The Changelog: Software Development, Open Source - Civic Hacking and Code for America (Interview)

Episode Date: July 3, 2013

Adam Stacoviak and Andrew Thorp talk with Michal Migurski (CTO) and Ezra Spier (Fellow) about civic hacking at Code for America, technical sustainability in government, skill gap for more modern software in government, open city data and more.

Welcome back everyone. This is The Change Log where a member supported blog and podcast that covers what's fresh and what's new in open source. You can check out the blog at thechangelog.com and our past shows at 5by5.tv slash changelog. The show is hosted by myself, Adam Stachowiak, and also Andrew Thorpe. Andrew, say hello. Hey, how's it going? It is going well.

You can tune in live to this show every Tuesday at 5 p.m. Central Standard Time right here on 5x5. And this is episode number 95. We're joined by Mike Magursky. He's the CTO for Code for America and Ezra Speyer. I said that wrong, Ezra. It's Speer. A bit close enoughra Speyer. I said that wrong, Ezra. It's Speyer, but close enough.

Speyer. He's also a fellow at Code for America. Code for America is a new kind of public service. It's by the people, for the people, and it's for the 21st century. I think it's the coolest line ever. Welcome to the show, fellas. Thank you. Great to be here.

Yeah, thanks so much for having us. Absolutely. So we've actually had you guys, not you guys in particular on the show before, but back at episode number 65, we had Eric Michaels-Over and Max Ogden on the show talking about Code for America. Great conversation there. I'd like to kind of extend that a little bit too, but definitely fans of Code for America and what you guys are doing. So why don't we take a minute and maybe introduce you guys. So Mike, we'll start with you. Sure. Yeah. Hi.

So my name is Michael Magursky, and I'm the chief technology office here at Code for America. Fairly new here, actually. I just joined the organization a few months ago. And prior to that, up until the end of last year, I was CTO and Technical Director at Stamen Design, a design mapping and information visualization firm located here in San Francisco. Did that for about, I want to say, nine years and did a lot of work with open source GIS, open source geography, open source cartography over all that time. Been a huge fan of Code for America for a number of years. We actually hosted the organization at our mission district office for its first six months of life. So it's kind of great to be here as a proper full-time member now. So you got some roots then. This

isn't, I mean, because you just joined the team two months ago, right? But your history with Code for America goes back pretty far. Yeah, pretty far. I've been friendly with Abhi and Jen and Megan and everybody else at CFA ever since they first got started. Cool. All right, Ezra, how about you? Sure. Yeah. My name is Ezra Spear and I'm a fellow here for 2013 at Code for America. And I think maybe a little bit later we'll describe what that actually means because it's a pretty interesting and unique kind of position I have. But before I was at Code for America, I started in January. I was living in Portland, Oregon and working as a developer at a coffee company called Sustainable Harvest Coffee Importers, where I was actually building applications to track coffee as it moves around the world, increase traceability in that industry.

When you hear about coffee, you think a lot about roasters and cool coffee shops and stuff, but not so much about the people that actually grow the coffee. So I was building apps for those folks, even some really cool educational apps used in places like Peru and Tanzania. And now at Code for America, I'm sort of applying the skills I learned in that context and bringing them to local governments around the United States. So it's a lot of fun, and I'm excited to chat with you guys today. Absolutely. Maybe let's open it up like this because I think it's a little – since 65 is far back. So it's been a while since we had this topic on the show, but as I understand it, Code for America is a startup, right?

Did you not hear what I said? I did, yeah. You were asking whether Code for America is a startup. Is that right? Yeah, I'm trying to figure out what Code for America is. I mean, I remember the conversation back in the day, but it's civic hacking, so it's for – it's to develop the technology space for our cities and our government. But I'm kind of curious, is it a startup?

What is the makeup of Code for America? Yeah, I mean we're structured around a fellowship program. And what that means is that every year for the entire calendar year, we have a class of 25 to 30 fellows who join us. These are people who apply to join us. They go through a fairly rigorous selection process like Ezra did. And they come to us as designers. They come to us as developers, urban planners, and other professions, typically, you know, kind of in the middle to early part of their careers, looking to make a change or transition into government. On the other side, we work with cities who also apply to be Code for America cities. This year, we have a mix of nine cities and counties, including New York City,

Oakland, California, Louisville, Kentucky, Summit County, Ohio, and a few others. And they come to us with very specific problems. Some of them want to work on things like criminal justice. Others want to work on economic development. Still others want to work on food availability and food banks and things like that. And essentially, we create kind of a design camp, design process built around this, where for 11 months out of the year, the fellows and cities work together to create a response to some sort of problem that the city is experiencing that could use a technical solution. So how many cities usually apply each year?

You know, I'm actually not sure. I want to say that it's something like 20 or 30 this year. We're currently in the process of narrowing down our selection for the 2014 cities. And I think we have it down to a list of something like 10 or so finalists. And so they come to you each year basically with a problem set maybe before the fellowship begins and part of the fellowship, this 11-month program, is about tackling and solving some of those problems? Yeah, we try to frame the problems in a somewhat general way. So for example, a city may want to work on economic development, but that doesn't mean that

they're coming to us with a specific application that they want built. We're not like a vendor or a contracting agency. We work with the cities to figure out what the real nut of the problem is and then develop solutions based on research that gets performed early on in the fellowship. So does one fellowship from one year to another kind of piggyback off of one other's work? Does last year's program kind of bleed into this year's program and then on and down the line? In many cases, yeah. The fellowships themselves will end at the end of the year, and then some people will stay on in other positions.

So, for example, we have two folks from last year who stayed on as a designer in residence and an engineer in residence after completing their fellowship. And in some cases, like for example, with Philadelphia, cities will come back for a second opportunity to work with us. Generally speaking, the way that the fellowship and the years feed into each other is through our very extensive GitHub repository of projects and through the startups and additional projects that fellows go on to do with their work after they're finished with the fellowship. That's pretty cool. So let's talk about, I guess, one of the topics that you'd mentioned

when we had a kind of a pre-call. You talked a lot about your desire to talk about technical sustainability as it relates to the different challenges you guys are helping cities that apply to Code for America to solve. What is some of the number one challenges that cities are, I guess, facing with technical sustainability? Yeah, I think that there's really kind of a broad set of challenges that cities are facing, but the two top ones probably are different expectations around technology and then also different expectations around things like budgets and how much things cost.

