Lawfare Daily: Hillary Hartley and David Eaves on 18F, Its Origin, Legacy, and Lesson

[Intro]

Hillary Hartley: The problem is that for decades, government outsourced its mission. And when you outsource your mission, that's when you're in trouble, because not only do you not have inside capacity to understand if the vendors are telling you that you're on the right track or what your actual problem is. You need people inside to understand your actual problem, what users actually need and what's gonna fix it. Then you go ask for that.

Kevin Frazier: It's the Lawfare Podcast. I'm Kevin Frazier, the AI Innovation and Law Fellow at Texas Law and a contributing editor at Lawfare, joined by Hillary Hartley, the former Chief Digital Officer of Ontario and former co-founder and deputy executive director at 18F, and David Eaves, associate professor of digital government and co-deputy director of the Institute for Innovation and Public Purpose at University College London.

David Eaves: Not everybody wanted them to come. Not everybody asked, but the places that did ask knew that they needed to figure something out and do something differently. They went where there were coalitions of the willing that wanted to change.

Kevin Frazier: Today we're talking about 18F, the now defunct digital services team that was tasked with updating and enhancing government systems and services.

[Main Podcast]

In a time of rapid changes in the federal government, you might have missed one of the most consequential shakeups: the closure of 18F. Between Senate speeches, agency closures, and mass layoffs, the dismantling of this small but mighty federal digital team barely made a ripple, but its loss may deprive the people of a wave of government savings and service improvements. So if you care about how government works—or more importantly, how it should work—this is one upheaval you need to understand. Today we're going to explore what made 18F special, why it faced resistance, and what lessons its story offers for the next iteration of digital government,

Hillary—onetime director, co-founder of 18F—let's start with you. What the heck is 18F and what did it do and what was your role there?

Hillary Hartley: Thanks Kevin. 18F was essentially a consultancy inside the U.S. federal government trying to build great digital services. And I used the word consultancy pretty specifically. You know, a lot of digital services teams have started since—GDS, the government digital service in the UK started since, USDS since—18F was always shaped a little differently and we always kind of had that consultancy bone in our body. You know, we really wanted to be the team that program owners could come to for, for advice, for help, for, you know, just encouragement on how to get started. And we would work with them to, you know, help them figure out their problems, figure out what they needed, and then build stuff.

Kevin Frazier: So, Hilary, you were obviously in the, the meat and bones of that effort of this consultancy.

David, what was your relationship to 18? How did you become familiar with this quirk of the federal government that people probably haven't heard of prior to its closure and perhaps haven't heard of since?

David Eaves: So I had been advising Code for America and had been responsible for the kind of education of the fellows in Code for America. And through that, when the Presidential Innovation Fellows program got started—which preceded all of this, but was a kind of first effort to bring, technology, talent into the government and, and, kinda start to think about how we could be doing things differently—I was invited to come and speak to the Presidential Innovation Fellows. In fact, the, the fellows requested me. I was the first person they requested to have come in. So that was like, that was a real privilege. And I was like, oh, these are really great and interesting people.

And then the following year, Jen Pahlka, who had founded Code for America, then moved to the White House and said, would I come and do kind of a two, like, you know, a one or two day training for the Presidential Innovation Fellows? And that's when I met many of the people who would go on to become kind of instrumental in the creation of 18F. That was not my first contact with 18F, but my first contact with the people who would make it happen.

And you know, some of them I knew from before, some of them I met there and I was just kind of following them in their journey, which continued that—they, they founded it. I wanted no role in doing that, but was always very supportive, but then I was brought in every year subsequently to help, you know, do a training program for the Presidential Innovation Fellows and had a lot of friends there. And then ultimately ended up at Harvard where I was convening digital teams from around the world. And naturally 18F was a, you know, core part of those people coming together.

Kevin Frazier: So of, of the people who know about 18F, I'm gonna put you all at the top 1% of 18F knowledge. So we are very lucky to have you help explain its importance and the significance of its closure. With respect to that significance, I wanna point out that if you pick up the Constitution as I do too frequently—or perhaps not enough, depending on who you ask—you won't find in Article I, Section Eight, some mention of Congress shall create a consultancy that will help it improve digital services. The founders apparently weren't ready for the digital era.

