Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
PolicyViz Podcast Episode #16: Justin Grimes
142 Plays9 years ago
Justin Grimes is a statistician, open data proselytizer, Code for America Brigade leader and all-around fun guy to be around. In this, the 16th episode of The PolicyViz Podcast, Justin and I talk about effective open data strategies, civic technology,...
The post PolicyViz Podcast Episode #16: Justin Grimes appeared first on PolicyViz.
Recommended
Transcript
00:00:10
Speaker
Welcome
Introduction and Guest Background
00:00:11
Speaker
back. This is the PolicyViz podcast. I'm your host, John Schwabisch. I'm here with Justin Grimes. Justin, how are you doing? Welcome to the show. Doing good. How are you doing? I'm doing great. I'm doing great. It's July. It's summertime. It's raining, but it's not 125% humidity here in DC.
00:00:28
Speaker
Yeah, always good, not so much to complain about. So, for those of you who don't know, Justin is a statistician at the Institute for Museum and Library Services here in DC. He's also the co-founder and the leader for the Code for DC Brigade.
00:00:47
Speaker
I met Justin through a number of different initiatives, specifically the Open Data Initiative here in
The Importance of Open Data Discussions
00:00:52
Speaker
DC. So why don't we start talking about open data and then we can talk about code for DC in a little bit. So what's your view of open data in
00:01:03
Speaker
We can talk about DZ, we can talk about the world, we can talk about whatever. What's your view of an open data strategy for organizations, either as they try to open their data within the organization and share it between different parts of the organization, or as they try to take data and push it out and let people outside their organization use it or view it or what have you? Sure, no. I mean, for those who don't know me, I'm a huge open data advocate. I see a lot of potential and promise in open data.
00:01:32
Speaker
I think that for me, I think there's no question. I think every organization should be having a conversation about open data. Whether they are a government organization, whether they're a commercial private sector, I think everybody needs to be having this conversation.
Enhancing Operations with Open Data
00:01:48
Speaker
I think there is a lot of potential. I think that when it comes to, and I mean this in a variety of different ways, I mean in terms of
00:01:58
Speaker
If you are already disseminating data, I think that thinking about it in terms of open data gives you a better approach to data dissemination. I think that if you're not already disseminating data, I think there's a lot of advantages, both internal and external, to sharing your information. I mean, the popular example is always that kind of, you know, by opening up yourself to others, you allow others to, like the kind of approach to like open innovation,
00:02:29
Speaker
that kind of strategy of maybe there's an outside organization that can see something that you don't see with your data. Honestly, if you've ever worked in a large organization and you need data within your organization, you'd be surprised how hard it is to get data from within your own organization.
00:02:49
Speaker
So if you make it better for external parties, you're clearly making it better for internal parties as well. Right. So let's talk about government for a minute.
Challenges in Open Data Audience Understanding
00:02:56
Speaker
Since you work for a government agency, I used to work for a government agency. When it comes to open data from a government's perspective, either local or federal,
00:03:06
Speaker
Do you sort of feel like agencies are trying to figure out who their audience is at this point? I see a lot of open data portals from different government agencies and it seems like a lot of them are sort of just put out there because a lot of it's just like open data means put everything out there but don't really consider who's going to be using it. Yeah, no. I mean, that I think is a very, very common question that gets asked all the time in this space.
00:03:36
Speaker
And I always have a very, very specific response to that. I mean, my stance has always been, I mean, first off, I would say that like open data is a very trendy topic. And that's independent of whether people actually see the value in it or not. I think that, you know, I can speak for the municipal government in DC was one of the first municipal governments in the world to actually start making, to push this open data, have a data portal to try and maximize the amount of data that they were making available.
00:04:06
Speaker
Um, and then you start seeing this trend crop up in almost every municipal city, which is almost kind of like, well, if that city's doing it, we should be doing it because they seem to be getting
Understanding Stakeholders and Data Intent
00:04:16
Speaker
a lot of good press. It seems to be very positive experience. So there's a little bit of kind of, you know, following the leader in this regard. Um, however, I mean, my thing about it is I'm, I'm a statistician. I collect data all the time. I mean, this is my job. I mean, so from every agency, every organization collects data for a purpose.