So in terms of expectations around technology, as you might imagine, the people who join us from industry as fellows are frequently working with application stacks like Python or Ruby on Rails or Node.js, fairly new, fairly modern technologies, while on the flip side, city IT departments tend to be composed of much older technologies. So they might be exclusively Microsoft shops, or they might require that things be built on an Oracle database and so forth. And so there's a huge sustainability challenge in figuring out how to kind of draw those two worlds together and make sure that the kind of open source methodologies that we use for all of our work

can be adequately reflected in the environments that cities are deploying in. You mentioned your GitHub page there. And obviously we kind of know on this show that GitHub kind of represents open source software, right? Absolutely. How much is, I guess, over the years of fellowships, how much have you seen open source begin to form and shape and reshape ideas for city government? It's been colossal. I mean, we have projects from our very first year of fellowship back in 2011

that are still being worked on and extended and redeployed by fellows today. So I think that the idea of open source is something that's really informed the entire fellowship. Cool. You mentioned people that join or apply to be a fellow for this live month program that they're either wanting to make change or get involved with government. What are some of the things that some of the fellows have gone on to do? I know Eric and Max, we mentioned they've been on the show before. What they do now, I think Eric is actually in Berlin. I know Max has been working hard on some other fun stuff in open source.

I'm not sure he does full-time, but those are just two examples. But what are others doing, I guess, as they move on from their fellowship? They go all over the place. It's actually really fun to see. We've had a couple fellows from the very first year move on to actual positions in city government over in Boston. We've had a number of fellows, especially from last year, who have turned their projects via our incubator program into proper startups. around blight monitoring in New Orleans or text messaging in the case of Textizen and have turned those things into startups with actual revenue and business expectations around them. In other cases, we've seen fellows move on to open source companies

and other technology companies. One of our fellows from last year, Jessica Lord, has actually moved on to GitHub itself, so it's kind of interesting seeing other graduates move on there. Other people have gone on to companies like Airbnb. I think Google might be one of them. So people go on to do a lot of interesting stuff from here. That's cool. I wanted to ask, how do the projects actually get selected and which fellows work on them? And what does that process look like in-house? Yeah. So, I mean, Ezra, you're actually in the midst of the fellowship. Do you want to maybe

talk about it? Sure. Yeah. So, you know, the application process, I'm going to put a quick plug out here early at the beginning of the show saying that we're actually looking for fellows for next year. Right now, the applications are due at the end of the month on the 31st, codeforamerica.org slash apply. But fellows come from a variety of backgrounds. We have folks who are hardcore software front-end, back-end developers. We have folks who are coming more from a design side or more from a government side. And the way that our teams are formed, they're sort of cross-disciplinary. My team has two developers and one designer on it. And we sort of knew coming in that we were assigned to New York City, which is where I'm working,

and we knew that our project focus was going to be criminal justice. And in particular, we were looking at a particular part of criminal justice, the time between when a person is arrested and when they're sentenced. There's some basically some new theory saying that these are this is sort of the time within the criminal justice system where we can really have the chance to help people stay out of trouble in the future. And knowing that, we sort of all came here in January when the fellowship started. And the process begins with a sort of month-long training process. And so we all come from so many different backgrounds, and a lot of us haven't actually worked in government before.

It's sort of like a month-long crash course in how government works, how open source works, different strategies that have worked and haven't worked in trying to engage with governments. We even had a day-long negotiation training, which sounds like sort of a weird thing, but it really gave our teams a good ability to learn how to talk to people. And we're sort of realizing, we talk about open source in this community like it's sort of just something we know is good. But when you're talking with folks that maybe aren't as familiar with it or see it maybe even as something that's threatening the way they do their work, you need to be really cognizant of where they're coming from and how to talk to them and really understand them so that you're not actually just being rude or dismissive of the kinds of needs they have. But in terms of your question of how do actually projects get selected, after a month-long training, the three team members for New York and the 27 fellows this year who are working with nine different cities and counties basically did a month-long intensive residency where we flew out to our cities.

We lived there. We lived all together in houses in each of our cities and basically did a month of interviews where we talked with people that work in the city, citizens, community groups. In our case, since we're focused on criminal justice, we were talking to judges and people in courtrooms and lawyers and those sorts of folks to try to ask the question, what's broken? What's not working? What could be better? And from there, we sort of used a collaborative process of deciding, you know, what problem we want to tackle and what apps we want to build.

This is really different, by the way, from the way that most governments work with technology. Most of the time, there's some need that people sort of see. Let's say it's, you know, driver's licensing, you know, data management or something. And then they'll issue a, you know, RFP, a request for proposals, and vendors will bid on those proposals. And then they'll choose someone that builds the application to a, you know, potentially enormous spec for a potentially large amount of money. And we're sort of trying to demonstrate that we can look at problems from the ground up. Use sort of this agile idea, not just in software development of problem solving, of starting with something small, building something and testing it to try to solve a small problem, and then building off of it and expanding it to fill other things as well. So it's sort of a new idea with a lot of governments, but folks are starting to get pretty excited about it. I'm kind of curious about, I guess, city collaboration.

You know, when I think about, you mentioned, you know, driver's license and data sets and stuff like that around that and different applications that might make sense. At what point do cities begin to speak at the government level around technology? Is there some sort of, I mean, is that what our national government helps to accomplish or is it up to the cities and individual states to help kind of lift those cities up and speak to one another to maybe have a unified system for dealing with driver's license, for example? Well, I think it's sort of a mix. Think about it this way. If you're working at a startup or any other company, who's going to make the decisions about technology? It'll be the people there. It won't necessarily be anyone telling you what to do there, though you might have a sense of what the best practices are in your particular field or industry. of the way that procurement, that's the way that government entities select vendors to build software, but because of the way that works, oftentimes cities and counties and states and the federal government are sort of going it alone and looking for products

that meet their needs. From the open source world, we know there's this huge advantage of actually collaborating to solve problems and to build something together. But in the government world, that's a little bit more rare. And that's one of the things that we're trying to do at Code for America is by working across multiple cities and counties and states at once, we can sort of start to see those similar problems emerge and then find ways to bring people together to look at them together.