So Hillary, can you help me understand? When did 18F emerge? Did it just show up one day? People said, hey, I'm smart and I know tech, can I help the government? What gave rise to the creation of 18F?

Hillary Hartley: It was really a confluence of events and a lot of different factors. So I was a Presidential Innovation Fellow in the second cohort. So the, the first cohort Dave talked about was 2012, and then I was a, a fellow in 2013. And 2013 was a bit momentous in a few ways. One, there was a website that failed fairly spectacularly—

Kevin Frazier: What? No way. Not a federal government website. Truly, surely not.

Hillary Hartley: —that happened to be pretty critical to President Obama's mandate. So of course I'm talking about healthcare.gov.

Pretty much at the same time, if I remember the timing correctly, there was also a government shutdown, which I don't think had happened in a while. There, there were a lot of us furloughed for a little while. There were also folks that had been sent off to the rescue effort for healthcare.gov. And there was just this you know, this like itch, we can do this better, we know we can do this better.

There was a lot of kind of behind the scenes momentum in that period. But what I'll say is that, that by no means was sort of like the, the seed of all of this, because Jennifer Pahlka, who was the founder of Code for America and had worked with David and brought him in to train the fellows—she had been essentially brought into the White House as a Deputy Chief Technology Officer that year 2013. So we were serving at the same time.

And I, you know, one of her mandates was really think about—you know, she had gotten to know the people behind GDS in the U.K. She'd really been able to see the positives and the negatives of the of the fellowship model, and knew that really to drive this long term, we had to think about capacity differently and we needed to be thinking about in-house capacity. So that was really something she was focused on during 2013 when I was serving as a fellow. So by the time all of us were furloughed and sitting around and, you know, agitated and trying to figure out what we were gonna do next, there was always a, there was already a lot of groundwork laid for this to happen.

So a lot of fellows had been hired into the General Services Administration. My role was inside GSA, and there were several of us inside GSA that year, and the administrator, Dan Tangherlini, became a champion, you know, kind of in conversations with Jen and others about how are we gonna do this? For a while we were jokingly calling, you know, starting a team and calling them perma-PIFs. Like, you know, he was very interested in, okay, how do we get this capacity to stick around and do great work?

So that was really kind of how it happened in, in October and then through the end of the year, we started having serious conversations about what it would look like to get a group, you know, a cohort of folks to stick around beyond their, the end of their fellowship date and become perma-PIFs. That was the seed of it.

Kevin Frazier: I, I'm looking forward to one day finding on someone's resume, perma-PIF and being like, oh, I talked to Hillary Hartley. I know what that means. Congratulations.

But I I, what's really intriguing to me is you're pointing out, obviously the severe need—if we have healthcare.gov, crash, alarm bells going off across the country, we can't make a website, right? And that's just obviously gave rise to a clear need for something like assistance from smart tech folks to come in. And what's crucial here is it, it wasn't a call to Booz Allen, or it wasn't a call to McKinsey; it was a call to folks already in the government.

But Dave, can you, can you help us understand why the existence of talent being there already, already having those PIFs, those Presidential Innovation Fellows in place already having kind of tech savvy people be a part of the government was so essential to getting 18F off the ground to begin with and then to, its kind of early successes.

David Eaves: Yeah, so two things I wanna talk about first is I, I do wanna just say there's like an international context that matters here as well. Like it wasn't just the United States that was facing this problem.

I wanna remind everybody, like we think of healthcare.gov as like kind of recent history, but if you're a 25-year-old like in my class doing a Masters of Public Administration, you were like 12 years old when healthcare.gov happened. Like it's like, it's as deep history as like the 1970s as to, you know, people listening to this, to the podcast.

But I, so I just wanna remind some of the listeners on this call, like this was a transformational moment where there was a real crisis and a core policy that was gonna benefit millions, tens of millions of people could not happen because the technology was not working. It really presented a crisis of confidence in the government's ability to execute.

And the United States was not the only country to have this problem. There had been a similar crisis in the U.K. actually involving healthcare as well, that happened about 10 years earlier or so, where they spent like close to $10 billion on a healthcare record system that basically didn't work out. Like the money just was all wasted.