00:04:36
Speaker
So when we talk about trying to identify the stakeholders, I'm always taken aback a little bit about that because no one collects data just to collect data. I mean, you had there was original an intent and a purpose behind that. So you should already have like some general idea of why the day is collected and how it's being used and who's, I mean, what's the genesis? What's the provenance of the data? So that should be your first step. Now, I think that hopefully when people ask that question of, you know,
00:05:04
Speaker
Who are the stakeholders with it? I mean, who should be, you know, who should we get into use our data? It's a question of who is not already using our data and how can we make them aware of that? I mean, I hope that's a conversation, but sometimes
User-Friendly Data Portals
00:05:15
Speaker
it's not. Sometimes I think there's a big disconnect between the collection and the dissemination and those that are disseminating are used. I mean, like people that are used to often setting up the open data stuff are usually more on the IT side. They're not connected to like the, you know, the day-to-day activities. So it's just a thing to them. They just throw it up.
00:05:33
Speaker
but it's disconnected from the policy, the procedures, the business practices.
00:05:38
Speaker
Right, because I think, take this example, so a Department of Transportation collects data on traffic flows and you know some guy in Northern Virginia say wants to figure out the best and fastest and easiest way to get to work and so there's some open data portal on the Department of Transportation website and so the person goes in and plays around with it but sometimes those data portals are sort of hard to use or
00:06:05
Speaker
things are listed as variable names and not as sort of things that are easily identifiable. There's code books that are buried in other places. So I sort of feel like the data is all, in that case and similar scenarios, the data are collected for the government analysts to sort of better understand traffic flows.
00:06:24
Speaker
And then they're put out as sort of a service to allow others to either understand their personal situation or to make comments back to government, say, hey, there's this big pothole or I've noticed on this road that there's always a traffic jam or something like that. But it seems like those two things don't always sort of blend together because the open data portal is not very useful for that sort of novice user. And maybe that's OK. I mean, maybe the open data portal aren't met for those people. I don't know.
00:06:54
Speaker
Yeah. Well, I mean,
Improving Data Portal Usability
00:06:55
Speaker
there's a couple of thoughts there. I mean, my, my first thing about it is, um, so, I mean, you have to disconnect, I think in some conversations, the data that's inside the portal from the portal itself, the presentation, the tech. So I think that for, for it's, I mean, it's, it's like how any, it's like any type of information or delivery of that information is it really depends on the context of the user. So I can't write a report that's, that's intended for every specific type of user.
00:07:24
Speaker
The same way as I can't disseminate one data set that's going to be perfect for every use case. I mean, especially with data because I mean, think about how often you have to transform or change to do something to get it to a point where you can actually ask the questions that you want of the data. But I mean, so that's, I mean, that is a challenge. I mean, you know, and everything you described there is serious issues because what is how, I mean, often how data is collected is, as I said, designed for a very specific purpose. And it's not really designed for,
00:07:54
Speaker
I mean, most of these, unless it's being collected today, most of it was designed for a very narrow, very specific purpose. And so there's like insider jargon and all that type of stuff. And I mean, so I mean, there's definitely steps that can be taken. I mean, I think people can think about why they're disseminating data to make it better. I think, you know, better metadata, tightly copying that information together. Maybe it's not releasing one dataset, maybe it's releasing the same data, but in multiple formats and multiple
00:08:25
Speaker
multiple arrangements. I think for the portals, I think you're right. I think the portals are, I mean, we're, we're in a very arcade. I think that we're in a very new period. I mean, open data hasn't been around for that long. So we're still, we're still getting our sea legs here.
Engagement Through Open Data Portals
00:08:40
Speaker
So, but for me, I think that it's, it's the same way as like, think about going to, you know, think about going to a library and how you check out a book and that experience of, you know, we have,
00:08:50
Speaker
you know taxonomies and do we decimal systems and all these things in place that better assist you know somebody looking for a book or serendipitous browsing of books I mean look at how we present data and data portals I mean there's there's so much that could be done in terms of the interface or you know maybe it's you know maybe it's improving the fasted classification or improving the metadata or you know there's so much promise in the area of linked data connecting data sets together I mean there's so much potential there but I think that we're still in that kind of
00:09:19
Speaker
you know, this is like open data of, you know, 0.1. Right. Like, especially when it comes to data portals, there's so much more that can be done.
00:09:28
Speaker
Yeah, I think you're right. We're just starting to figure this out. And you're seeing more and more, at least at the city level, more and more cities starting to figure this out. And it's really interesting to hear some of these sort of anecdotal stories of different sort of city agencies working closely with communities to improve the data and how they're exporting it and who it's targeting. So it's an interesting field. I'll add one more. Yeah.