What kind of a response do you tend to get from the cities themselves? Are they interested, excited? What kind of response do you get when you're actually talking with the government of the cities that you're going to be working with in your experience, Ezra? Well, I think you'll have a multitude of responses. And it depends on who you're talking to and why. Because remember, city governments are not unified. We're talking, especially in the case of New York City, is thousands and thousands of people, all with different roles and different needs

that they have to do for their work. I think most of our cities we partner with are ecstatic to be participants in the Code for America Fellowship Program and the other programs we have because they see us as a way they can do their work better. They have things they do, whether it's driver's licensing or having a police force or any of those other services that we think of when we think of government.

But they never have enough resources to be able to do everything they possibly want to do, and they see technology, and particularly our kind of open-source technology, as a way of doing better work maybe for less money, and particularly our kind of open source technology, is a way of doing better work, maybe for less money. And that's awesome. When it gets to actually building software, though, it's a little bit more complicated. As Mike was saying earlier, folks that run IT departments have particular methodologies and stacks and things they're really comfortable with. And sometimes us coming in there shakes things up a little bit. But because we're sort of engaged usually at a higher level on the political side, but also working with people

who are doing their day-to-day work, we're usually able to make things work. Right now, working with New York, I'm in the middle of an IT process. We're trying to get servers provisioned, trying to get Linux set up and all that sort of good stuff. And I'm used to, in my normal life, just firing up Heroku or getting an EC2 instance or doing what it takes myself just to get the infrastructure I need. And that's something that is a little bit harder, I think, in a larger institution like government right now. So we're trying to sort of show what's possible with a small amount of resources. Yeah, and I just – I'm like having a little bit of a hard time visualizing. So essentially you – when this started for you in your first month when you just kind of got dropped into New York City, what did you – like where did you go?

What did you do? What was your – I mean this is so interesting to me, but what was your first, you know, thought of where to get started? So the way that we work is that with each sort of local partner, and we've, right this year, we're working with nine different cities and counties across the United States. From New York, where I am, to San Francisco, Oakland, Las Vegas, Summit County, Ohio, which is the county around Akron. We're working in a ton of different places, and as Mike was saying, each one of these governments has sort of chosen one topic that they want us to work on for the year. The way that our partnerships work is we generally have one or two primary people within the city government who serve as our main contacts and sort of our liaisons. Usually one of them is working in the topic area that we're working in. So in the case

of Summit County, they're doing a project related to parks. So they have folks in the parks there that they're working with primarily. But then they also have contacts in the IT or technology departments of those cities as well. So we have sort of a liaison both on the sense of understanding the problem and also understanding how technology gets built. So when we showed up at the beginning of February, our primary contact is a fellow in New York City government underneath the mayor's office in an office called the Office of the Criminal Justice Coordinator. And I don't know if you've ever watched Law and Order or any of the other New York cop shows, But one of the things about criminal justice, it is so complicated.

It's especially complicated as a person with no background in that system. But even, I think, for lawyers and other people, it's pretty complicated. And our main contact has been working as a lawyer for many years in New York City. He understands the system really well. He's been a defense attorney.

He's worked for the attorney. He's worked for the government. And he knows everybody. And so working with him and with some of our other friends, we were able to set up meetings with people all across the system. So police, courts, probation, district attorneys, all the sort of players that are involved in the process. And by talking with each of them, we can start to get a sense of what's going on. So what projects specifically have you been working on in terms of open source with New York? So we're still sort of figuring out what exactly the end result is going to look like. I think a lot of the work that Code for America has done in the past has been on

open data. I think we'll come back to that in a second. But you know, we're talking about criminal justice, we're talking about people who are being arrested, who are having some, you know, really serious things happening in their lives. There's a lot of private and sensitive data that is a little bit harder to work with. We can't totally do that out in the open because we don't want to share all this data. It's not legal to share it in many cases. And so the kind of work that we're doing is applications that help bring data about people who are being arrested to folks who can really help them out. And working alongside the

infrastructure that the city has already sort of put together to help make the system faster, building applications that provide new views of data and better streamlined views of data to the right people. And that's a little bit of a vague answer, but it's sort of building on the shoulders of the kind of work that's already been happening in New York to make the criminal justice system smoother than it is right now. Gotcha. So you're being named Code for America and obviously getting involved with government. How do you, and maybe this is a question better for Michael, but how do you kind of get involved on the like ethics side and the government side with,

like I would imagine you try to avoid the politics side of things. Yeah, that's absolutely right. I mean, one of the things that we discovered very early on in the program three years ago was that we had to for example, on its way out due to an upcoming election. And one that can actually, you know, get people into rooms for meetings. It's really important that fellows like Ezra be able to have access to the people in the city that they need to have access to. So in terms of politics, what we're looking for is someone like a CTO or a CIO or a mayor's chief of staff or somebody at that level of government who can basically smooth the runway for us and let the people in the city know that Code for America is coming, answer their phone calls. They're here to help. Yeah, pay attention.

Yeah, exactly. That's cool. I mean, at least you have that kind of clout to be able to do that. I mean, I can imagine that your success year over year has kind of afforded you that ability too. Yeah, I think so. I mean, you know, obviously I'm new here, but I've also really had a nice vantage point over the past few years to watch how the team has been working. And I think a lot of that success is really encoded in the happiness of CIOs and CTOs that have worked with us. You know, I've spoken to people like John Tolva from Chicago, and they've told me that they've just been really thrilled with the Code for America mission. And in a lot of cases, it's not even so much about the specific code that gets generated, although that's a large part of it.

But it's really about the process that Code for America brings. It's being able to spend almost an entire calendar year just focusing on a city's issues and problems and working on a variety of different solutions for them and trying to kind of draw connections between disparate systems and data sets. That, I think, is really where a lot of the value comes in. It's that ability for us to act as sort of lateral thinkers across a government where the people that we're dealing with are potentially, you know, members of a hierarchy of sorts, you know, in the IT department or in the criminal justice department and really unable to forge some of those connections across government that they would like to.