And it also, it was a huge crisis of confidence to the government and there was this realization that we can't just outsource everything. We have to have some capacity—if only to understand how to define the problems and choose people to do the work—we have to have some internal capacity, and it may be that we need to have some capacity that actually can build it. It might be cheaper and better and easier if we build some things in house. Partly just to do with that, but partly just to keep the capacity so we can evaluate the people we are outsourcing more effectively.

And so that realization was starting to dawn, but I think the just—really, really kinda the big thing was going on was governments were finally wrestling with the fact that the internet was a thing, that people expected to get services on it. And they were very slowly realizing that the entire government actually runs on technology. And now citizens are interfacing directly with that technology to get services, and we don't have a ton of models or ideas about how to operate in this way.

Kevin Frazier: And Hilary, it's hard to have this conversation and not mention DOGE, right? I'm sure for a lot of listeners right now, the thing that's already come to mind is, oh, so you're telling me a bunch of tech oriented people showed up to the government one day and said, hi, we're here to fix things, and we know more about the latest technology than you do.

What is distinct about how 18F operated as opposed to how DOGE is operating right now? And to the extent you wanna explore DOGE and comment on that, that's fine. If you'd prefer to focus explicitly on how 18F functioned that's also fine—I'll, I'll, I'll leave it to you.

Hillary Hartley: I will be clear that I, I, I don't know if I know how DOGE is operating, and I don't know if they know how they're operating in terms of this work, in terms of what they're gonna do in terms of building. I've been paying attention to some of the news I've been following along with what the folks inside GSA currently are talking about, and really they're putting all of their eggs in the AI basket, and that's a whole other podcast.

And so I, I, you know, I don't know that I know enough to understand really what they're thinking or how they're operating, but I do know that 18F, USDS, public servants writ large—when you go into public service, you take an oath and you understand, you know, you're, you understand probably the the pay cut maybe that you took, you understand that you are there for a, a bigger mission.

And that was we, we felt that. You know, it, it is one of the reasons that so many of us wanted to, to stick around after our fellowships. You know, I was on a leave of absence from a technology company and didn't wanna leave. You know, I didn't wanna leave. I, I guess I didn't wanna go back. I wanted to stay and, you know, keep making a difference inside government. You know, it takes a lot to push those boulders uphill, but it's worth it.

And I think one of the things that we all recognized was that, you know, David talked about, you know, outsourcing capacity. But the problem is that for decades, government outsourced its mission. And when you outsource your mission, that's when you're in trouble, because not only do you not have inside capacity to understand if the vendors are telling you that you're on the right track or what your actual problem is. You need people inside to understand your actual problem, what users actually need and what's gonna fix it. Then you go ask for that.  And government didn't have that for a long time.

And you know, if there, if there's a kernel of what I think all of these groups are trying to do yet we're building things, building some awesome things sometimes. But I think that's like the thing that if we can turn that Titanic around, that's the thing that really shifts the conversation, and that's what I don't know if DOGE really understands.

Kevin Frazier: Yeah. Dave, you wanna jump in here?

David Eaves: Yeah, I love this thing about the mission. I think, again, I wanna link it back to like understanding how government works in a digital era. Because you have to understand, like we've had computers in government since the 1950s. It's mostly been running back office systems.

What shifted kind of mid-2000s and then particularly took off after 2010, was this notion that people were gonna interact directly with their government over a website or a mobile device. They understand—like in the 50s, 60s, 70s, 80s, if you interacted with a citizen, they came into your office.  So you, you learn something about them. Like you knew these people, like you knew what their needs were, you knew what their problems were, you knew what the edge cases looked like, and, and the people who were like the public servants sitting in those office, they knew how to navigate those edge cases, they knew how to deal with people.

But when everything moves to a website, the people get abstracted away. You don't actually know where they're struggling unless you are building monitoring into your website. You, you don't know where the edge cases are. You don't know who's dropping out, and it's, it's, it's this weird, like this implicit skill of understanding people which existed distributed in these offices disappeared and you now have to make this implicit skill an explicit skill of the government. And so part of what, what Hillary's talking about is like you don't know your mission. Part, is you don't know your citizens anymore. You don't know your residents. You don't know the, the public because you are not touching them.