00:09:56
Speaker
You hit it right there in the last comment that the most important thing, putting the data out there, disseminating the data, putting it on a portal, those things are just a system of steps. The real value is the engagement. The conversation is the most important thing about open data portals, is the opportunity to be able to continue a conversation
00:10:20
Speaker
And that's, and that extends not just beyond like the comments and the, you know, on the open data portals, but this extends to the agencies releasing the data is it's not just enough to throw your data up there and say, Oh, it's open data. I'm done. I mean, it's, it's got a, there's feedback loop to that. That's a participatory process from, for which you can learn more about the community you serve and then, you know, iterate and change your processes. And then it should be a very, it should be, it should be
Measuring Success of Data Portals
00:10:45
Speaker
like a feedback loop. It should be a very cyclical system. I mean, it should be a cycle.
00:10:49
Speaker
I provide data to somebody, I learn something, maybe that changes my practices or what I collect or how I disseminate it. I change my process and then it goes back out to the citizens. It's very adaptive. It should be, I should say.
00:11:02
Speaker
No, the engagement is probably a key piece. And there's always a sort of conversation about data and data visualization as people put things out. How do they measure success? How do they measure that something is having an impact? And I think with these open data portals, you have the opportunity to actually measure success as you engage with the people who are using the data and see what sort of products or analysis or conclusions they're coming to as they use the data. Yeah.
00:11:31
Speaker
And I mean, like I said, this is one of those things like, you know, I mean, for example, most agencies didn't keep probably good track of, you know, data usage before. I think data portals make it easier. There's a little bit of a like, most data portals have an analytic back end that you're getting a little bit more greener information about the use. I think that even that still is still pretty basic. I mean, we still
00:11:56
Speaker
I mean, I know lots of organizations are like, oh, our data set's been downloaded 1,000 times. I'm like, oh, that's good. And they're like, oh, this data set's been downloaded 2,000 times. I'm like, well, that's also good. Because the question is, for when you, and it comes back to a very, very basic core question is, why are you making the data available? What is your goal? What is your organization's goal? Is your goal to spur innovation?
00:12:22
Speaker
Is it engagement? And you can have multiple goals, but knowing what your target is helps you to create better metrics. Because we're still talking about just inputs and outputs. We put a bunch of data out there. That's great. A lot of people love the data. That's great. But what are the actual outcomes? And that's really what people care about at the end of the day. That data set may have been downloaded 1,000 times, but somebody in the community built an application that led to 20 lives being saved or something like that. And now that's an outcome.
00:12:52
Speaker
That's an actual positive thing that you can point to. And that's the story that I think people really want to hear. Yeah, and I've also heard you say, what is your goal at the Open Data Portal? Do you want to have a thousand downloads or do you want to have five downloads? And really, whenever I see you ask that question, I see the audience sort of, they pause for a moment. You can see them thinking.
00:13:13
Speaker
It's not one or the other. You want to have the five people are really invested in those data versus a thousand people who are just browsing it for fun. You want to have those five, that's the core users you're targeting and that's successful whereas a thousand people just browsing it and just looking at it but not actually doing anything with it is perhaps less successful even though you have more hits or more downloads.
Defining Open Data Goals
00:13:39
Speaker
Yeah. I think it's not like a
00:13:44
Speaker
A question that every unique organization has to ask and be able to answer that question themselves. For me, maybe it's just making the data available increases citizen trust, because even if they don't do anything with it, it just increases trust by making more information available. Or perhaps maybe it is putting that data in the hands of not of actual citizens, but intermediaries, journalists, people in the civic tech community, researchers,
00:14:13
Speaker
who then take that data and then add some value to it, and then return it back to the citizens. It's almost like a transitive benefit to the citizens. So citizens may never touch it to open data portal, and that's completely okay. I mean, you just recently did a little survey where they asked people about their thoughts and opinions on open government, open data. It was, I think, a telephone-based survey. And the numbers were really, really low. I mean, it was just like, I mean, even if you look at specifically the questions they were asking,
00:14:43
Speaker
how often do you actually look at government data and do you do anything with government data? It's very low. It's surprisingly low. But again, that doesn't mean like it's not a success to me. That just means that there could be work to improve that. But there's lots of stuff that never, lots of policy topics that never touch the average citizen directly. They're not the direct, directly connecting the citizen to the portal. But I mean, they benefit from that, not even knowing how.
00:15:12
Speaker
Think about how much government data that exists in this world. I mean, satellites. I mean, the atomic clock, what is the national standards? Whatever is the vice president's, I don't know what it's called. The vice president's response. Yeah, I mean, I always used to joke with people, like, what is the most important government data set? I mean, what is the most important government data? I was like, GPS, time. Right, right. I mean, if it wasn't for time, you wouldn't have GPS. You wouldn't have the internet. I mean, because you've got to keep the clocks running.