And so we almost become kind of like a, I don't know, like a cut across an organization in a way. Yeah, it's cool because it's almost like the city itself, I'm sure it's refreshing to them to be able to actually, you know, see things start to happen without as much of the red tape and the huge grants that need to go out and, you know, this and that. So to see problems, real problems, get people dedicated to working on them, just for the sake of fixing problems, I'm sure is refreshing to those cities that apply. Yeah, that's definitely something that we've been seeing across the kind of civic hacking and civic technology world as well, is that one of the reasons why this is such an interesting space for a lot of these folks is because it's a way for them to look at problems at an angle that they haven't really seen them at before. And that ability to kind of dive in and try solutions in a way that's much more influenced by, you know, like lean and agile methodologies of code applied

to civic problems, I think is quite interesting for them to see. Has there been any like – so I hate to go sports here, but during the last 15 years of sports, it's gone from like a gut-fe these problems from, as you pointed, like a data mindset, has there been a reaction from – or a response from people in the cities to want to go in that direction rather than what politics seem like, government seems like so much is a gut feeling and that sort of a driven field to switch that up and go into like a statistic and data driven environment. Yeah, the reactions have been quite positive. I think that, you know, the particular kind of data that drives us is really based on kind of user research and needs finding exercises. So it's about figuring out what a kind of minimum viable product or prototype is

that we can deploy in that city really early on in the fellowship, and then sort of iterating on that and working on new versions of it. So it's really close to the kind of, you know, open source way of doing things where you have a general sense of where you want to go, an ability to actually push a piece of code or a design solution or some sort of prototype in that direction, and then very rapidly gather information about how people are responding to it, whether it's working for them or whether it's falling on its face. Oh, yeah. I mean, you said something there when you talked about what we in the technology world

know is the minimum viable product. And I think that in politics, in government, that is something that is not even considered most of the time, right? So how often do you hear about these huge grants, these huge proposals that go out to accomplish these massive tasks, and these tasks take, you know, 10 years, and if they make it to completion, they took 10 years, but if not, they get abandoned at five years. And if they make it to completion, they took 10 years, but if not, they get abandoned at five years. And, you know, these, the thought of like a minimum viable product, it, you know, it almost seems like if we can get people in the government, people in, you know, the public sector to, to consider solving the minimal, you know, solving these problems, like that gets approaching

a specific problem and solving it with a minimum viable product. It almost, you know, you could, you could kind of extrapolate that and say less money could get wasted, less, you know, projects could get abandoned, and more work could actually be done. And if nothing else, I would think Code for America kind of bringing that mindset to cities is a very, very positive thing. Yeah, it's really about putting everybody in a position where they can learn as quickly as possible. One of our board members, Eric Reese, who created a lot of the terminology around the lean and agile startup, talks about the idea of creating, I think his words are like an alternative space for doing kind of testing and accountability of these things. And what he's really talking about is a way to, you know, modify normal expectations around technology being this like, you know, one huge giant release slug that happens at the end

of a multi-year process into something where you're doing smaller releases more quickly, you know, closer to the gate and then learning from them as you go in order to adapt to what the reality is. Yeah, this is oddly enough, not oddly. I mean, obviously he's been incredibly influential in this world, but I think Eric Ries has been mentioned on like four of the last five shows we've done. So that's pretty cool. Huge fan of Eric Ries here, by the way.

Like earlier when we chatted when we were doing a sound check, you mentioned a topic that I'm not too familiar with, but I'm hoping you can expand on it, which is something a little closer to your heart being the new CTO at Code for America. So you mentioned deployment environment is one of the points that you wanted to talk about on the show. What did you mean by that? Yeah, I've been thinking a little bit about kind of what the environment is that we push our software into. And I'm trying to create sort of a language around it or a way of thinking about it that encompasses both the technology deployment environment,

which is to say, you know, Linux versus Windows versus Rails versus Python, all that usual kind of stuff. And then also the kind of political and emotional and human environment that that lives inside of. You know, so one of the things that I'm looking at in the cities that we're working with is what actually happens to this code afterwards. And the answer to that question can be a lot of different things. You know, Ezra could probably talk a little bit more about how New York does things.

But, you know, you can imagine that for a city like New York, I mean, they have, you know, hundreds, if not thousands of people that are working in IT roles and working with databases like Oracle and gigantic Java-based systems, you know, very classic kind of big-ass IT enterprise kind of stuff. So we're working with New York this year, but then we're also working with cities all the way down to the scale of like a South Bend, which is a tiny city with something like six people and their IT staff. And yet they're still very, very interested in kind of taking the applications and work that we're doing and running with them. So we're thinking about deployment environments as kind of a combination of the technology of where you deploy to,

but then also the sort of political and social environment of where you deploy to. And so what are the, what are the, some of the, I guess, hurdles you've been able to identify or look to overcome soon in your, in your tenure as CTO? Yeah, I think the biggest hurdle that we're staring down the barrel of right now is that the, you know, 2013 is a very challenging budget environment for basically everyone in any kind of government anywhere in the world. You know. I talk to some cities that have literally $0 training budgets for their IT staffs. I talk to other cities that have no IT staff whatsoever, just sort of enthusiastic people that want to help but aren't really properly staffed to do things. So I think the real challenge that we're thinking about is

how do you transfer some of what we've seen in the open source and startup universes where things have gotten kind of enormously cheap over the years with products like Heroku and EC2 and sort of, you know, push button data services and things like that, and transfer those to a city environment where they're perhaps used to buying hardware and provisioning racked services and, you know, paying a lot more money for things. So really thinking about how do you save these people money and get them a better type of technology deployment if they're dealing with very, very challenging budget situations.