It's weird, like in the, in the digital era, we have to build all these new skills, part of which are very technology focused, but part of 'em actually are very, very human focused because we have to make the human part even more explicit.

Kevin Frazier: I do wanna—and I apologize, I promise this won't just become the DOGE podcast—but something I wanna highlight that Hillary I, I'd love your comments on, would be I think, everything all else equal, would say, do you want more tech savvy people in government? They'd say yes. The second question now may be do you wanna make sure they are accountable and transparent in their actions? They would also say hell yes or some variant of that.

So can you describe a little bit more about how 18F interacted with agencies, how there was a sort of heightened accountability and transparency built into its work to give people an understanding of what it actually meant for the 18F to to show up at an agency's door.

Hillary Hartley: Yeah, I think it's a couple factors. One is from the moment we really wrote our first or second blog post, we talked about being an open source team.

Kevin Frazier: Well, I just wanna pause there, sorry. You had a blog.

Hillary Hartley: Yes.

Kevin Frazier: Let's just focus on that, right? Let's, let's point that out, right? Openly accessible; I'm guessing you wrote in plain language, right; available to anyone who wanted it. So even that fact I think is worth calling out. But please continue.

Hillary Hartley: Well, you're exactly right. So we had a blog and you know, we had to go through GSA comms occasionally to get things published, but they were usually really happy to work with us on what we wanted to put out.

And our, and our thesis around the blog was building on what we had seen, you know, be successful in other jurisdictions was that it was a tool for us to document kind of how we were doing and what we were learning, but to share that, you know, so that, and to try to be open about successes, failures. Nobody ever blogs enough about their failures, but we sort of tried to talk about the meet. We tried to talk about the work, not just put press releases out.

And so that really was at the heart of the, at the 18F blog was, was the work. You know, here's how it's going, here's what we're doing, here are the tools we're using. And it's powerful as a transparency tool. I'll be honest, it's powerful as a recruiting tool, you know. I mean, even simple things like you, you know, maybe you see a photo in one of the blog posts and somebody is using a Mac and people think the government teams are using Mac. I, I didn't know. So it's, but it's little things like that.

So one was we tried to be very public and we were an open source team. Everything that we did started as a GitHub repository. Whether it was a, you know, a, an idea that somebody had or whether we were starting a new project, there was a repository for it. The public could follow along and comment and code along with us if they wanted to. So, open source was really at the heart of our value system.

And then, you know, the second piece in terms of transparency to our partners and other agencies I think was the fact that we, you know, we weren't sent to their door because something was on fire. We weren't that team.  But we were a team in GSA, acting like a consultancy.

So the General Services Administration does a lot of business on behalf of other agencies in a lot of ways. You know, they're the, they're the landlord for the federal government. They're the buyer of everything from pencils to tanks for the federal government, so they act on behalf of the federal government and often there are, there are fees associated with that. There's all kinds of ways that GSA operates. And so we were in that machine, which simply by that fact, made us different.

And, you know, the way we were created and funded was through a fund inside GSA called the Acquisition Services Fund, and it's essentially their, their fund that allows them to be a buyer, make investments, and know that they're gonna get paid back for those. And so since we were funded by that, we were kind of beholden to that same, you know, statute, which was, this is money, but it's gotta get paid back.

So, we had a relationship with partners and partner agencies where we would do work with them, but we were consultants. They had to pay us. So we, we, you know, we did the work for a fee. And I think that creates, it created both accountability to our partners because there is kind of this client relationship, but it also meant that those folks had skin in the game in a different way. You know, we weren't just showing up at their door with, you know, our fire hats on, but we were showing up, like really wanting to help them.

Kevin Frazier: I don't know. I mean, showing up in fire hats could be a compelling thing. Just to, to send a message.

Dave. I, I'm sure listeners at this point are like, wow, 18F great—I probably should have known about it before this podcast, glad it existed. very sad it now doesn't exist—but are still left asking well, what did it actually do? Give, give me an example of a project or a project or two where 18F was actually doing the work we've talked about. What did that work actually look like, and critically, as you all point out in your forthcoming Lawfare piece, how was that work essential to then setting up the government generally as having positive externalities on the government work going forward.

