Hewett: Is that a report that will be publicly available?

Noonan: It will be available at the end of next week, yes.

Raleigh: So I don’t get to play with data anymore, but I will totally be sending links to all of the research that has been shared with these two folks because I think setting up a system for D3 where we can continue to be thoughtful around the participatory side of things and making sure that we’re setting aside that time in all of our projects is something I would like to make sure we remain committed to as we embark on our next adventures.

Webb: And I will be with my team getting set for the Canadian election next month and then moving towards to US primaries and general election and other international elections next year.

Hewett: Terrific. All right. Well please join me in thanking our panelists.


Tags: , , , , , , , ,