You mentioned budget there, and it kind of bleeds into the other topic we want to talk about too, which is the skill gap, I guess, for cities, especially as you mentioned earlier, like more modern tools like Rails or Python. And you mentioned deployment environment. That's both the human side as well as the technology side. What are you guys seeing in terms of, and Ezra, feel free to chime in too on this, but just because you're probably more on the ground talking to people, but what is the skill gap problem for cities when it comes to more modern technologies that we,

like Andrew mentioned, you see more of on GitHub these days or in open source that people are using and forging and modifying and changing rapidly versus older technologies and some of the more modern development tools? What's the gap happening there for cities? Well, I think one of the challenges is that when you're working for a government, I think oftentimes it's easy to become really risk-averse because you hear all these sort of stories in the news about government spending too much time working on something or spending too much money. So when you're in that sort of environment, you use what you know,

and it's hard to blame anybody for doing that. You don't want to get in trouble. So I think a lot of times folks that have been in cities for a long time continue to use the things that they know really well. And so that sort of maybe does not lend itself well to a more rapidly paced, maybe open source based software environment. Because it takes longer for things to filter down um i think in in many cases people are interested in open source tools but then they say you know who's going to support this who can i pay to support this application um after it's after it's deployed uh so maybe i wouldn't want to just install WordPress on my server. I would want to hire someone to start and manage the servers that my WordPress instance is deployed on.

So it's just a little bit of a different mindset for one thing. But I think, as Mike said before, there just isn't a whole lot of budgets for capacity building and for training within IT departments as well. So it's a tough situation. It is a tough situation. that meetups might even be flooded with people. Maybe they just don't know where to go. What do you see whenever you talk to some people in terms of some of the newer technologies you're bringing in? I mean, is that mainly what Code for America helps to establish, is moving in more modern things, or is it bringing in help that can even help revive some of their legacy stuff? You know, it's got to be a combination because in the work the governments are doing, they're doing work now. And we don't just want to build things that are going to be totally new, that they're not going to be able to carry over their data from old systems or things like that. So I think a lot of the sort of emphasis that we've put on are on more strategies than particular technologies. So things like open data, things like APIs, things like tools that

help computer systems be a little bit more flexible, let them work together, and maybe not be so entrenched and static. We've seen an enormous growth in the idea of open data over the last few years. And basically, that's the idea that the data that government uses to run itself should be available to the general public as well. One of the sort of great examples of open data at work is, you know, these days when you open up your smartphone, you go to a new city and you open up Maps, and you say, I want to go from the airport to my hotel. And the phone will tell you exactly which bus to take or which train to take and when it's going to come. It wasn't always that way.

It's not hard to remember. Up until a few years ago, you had to go onto the local transit agency website, look through some inscrutable timetable if you could even figure out which thing to take. And it was pretty difficult. But a number of years back, Google and the transit agency in Portland, where I'm from, worked together to create a standard called GTFS. That's the General Transit Feed Specification. Basically, it says, if you're a transit agency, you run buses or trains, and you want people to be able to know what your schedules are, just put your data, put your schedules in this simple format, and publish it online. Now all of a sudden, hundreds of transit agencies are in

Google Maps, they're in all these other applications that people have built to help you get to where you're trying to go. So we're trying to help create more opportunities like this. One area is in 311. So in New York City and San Francisco and a lot of other cities, you can call 311 to report a problem in your city, maybe a broken street light or a pothole. And a number of different companies have started to build applications

that use a consistent API and can let you report a 311 complaint from your smartphone without the city actually even building anything new at all. So we're trying to find ways to help government be more like an API, more like a platform. This is sort of Tim O'Reilly's idea of government as a platform. Just a little search for GTFS. I didn't know that at all, but it seems like it's pretty simple data, too, that at least the spec looks like it's basically CSV. It's a simple specification, but by more and more different agencies adopting that, that specification becomes really powerful.

And it doesn't necessarily even take a whole lot of work for a government or an agency to adopt the standard. But it takes sort of the network effect of that thing, of that thing happening. I think we're going to see more and more standards like that. In January or February, Code for America announced a standard called LIVES that works with Yelp and food inspection data here in San Francisco. So before too long, you'll be able to go on Yelp and see whether the restaurant you're going to met the food inspection last time they were inspected. Similar process is happening right now with housing inspection data. So we're trying to be at the center of bringing folks together to come up with these kinds of simple standards. You mentioned open data as a piece of this, and I got to say that I was pretty excited about the traffic of Chicago when they kind of splashed onto GitHub with five different data sets,

probably all stemming from the same spec that you're mentioning, but the impact it had. But I thought it was kind of neat the way that they, some of the things they had said, but one of the things they said that stood out the most to me was when you want to improve our data, just it. I mean that's a term that we I guess modern software developers have been enjoying for a while

with GitHub and the past four or five years and what that's done but I think it's just kind of neat to say that if you're passionate about open data for the city of Chicago and you want to make a change to some of the data, just fork it. We talk a lot about that at Code for America with our founder, Jim Palkus,

talks about community as capacity. The idea being sort of that if – I think you should mute your computer, Mike. I think we're going to have two streams in from our two computers. So can you mute it? Yeah, I can mute it. Cool. There we go. Sorry about in from our two peers. So can you mute? Yeah, you can. Cool. There we go. Sorry about that. That's okay. We had some technical difficulties.

We got Mike back in. It's glad to have you back, Mike. So, you know, that's one of the things we talk a lot about with Code for America and sort of our approach is this idea of the community is the capacity. In a lot of cities, there aren't resources to immediately fix every pothole in the street or clean up every corner all at once when things are left out there. But if we can help bring communities together and help citizens themselves

be part of the process of making things better, it helps create more ownership. It helps us feel like we're part of the cities and counties we're living in. And if we can give people tools to do it in a way that's legal and right and help people do it the right way, then people probably will because people care about the places they live. Developers have all these really particular skills that can help in even more significant ways sometimes. So we want to help make that possible.

Yeah, I mean, just keying off of that, it was actually one of y'all's projects, the Adopt a Hydrant project that was a fort, and then Chicago turned that into shovels, Adopt a Sidewalk. And it was like, you know, rather than just stepping over the crack, you can actually adopt that sidewalk and make an impact. And it's this code that kind of stemmed from the things that y'all are doing to help cities take that ownership. It kind of reminds me, Andrew, of something that we talked about at Pure Charity a long ago, which was I went to this gala for Living Water here in Houston.