David Eaves: Yeah, so I love this question. So maybe the first thing just to kind of maybe also contextualize even more of kinda what Hillary was just talking about, which is, so I've been studying governments around the world and how they're kind of grappling with this problem, and, and, and the U.S. government was kind of unique and it had two teams. Not just one, but two teams that were kind of doing this.

And, and, and one of the things that's so powerful is, you know, you have this other team called the United States Digital Service, which is where DOGE, DOGE has kind of taken over that part. And it was a group in its early days, it was really, it was the, the group with the fire hat. Like they would go often, sometimes when directed to projects that were in crisis.

And I kind of jokingly in the piece, I was talking to 'em like they're the Marine Corps. They kind of like, they show up and they're like, well, we're here to help. And if you kind of say, well, I don't think I need any help, and you're like, then you're really not gonna like the next phone call you get like, it's like we're, we're not really asking. That was very powerful. And it gave them, you know, real leverage.  But they were really there to go solve a problem, you know, whether you liked it or not.

Whereas I think what was so powerful with 18F is they were the Peace Corps. They only went where they were invited and they were gonna work with you and they were only gonna work with you if you asked 'em to come in. And so not everybody wanted them to come. Not everybody asked, but the places that did ask knew that they needed to figure something out and do something differently. And so you had kind of like, they went where there were coalitions of the willing that wanted to change. And I think like that's a really wonderful story. And, and, and part of what makes it wonderful is that they changed the way people worked and thought.

It's hilarious. Hillary doesn't even know this. Just yesterday I was doing an interview for another case study that I'm writing about DOT Flow, the Department of Transports flow project, which is this kind of data, data trust that they've created around shipping data that was used to help prevent the backlogs at the ports. So like nobody knew what was going on the ports, nobody had any data, and so they created this kind of data trust where the shipping companies actually trusted them to hold the data so they could analyze and predict when there might be problems.

And in the middle of this interview, a DOT employee suddenly says, yes. We invited 18F in and they walked us through the way to think about this. 'Cause they're like, we are regulators. And suddenly we weren't using our regulatory tools. We had to go build something that people wanted to use, and that had to be compelling and interesting to them. They were gonna have to volunteer to participate and we just had no way of thinking about that. And I, I wish, I wish I almost had this person here 'cause they were genuinely transformed by the experience, and, and the way it made them think about their job and the way they interact with the people they serve and the way they think about their relationship with citizens and the way they think about their relationship with technology. So that for me is a great example of kinda like the, the power of what 18F does when it was operating its best.

But the second thing that I really want to hammer home on, which I don't think was intended when 18F started, was the way it evolved to become something I think even more powerful than initially conceived. So what we've really talked about are these kind of interventions. And so 18F would get invited in and it would do kind of project work, and that's where it started.

But I think the thing that then happened is that the team started to see patterns. Like the problems didn't just occur once. It was like, oh, we had the same problem here and the same problem here, the same problem. So then they began to write guidance. Hey, what? What is the way that we can solve this problem to help their own teams? But then they're like, because that's, Hillary said, let's make it all open. So then they shared with the whole, the whole world, not just the government, the whole world. And that helped scale these solutions.

And then they said, wow, we keep building the same solutions over and over again to solve these problems. So then they started to make products. And so now you had a product line where you were starting to build products that people were using and they opened those products up so anybody, even people they weren't working with, could come and download and use those products.

Then some of those products became so valuable and so important that they eventually kind of transformed into a kind of infrastructure where they were core tools that the government could use and rely on that were being run internally at a fraction of the cost of procuring them.

And I think that evolution—of taking things from advice to consulting, to product, and then to infrastructure—is almost unique. From what I've seen around the world, there are people who get into the infrastructure phase, but it's through a different but very, very unique story that I think is a, a, a particular contribution of 18F.

Kevin Frazier: And without getting too political, because we've, we've talked about a lot of positives for, at 18F. Hillary, I'm, I'm gonna put you in a tough position because I'm a law professor and I get to cold call on students and it scares them, and now I'm partially cold calling on you. So, what's the argument against 18F? Who, who was opposing this work in terms of creating novel products, in terms of collaborating with agencies, identifying these positive ways forward?