00:15:42
Speaker
Yeah, so I mean, and that's like or weather or I mean, and no one acts like who very few people access raw time or weather data or GPS. I mean, but you I mean, everything is built on top of it. I mean, and that touches everybody. Well, if you think about just your standard sort of 311 call, if you're just getting information, I mean, I'm assuming that 311 is is more and more being based on the the data portals that are available through government. But
00:16:08
Speaker
If I call 311 to find out why there's construction down the street for me, I may not think of that as using an open data portal, even though the person I'm calling is looking at some sort of spreadsheet or data portal that he or she didn't have five years ago. Yeah. No, I mean, everybody should look at the PSA and the survey and the question they ask, because I have some thoughts on that as well. But I mean, the question of it is, well, what is government data?
00:16:35
Speaker
Like, is government data, like, are we talking tabular data that's in, you know, columns and rows, that type of stuff? Or, I mean, does government data take different forms? Like, what do people recognize as government data? Because, I mean, some people probably wouldn't recognize weather data as, like, that's, I mean, when weather information is presented to them, they wouldn't think, oh, that comes from government, or time comes from government, or satellites were military and GPS, and blah, blah, blah, blah.
00:17:00
Speaker
So I mean, I think that it's a hard time for, I mean, I think some people have a very, I think some people in the populace have a very narrow perspective of what government data is.
00:17:11
Speaker
No, I think that's right. So we're going to run out of time, but I don't want to leave without talking a little bit about Code for DC because I think a lot of things that we're talking about leads right into the sort of events and things that you guys are doing in bringing it together, all these different groups.
Code for DC Projects and Civic Tech Impact
00:17:29
Speaker
So can you sort of talk a little bit about the sort of projects that you've been doing with Code for DC over the last six months or so?
00:17:36
Speaker
Sure, so I mean you and so I'll mention COVERDC, we are a brigade of Code for America based in the District of Columbia. We are an all-volunteer group where we take technology and open data and try and find ways to improve the local community. And so this, you know, it works, we work with, volunteers work with government agencies, we work with nonprofits to try and come up with novel solutions to try and fixing issues in the series.
00:18:03
Speaker
So you can find more about it, code for DC.org. I highly recommend you check it out. So my thing is, some of the examples of applications we built were caveats. It was a map where you can, loading historical capital bike share data, predict the availability of a bike at a station at a given time based on historical trends. So I mean, for trip planes, you could be like, oh, I need to be at the station at 7.50 to be the rush hour commute. So some of our other applications were
00:18:32
Speaker
taking data from education data and education spending on students. There was a particular program in DC where they were spending additional funds on at-risk students and one of our members built a little application with a nonprofit here in the city to show how the funds are not being allocated appropriately. So it was more kind of like showing using data to assist the populace's understanding about a problem and then guiding them to
00:19:02
Speaker
making a decision like talking to their local council member.
00:19:07
Speaker
Do you think that code for DC and similar sorts of groups with meetups and all these sorts of code for progress and all these sorts of code groups, do you think that's the future of civic tech? Is getting these sort of volunteer organizations and pulling people together or is it a more sort of hierarchical organization, for profit sector that's going to drive that area? No, I think it's a great question.
00:19:36
Speaker
I don't think that, the short answer would be no, I don't think so. I think that it's, you need a holistic approach to this. I mean, it's the same way as we don't rely on any one organization to solve big problems that we're tackling as a society. I think that definitely this volunteer civic hacking, whatever form it takes, I think it's definitely a component that should exist and has value in this space. And I think it should continue to exist.
00:20:05
Speaker
seen through my own work through it, that it does resonate well and does make a difference. But I don't think it can't be the only thing. You can't shift the burden entirely to all volunteer meetups or that type of stuff, nor could you shift all the burden to startups in the tech community, nor shift all the burden in government. We're tackling these problems as a community, and it requires a very holistic response from folks. Well, good. Well, on that note,
00:20:33
Speaker
That's a great way to end it. I'm going to end it on that note. Justin, thanks so much for coming on the show. This is really interesting. We could talk and think about open data, and we have. But I think we could talk about it for a couple more hours, probably. No, no, definitely, definitely. Thanks for having me on. You know I'm always eager to come out.
00:20:50
Speaker
All right, well, I'll have you back on then. We'll do it again in the fall for the next event. And thanks to everyone out there who's been listening. Thanks so much for listening to the show. If you have questions or comments, please log on to the site at policyviz.com or hit me up on Twitter. And I'm John Schwabisch, and this has been the Policy Viz Podcast. Thanks so much for listening.