There's this nonprofit called Living Water. And when I went there, part of what their mission is, is to go into various countries, both here and abroad to, to help bring clean water in a place. And they found that just going there and putting a well in place wasn't the solution. It was going there and helping that community lift itself up and understand that they have to take care of the water. They have to learn how to operate the well. And it kind of seems like that's a similar thing in this case, you know? Yeah, definitely. And I think, you know, the Adopt-a-Hydron app became Adopt-a-Shovel, and in Hawaii, it became Adopt-a-Siren. They have tsunami alert sirens around the island. And part of implementing that project, it's a Rails app that's available on Code for America's GitHub page, was training up people in Honolulu about how to deploy Rails applications. Us fellows

generally have those kinds of skills, and we're more than happy to share them with anyone who's interested. So sometimes when there is an application that we're building for one purpose, someone working for a different city or a different organization will see it and think, hey, I want that, but I wish I could tweak it this little bit. And we think that's the power of open source, and we're trying to definitely facilitate that stuff to happen more often. It kind of goes back to that, what I mentioned earlier, and I asked if cities are collaborating with other cities, and it's just kind of that effect. It's like Chicago sees what y'all are doing with Adopt-A-Hydrant, and then Honolulu sees that. Andrew, that's pretty close to you because you just came back from vacation there.

But just kind of seeing the mentality of forking code. In this case, they forked the code base, but they apply it in a different way of adopt xyz and whatever that xyz is is something new for that city it's that's really neat too to even see that and i'd imagine that the issues for those um those code bases are pretty filled with lots of collaboration as well yeah definitely i wanted to jump in the one of the projects that looks like it got pretty popular was honolulu answers and looks like it's i mean it's obviously still going but um i think it's neat to see and i wanted to say this before you even uh mentioned that last thing but it's it's cool because honolulu answers

kind of started to grow and is still growing and then oakland uh forked in and is and they're doing theirs and salt lake city forked it and they're doing theirs so they're like figuring out how to do these open source forks of these different projects I think that's a really neat thing to see that happen, that's a good trend that I'd like to see continue in our country. Yeah, thank you

that's actually one that we're particularly proud of, not just because it's gotten forked a couple of times but also because of the particular kind of social activity that that application engenders. You know, the Answers application basically allows the city to hold write-a-thons where people come in and answer questions very directly for the website of the city. And so in the case of Oakland, where I attended a write-a-thon, or Honolulu, where I saw records of the write-a-thon. What you saw was just regular

people who actually knew things about the local city government coming together to, you know, talk about what kind of questions people might have of the city website and then actually do research and actually write responses to those questions right there at the write-a-thon. So it's built on the hackathon model, but in a way that's completely accessible to regular people who don't have coding skills. In the case of the Oakland one, I think we attracted something like 50 or 60 people throughout the day on a Saturday afternoon when we were holding the write-a-thon together, and a ton of stuff got answered.

Yeah, it's really cool. I wish I would have. I was just in Honolulu a few weeks ago. It would have been nice to have this when I was there if I would have known about it. Yeah, I like this. This is, that's pretty... I did want to ask you, though, because some of the projects,

I've been just kind of clicking through a lot of the projects on the Code for America GitHub page, and one thing I did notice was the Adopt-A-Hydrant. Eric Michaelsover still is committing to that to this day. So that's pretty neat. He had the most recent commit.

Yeah, he's one of those guys that he's just like always coding all the time. He totally is. Probably half of the repositories I look at on GitHub, the most recent commit will be him. But no, I mean, so I've seen, you know, so I look at most of these repositories and the majority of them, you know, the ones I'm looking at now, anyway, closer to the top, the ones that are just kind of seems like they're getting up and running. The majority of the contributions are from what looks like to the team that's working on that city. How often do you attract just a non-Code for America fellow to come in and work on a project directly without forking it for a specific city? So for instance, like

Honolulu Answers, it seems like it has a lot more contributors than some of the other projects. So it would appear that maybe some developers that just have a heart for Honolulu or maybe some developers in Honolulu got wind of this and said, hey, I'm going to help out. And how often does stuff like that happen? Well, for the sort of primary fellowship projects, oftentimes those are sort of really sort of researched and run by the fellows themselves. And occasionally people will get pull requests and issues opened. And that's definitely happened a few times this year, but it doesn't happen all that often. I think we'd love to sort of structure the projects in a way that it could be a little bit more open to other contributions so people knew what to do. But one of the ways that people really can get involved and are really, really involved is with our

brigade program. It started up about a year ago. We realized that in a lot of cities around the country, there were sort of communities coming together to build civic technology on their own. We wanted to help them out. So in Oakland, there was Open Oakland. In New York City, there was the Open NY Forum. And now we help them out with what we call the Code for America Brigade, which is a sort of community volunteer-run group that lets just normal folks become civic hackers. And we have them up and running in a bunch of different cities around the country.

All of our fellowship cities are sort of working on building them right now. This is a way for just normal people to be engaged. Usually there's a hack night every week or two. Sometimes it's a little bit more tech-focused. Sometimes it's a little bit more sort of community organizing-focused. And some of those organizations have their own GitHubs open. A lot of times it's focused on redeploying and maintaining applications that come from other places.

This sort of civic tech community is a lot greater than just Code for America, though we like to help people get together as much as we can. And the brigades are a really awesome way to get more people involved. So you can check out more about brigades and see where we have them if you're just joining at brigade.codeforamerica.org. We're also right now in the middle of a civic coding campaign, the Great American Civic Hack. We're trying to get more coders involved with working on some specific repositories that we think could use a little bit of love. So if you check out that website I just mentioned, you'll see that there's a few different repositories that are think could use a little bit of love. So if you check out that website I just mentioned, you'll see that there's a few different repositories that are ready for your

contribution. So if you hear about this and want to get involved and maybe can't devote a year to be a member of the next year fellowship class, it's a really great way to get involved. You'd mentioned like the local scene, just learning from the Code for America Twitter account, you guys just retweeted something from right here in my neck of the woods. Our mayor here in Houston just announced an implementation of a City of Houston

hackathon project that's kind of neat that just came up. It's like budget boot camp, I guess. I just learned about this. And a few months back we had this hackathon here, And I was like, wow, this is really neat. I missed it, sadly, because I didn't hear the news. But it's so wild to see these kinds of things actually being spawned by local city governments. And then, like you said, Andrew, just kind of getting the general Joe, for lack of better terms, or Joette, involved in code and then also in civic hacking rather than just leveraging only the fellows. Yeah. I wanted to ask the project here, and I just kind of found this little – I could run down this little monkey – I don't know.