So if you could walk us through both when 18F was at its zenith, things are going best, who's opposing 18F in that scenario? And then if you could also walk us through 2025. What pushed 18F onto that cut list and ultimately resulted in its demise?

Hillary Hartley: Yeah. If you do a Google search for 18F—well now the results are all about what happened in March. But before that you would see, you know, the kind of articles that would run the gamut from talking about the incredible work that the team shipped in terms of the kinds of things that David was just talking about. So, building cloud.gov, you know, a platform as a service for the government. Just incredible. Login.gov analytics.usa.gov. And then, you know, not to mention all of the sort of program specific work that the team did over 10 years. So incredible work. Lots of coverage on that.

But two things, kind of always dogged the team. One was just, there was resistance from the beginning, from external D.C. media vendors, you know, to see this group of kids come in wearing their flip flops and shaking things up. And you know, there were—even though I truly do not believe it was warranted—-there was always chatter about bringing these kids in from Silicon Valley who think they know better than us. You know, and the, the number of folks actually from Silicon Valley was pretty small.

But that was the, that was the mentality of, you know, where are you bringing these folks in? They don't know any better than us. They don't know our business. So folks external to government vendors, D.C. media, were more than happy to pile onto that because there was, I think, just resistance to a team of smart, capable folks inside government building things at a fraction of the cost. So there was resistance whether it was from vendor community or from media, or just from other parts of—well, if I'm honest, other parts of GSA.

We always tried to build a narrative to overcome that. And, you know, and to really go into meetings with current and new partners, you know, in a way that always showed empathy, always showed that like, you know, your business, here's what we know, let's work together because that's gonna make magic.

Kevin Frazier: If I were listening to this and I hadn't heard you say the words 18F and I just really want you to double down on this because it'll tee us up nicely for, for the end here. Which is, so you mentioned a bunch of kids in flip flops. They're younger, they say they know how to fix things. They're kind of, pointing out wild suggestions that no one's heard of.

Can you just describe again, so we have—you all were transparent with the blog, you only came at the invitation of the agency. Is there anything else that you really think made the 18F distinct from a sort of move fast and break things, because that may be the connotation that people have coming into their mind. What else made it a partner more so than the angry consultant? Let's imagine a consultancy that comes in and says, we're just gonna tell you what to do. Are there any other sort of formal or informal safeguards that you saw that made 18F more in that collaborative fashion?

Hillary Hartley: Yeah. 18F and USDS both often started with, which is kind of a, an industry term—if you're building products in the private sector—but we would often start with a discovery sprint. So that's a, you know, it's, it's a couple weeks where you're just really getting to know a partner, their problem, some of the users, et cetera. Like you're just trying to wrap your head around the problem. You're doing some user research. Maybe during that time you, you do a couple paper prototypes so you can try something out with actual users, but, but the idea is that you're trying to wrap your head around the problem. You're not coming in claiming, claiming to know how to fix whatever you're there to do.

And I think that was, that's an empathy tool. It's an empathy building tool, you know, and it is what good product developers do. You know, they, they do use the research. They try to wrap their head around the problem, and they identify the user stories that are going to actually create good outcomes for people.

Kevin Frazier: I, I really appreciate that point because it, it really crystallizes the difference between identifying a solution and telling the agency how to implement it, versus hearing about a problem and working through what the solution could look like. And I think that's a, a, a critical distinction.

Hillary Hartley: Yeah. And we didn't have anything you know, out of the box or that we'd built before, you know, so we, we were different from vendors. We weren't coming in saying, we've done this with this thing before, and all you have to do is turn it on. We know your problem so well, just turn it on. So it was a big difference.

Kevin Frazier: And Dave, I, I cut Hillary off earlier. I'd love you to build upon that and if you could also talk about what ultimately motivated the demise of 18F in March?

David Eaves: I do think like the critiques of 18F—I think it depends on where you're standing, but there, I think there are some, there's always some challenges teams like this face.