Rabbit trail. Gosh, I'm losing my words here. So Blight Status, looks like it started in Code for America, and it got popular with New Orleans, and they started a whole company, Civic Industries, and then they're kind of taking this project and expanding it into something even more. Is that a trend that you would like to see continue, or can you talk about the ongoing relationship you'll have with the three?

I imagine the three people that are starting Civic Industries were fellows. Yeah, all three of them were fellows last year working on the New Orleans team project, and they're all still here in the office. I think they're actually finishing up their incubation period pretty soon. But they were one of three startups, Local Data and Textizen were two others from that year, that were kind of the founding members of the incubation initiative that we have, where we essentially take what look like really promising applications from the fellows and turn them into full companies with a little bit of funding and runway. So you actually, okay, so you're almost, like you said, an incubator. So you're actually watching these and helping them grow, which is cool then.

So that's a way that you can kind of, you know, the fellowship program is really neat, and I think it gives the people a good, you know, the fellows like Ezra, a good base. But I think that to allow them and to help enable them to continue on is kind of a, it's even a greater calling, you know? Yeah, exactly. I mean, the fellowship is, you know, it's just the one year and it's really important and increasingly important over the years that the fellows that come and, you know, give their time and work with us for that year have some very concrete new things that

they can do and new contacts that they can make as a result of their time here. So we're really interested in figuring out, you know, what kind of companies they move on to, what kind of companies they create. And, you know, with luck, government becomes something that's actually attractive to all the different fellows that come through here. So we're really hoping to see more people go into government proper. How do you split the teams up? Like when fellows come in, you obviously don't want to end up with a team of three designers or three data monkeys. So how do you kind of split the teams up when it happens? I think it's a mysterious process involving a bulletin board and darts.

But no, we're all sort of, as fellows, we're given a chance to say what we're interested in, where we're interested in working and what kinds of skills we have. And then the staff sort of goes into a room and tries to take our preferences into account and build teams that have a good variety of skills on them. So it was a little bit interesting coming in without knowing the people I was going to be working with, but it's worked out pretty well. Yeah. What comes first, the fellows or the cities? Well, I think, you know, it's one of those situations where we all have to be able to get the work done. In terms of selection, right now we're in the process of choosing the cities for next year. And the fellowship application deadline is a few in the month. So we'll start reviewing them later this summer. Both will be announced later in the fall.

Well, and the reason I ask is, you know, I know that, you know, and in my circle of friends, I know a lot of people that have hearts for specific cities, right? Whether it's where your grandparents are from, where you're from, whatever the reason is. Let's say you get elected into the fellowship program, and for all of our listeners out there that have an interest in maybe just kind of wavering through life, I highly recommend looking into this. But if you get into the fellowship program and then you say, hey, I would really like to work with this city if that city is available. Is that kind of stuff taken into account at all? It's definitely taken into account, but I don't think, unfortunately, they can make any promises. So keep that in mind. If you only want to be in one place, it's difficult.

The truth is, is a lot of the folks, the average age of Code for Breckenfellows this year is, I believe, 30, maybe even a little bit older than that. We have folks who are sort of new to tech, people who've been around for a long time, people coming from big companies like Cisco and Google and Yahoo and places like that. So we have folks with an incredible breadth and depth of experiences. And a lot of folks are taking a really big pay cut to come here. This year we get paid $35,000, which for, you know, a lot of people in the tech industry doesn't sound like a whole lot, but people are here because they are really carry, they care a lot about what they're doing and they're passionate.

So, you know, we're really trying to bring in people who have a lot to offer. Yeah. And I mean, and so how, how many of the fellows do you get that are, that are working in another job and, and, you know and maybe deciding, hey, this is something that I just really have a passion for? And then how many of the applications are people that come out of college that have no idea where to start and are just trying to figure out how to get started? I don't think there's anybody this year who's coming straight out of college. Everybody's worked, has some significant experience or expertise that they're bringing to the table. We have a PhD anthropologist who's helping with user research.

We have folks who've worked in healthcare policy research and data analytics and all kinds of different fields. So maybe they're not too experienced on the tech side or on the startup side, but everybody comes in with something. That's not to say if you are coming straight from college, you don't have something to bring. Definitely encourage you to apply. If you have questions, you can ask about whether you're qualified. But we're looking for folks with a big passion and a big capacity to learn.

That's always good. That's always a good need right there to learn. That's certainly a bridge in that skill gap piece there, just bringing either learning on the job or for lack of better terms or helping educate those that are in cities that are using less modern things that would benefit from more modern things you know as one thing that was um uh that mike and i talked about briefly when we did sound check earlier um was just kind of going back to Chicago and their open data initiative that they kind of spawned back in March.

And you're there in New York. Mike mentioned you had some interesting things with what you have planned for New York. Can you share that with us? Yeah, I think New York is a really awesome place to be working on open data because about a year ago, a year and a half ago or so, they passed a law that says that all of their government data has to be open. And so you can go online right now to nyc.gov slash data. And there are thousands of different data sets, everything from 311 calls to GIS locations of subway entrances to all the different trees that they have kept track of all over the city. And you can build apps on top of them right now. It's just open to the public. And I think what you're going to see over the next few months is more and more

data sets coming from more and more agencies that just developers or anybody is going to be able to use for all kinds of different purposes. New York is not the only place that's happening. Chicago's done a lot of work. Seattle, San Francisco. A lot of cities are starting to create these open data portals and creating laws that make the government release this data. So it's a really exciting time if you're interested in building applications or using civic data. And I really encourage anybody who's interested just to go to their local city website and take a look. There's also work happening nationally at data.gov and many states. I know Maryland's doing a really great job and other states, New York State, that have

all kinds of data sets open. What about Texas? I don't know off the top of my head, but we do have a fellow from San Antonio. Awesome. Obviously, I'm a huge fan of Texas. Austin was one of our cities last year. Texas is usually so good at government, though,

so I figured they'd be the Chicago, you know? Bummer. The Chicago of states. Yeah. Well, the Chicago in terms of leading the way. I mean, they did a really cool thing with that. I mean, we had gotten such a huge traffic spike that day. I don't know if it was – and I think that's what's really interesting about this.