So the first is, is if you have a group of people where you say, this is kind of where the newness is happening, this is where the innovation is happening, this is where the new practices are, it is an implicit—it runs the risk of sounding like an implicit critique of everybody else. And, and a lot of other people, they're doing great work. They're struggling in the trenches and suddenly there's this other group that suddenly has a spotlight on them and is maybe given different resources—maybe it's even given more latitude—and that can breed resentment. And this is a challenge. Any innovation team, I mean, any enterprise, but especially in government, will suffer from this challenge. And so you actually have to kind of work harder to be friendlier. You have to work harder to be humble, you gotta work harder to kind of spread the, the kind of the, the, the limelight in order to not be perceived that way. So I think there's like, there's always that challenge that you face.

And then the other challenges 18F and USDS, but I think 18F in particular, really was challenging a consensus position that had emerged in many governments, which is the governments weren't really competent, they couldn't do things, and that things are best outsourced. This notion that actually it might be cheaper and more effective and create more value if you build things in the house, was very threatening to a lot of people and you couldn't allow that to succeed. 'cause if it, if it succeeded in this domain, my gosh, that that blows the entire thesis of this, of this consensus apart. Maybe government's gonna start building more capacity in other domains too. And so there were a set of private sector vendors that were deeply concerned about its mere existence even before it had any successes because any success would present a threat.

Now, I don't think you should underestimate the challenge that those two things posed. I think the sad truth is—like, it's funny, I was joking around earlier, like I've got my Clue envelope here and I think it says DOGE in the executive office with an exchange server, and I think I just won. But I think the sad truth of the demise is 18F was filled with—like, there's two ways of cutting what's happening with DOGE.

One is there's just a kind of let's shrink government. But there, there is a belief out there that we're gonna go after the parts of government where it is believed people who believe in state capacity and are like more skewed towards the left tend to be hanging out. And there are tweets and quotes from Musk basically saying, this team we've gotta get rid of 'cause it's filled with kind of like basically Democrats, are left leaning people, which I don't think was entirely true and I don't think was fair.

And so I think that—my, my big fear is that more than any kind of rational assessment about the capacity is what drove it, which is kind of doubly ironic because actually there is an element of this, of this administration that actually does believe to some degree, small degree in state capacity in certain ways. And so this team could have actually been serving that interest really, really well.

Hillary Hartley: And if I could, you know, quickly, Kevin, what, what David started with there in terms of you know, 18F, these digital service teams, I think kind of writ large, but their attitude is not always like move fast and break things, but it is move fast and test things. And it is a, a fundamental belief that we can do better, that government can do better, that we can deliver better service.

And I think that in and of itself, you know, it attracts a certain kind of person. It builds a certain kind of team. And you know, I think it was, 18F and a lot of digital service teams probably have been filled with folks that, like David started with, have just kind of said, this isn't going well and we need to do it better. And that's that's just a hard message I think for some folks to hear.

David Eaves: When we did the PIF trainings, we’d talk about—actually, I would do a lot of work with the PIFs and we talked about like the three archetypes of the PIFs, which is actually a little bit of the archetypes of the type of people who end up in 18F.

But, but with the PIFs in particular, there were three. There was the careerist who was like, this will just look good on my, my CV. I'm gonna show up, I'm gonna do a year of duty, you know, I, I feel some, you know, some sense of responsibility, I'm gonna do a year of duty and then I'm gonna move on to the next thing and it'll be great.

And then there were the, the believers. People who were like, I believe in product management. I believe in design, I believe in open source. And they were like, I'm gonna come and inject this belief into government 'cause I think it'll be better. And these people really believed that they could help make government better by bringing this knowledge they had and unleashing it inside this beast and making it better.

And, and then there were the patriots who were like I'll never be asked to join the Marine Corps—I'm not in good enough shape or I'm just like not the right person, that's not me, I'm not gonna serve that way—but I'm here to serve.

And you really saw those three people, and what was interesting is they would change. Like sometimes people would arrive as careers and be like, let's here for a year, and then they would become patriots. Like they'd do the work and they're like, I'm, I'm sold. Like sign me up for 20 years, I'm in. And some people would come as a, you know, as a believer and then they'd be like, they'd become a patriot or maybe they'd just become a careerist. They’d be like, this is much harder, this was harder than I thought. I'm gonna do my year, inject what I can and I'm gonna get out. But, but people, they showed up wanting to serve in some capacity.

Kevin Frazier: Unfortunately we have to move into the rapid question. Dave, I'm gonna start with you and I'm gonna give Hillary the last word here. So Dave, the 48th president of the United States shows up and says, Dave, I wanna restart 18F. What's one reason I should and what's one reform I should consider?

David Eaves: That's a great question. So I think the answer is definitely yes, and the reason you should is that pipeline that I talked about, about moving things from discovery and kind of advice to infrastructure is so important for a digital era government. We have to have government that—that was like a sensory, it was a sensory thing; it was go and find problems, identify problems, categorize and identify what problems had patterns, and then figure out how to solve those, those problems at scale, like how to address those patterns at scale, is such an important function to have inside any enterprise and most important inside the government and an organization as large as the United States government.

So that for me is like a very compelling reason. And that thing pays off dividends in the short term through small projects, but over a 10 year or 20 year period can pay off huge dividends as, as that infrastructure builds up and starts to solve lots of problems at scale. That's my, that's the big advice I'd give.

I think the, the thing that I would do differently is I would be pushing for ways to rethink how we do hiring, because the problem of at 18F is actually that it shouldn't just be in18F. We need to unleash people across the government. There's lots of great talent already inside the government that I think is constrained, but we also make it easier for new talent to come in. And I, that's never a fun thing for administration to do 'cause it doesn't get you votes, but it's the kind of thing that builds capacity, that makes government more effective.

Kevin Frazier: Okay, so Dave's calling for an 18G, an 18H, and an 18I. Hillary, how about you? One reason to, to restart it and one reform you'd consider.

Hillary Hartley: I think the reason to restart it is pretty simply just about delivery. You know, you, you need in-house talent, wrapping their heads around the problem and telling you where to go from there. You don't need vendors telling you what your problem is and telling you this is the best way to solve that problem.

Kevin Frazier: I, I don't know. New Yorkers might say, hey, they helped us point out that our trash cans aren't as readily available as they need to be. That was facetious, but yes, this is good, good.

Hillary Hartley: But, but I think that's, for me, that's the bottom line. It’s just back to that line—it was maybe Robin Carnahan that I stole it from—but government has outsourced its mission, and we have to insource it. We have to take accountability for outcomes; we have to understand what the problems are; and we have to do that in a user centered way, so that what we, the policies create, we create and the tools to implement those policies meet the needs. 'Cause that's, that's product development, but that's product development in government, and that's how it's different. It’s this kind of marriage of policy and delivery that various teams sort of skew toward one or the other and, and try to get it right. But I think we can continue to work on, on getting those two things to, to ship together.

What would I change? I think in addition to just really needing to break hiring for the federal government, and maybe that's the lasting thing that DOGE has, has left us with—you know, the, that there will be a lot that needs to be rebuilt. I think the creation of, of teams is what David was talking about. Like I think a centralized team that can drive that, that need around the marriage of policy and delivery is, is still super necessary, but we do have to be thinking about more diversified teams.

And in order to do that, you gotta think about a diversified funding model. So, you know, we can't just pray that each administration is gonna work us into appropriations and is gonna give us money. I mean, even, I think it was in October, USDS actually crossed the bridge from being fully appropriated into like doing some cost recovery work because they weren't sure about their budget. So we do have, we have to think about diversified funding and what's gonna lead to longevity.

Kevin Frazier: Praying for money hasn't worked for me yet, so I think that's a a particularly good reform. With that, Dave, Hillary, thank you so much for joining. We'll have to leave it there, but I trust we'll find some way to bring you both on again at some point down the road. So thanks folks.

David Eaves: Thank you.

Kevin Frazier: The Lawfare Podcast is produced in cooperation with the Brookings Institution. You can get ad-free versions of this and other lawfare podcasts by becoming a Lawfare material supporter at our website, lawfare media.org/support. You'll also get access to special events and other content available only to our supporters.

Please rate and review us wherever you get your podcasts. Look for our other podcasts, including Rational Security, Allies, The Aftermath, and Escalation, our latest Lawfare Presents podcast series about the war in Ukraine. Check out our written work at lawfaremedia.org.

The podcast is edited by Jen Patja. Our theme song is from Alibi Music. As always, thank you for listening.