I think it's a different side of obviously the government and things that we as citizens are able to interface with. We don't have lack of power. We just have lack of knowledge, and this open data starts to give us knowledge to either make our cities better by doing maps or just doing really unique, cool things with that data. I mean, you just never know, and I think that's something I wouldn't mind doing. That's why I'm kind of bummed I missed Houston's hackathon that was back in May. So I'm excited to see the next one and get involved.

But, Andrew, we've got some cool questions we always ask. What are those questions? Yeah, so, I mean, I feel like we could talk about this stuff forever. But at some point we have to stop it. But the questions, yeah, for the sake of people driving home from work, the questions that we like to ask, and if you are a listener to The Change Log, you are ready for this. If not, we're going to ask these two questions at the end of every show. And seeing as we have two people on there with us, I will let you both answer. So it's your lucky day.

But our first question is for a kind of a call to arms. So in the projects you're working on or maybe any other projects that Code for America, you know, has worked on or just anything, what would you like to see the open source community get involved with and help out with?

And I guess I'll ask you first, Michael. Sure. Yeah, I think that really what we're looking for is help and participation from the open source community. We're trying to improve all the different readmes and documentation around our applications. So if you go to github.com slash codeforamerica, take a look there and see if there's any issues you can squash or stuff you can help with. Another place to look is brigade.codeforamerica.org.

Brigade is sort of a wider ranging, much larger program that we run with people that are local to something like, I want to say, 25 or maybe even more cities around the U.S. So take a look at Brigade and see if there's a group near you. What about you, Ezra? Anything specific? I would say if you're interested, apply for a Code for America fellowship. It's been a really awesome opportunity to move to San Francisco, meet awesome people. Tim O'Reilly is around really frequently. Eric Reese came and talked to us. We get to work with some really amazing government officials all over the country. It's a pretty good opportunity

if you're interested in tech, cities, urbanism, any of the sort of city-like stuff. So codeforamerica.org slash apply, and applications are being accepted through 731. I was going to say 731 was, so that's, that's another reason why I wanted to get you guys on the show pretty quickly. I know we kind of expedited your, we wouldn't,

we would have had you on the show no matter what, but definitely that timeline was what wanted me to get you on because we haven't mentioned yet, but we're going to not have a show next week. Andrew and I are both traveling up to our day jobs home office, so we'll be there doing some hacking ourselves. But I wanted to get you guys on before that so that we can, I hope, spike the interest and hope that this peaks some applications coming in.

So the deadline is July 31st, right, so the end of this month. That's right. Awesome. Let's start doing it now. Don't wait until the 30th. of this month. That's right. Awesome. Let's start doing it now. Don't wait until the 30th. That's right. That's right. The second question we ask is for a programming hero,

so somebody that has been influential. And I'll even let you guys answer a civic hero if you would like, but we typically ask for a programming hero. So, Michael, what do you got for us? Sure. So my programming hero is Guido van Rossum, the creator of the Python programming language. He's my programming hero both because he's pretty awesome at what he does and also because I think I've just seen him exercise so many moments of admirable restraint in designing the language where it's been created to be this extremely easy-to-use, beautiful, extensible thing.

Interestingly, Kenneth Reitz, who is one of our frequent co-hosts on the show, he also said that the BDFL is his programming hero. And what about you, Ezra? Well, I've had the luck to have a bunch of different really awesome mentors and a lot of people I really admire, so I'm going to go a little bit of a different direction. I'm going to choose my friend, Lily, who I know from college. And she didn't study computer science or do anything particularly techie. But since moving out to San Francisco,

she started working at a startup on their support role. And she's really spent a ton of time diving straight into learning web development. She's at Dev Bootcamp right now. And I'm just really inspired to see how she's put so much effort and energy into learning web development. She's at Dev Bootcamp right now, and I'm just really inspired to see how she's put so much effort and energy into learning the skill. And more than that, she's an active member of a great organization here in town called RailsBridge, which does free training programs

for people who want to learn web development in Ruby on Rails, particularly women, who, as we know, are not as common in our industry as they are in the rest of the world. So she's just been doing really great work learning and sharing. And I just think it's a great standard for us to all work towards. Yeah, one of our coworkers at Pure Charity, Beverly, and we've talked about her a few times, she's involved with RailsBridge. That's a very cool movement that we're seeing happen.

So, totally. Yeah, Beverly's also a fellow change lawyer. She hasn't written on this subject yet, but I've been asking her to talk about some of the things they've got going on at her chapter of Rails Bridge because she does the leading in. I think she's actually the national leader, isn't she, Andrew? Isn't she? Or she's on the board or something like that.

Yeah, she's such a great teacher, and she always has a heart for helping people learn. So definitely big fans of Rails Bridge. So, Ezra, make sure you get me a link to anywhere I can find Lily or help me with her last name so I can make sure we link her up in the show notes. Will do. All right. Well, gentlemen, thank you so much for joining us on the show. This has been episode

number 95 you can find this show at 5by5.tv slash changelog slash 95 show notes will be there as well we'll link everything up but Mike and Ezra thank you so much for taking time out of your day to share this fun topic of civic hacking and

the efforts that both you guys are doing at Code for America. And say hello to all the fellows and good luck on getting some really awesome new fellows in for your program for this year. But we'll be back not next week, but the week after for the live show every Tuesday at 5 here on 5x5. So let's say goodbye. Awesome. Thank you, Adam. Thank you, Andrew. Thanks so much. Great let's say goodbye. Awesome. Thank you, Adam. Thank you, Andrew. Thanks so much. Great chatting with you guys. you