Enterprise View

Bray talks ELC, change agents and future of government

As government chair of this month's ACT-IAC Executive Leadership Conference, FCC chief information officer David Bray very much has the future of public service on his mind.

Bray sat down to talk with Senior Staff Reporter Carten Cordell about how the conference will brainstorm better solutions for government, the benefits of cloud computing, the promise of open data and his advice for stakeholders seeking innovation.

Well, to start off talking about talking about ELC, what are you looking forward to this year? What kind of things are you rolling out that are new?

This year's Executive Leadership Conference, the timing in some respects is almost perfect. It's two weeks before the election and whatever change comes with it.

Now, I'm a nonpartisan senior executive, and part of what I think is so interesting is with a time of transition, this is a chance to have conversations both with those people that are in public service on the government side, however, also recognizing that public service also includes those members that are from the private sector that work to help us get things done differently and better and then, finally, members of the general public.

So this year's ELC, while it is the Executive Leadership Conference, I've been trying to sort of use that 'E' both for executives, but also to mean 'Edge,' in terms of edge leaders. Because the reality is anyone can be a leader anywhere whether they're a member of the public or their member of the private sector or in government, and what we really need to do is tee up ideas on how we can deliver results differently and better for the next administration 2017.

So what have we done to make this different? First, instead of it being sort of what used to be in the past — which were a lot of talks that were monologues and went about 40 or 50 minutes long — this year instead everyone that comes to the center stage events, [those] people get 12 minutes to make a pitch as to how they would deliver results differently and better and then gong will sound and then we'll get eight minutes for Q and A, then the gong will sound again.

For the morning, this is meant to be ideas that are currently already working in public service that could scale in 2017 by OMB and by other parts of the administration if they want to. And they will actually be a voting app, and the audience will get to vote and which ones are they think are the best will actually get to brief the next administration in 2017.

Then during lunch, we're going to have what we called 'Experience versus Edgy' debates. By experience, I mean people that are boomers, and the Edgy, I'm talking more the Gen X, Gen Y. And these individuals will be people that have been listening to the conversation so far that morning. They will get to go up on stage and share what differences they heard. There will be a chance to hear, 'Well, how did you see this and what were the differences and why?'

Then finally on Monday evening, in that afternoon period, what we are going to have is what we call 'Perfect Pitches.' Again, it's 12 minutes, but it's teams of three to five that make a pitch that's something that's completely new that the new administration could do in 2017. Again, a gong will sound, and this time, we actually have what we are calling the Shark Tank judges. They get to give their little comments and feedback and again, the audience votes. Again for these Perfect Pitches, whichever ideas get the most votes will also get to brief the next administration in 2017.

You mention the presidential transition, we are also currently in the midst of this massive IT transition and modernization. Do you see these events as sort of a bridge and how do you connect these initiatives between administrations?

So the IAC side of ACT-IAC got started in 1979. One of the things we are going to do to kick things off on Sunday afternoon is we actually are going to have a very brief slideshow — about five minutes — that is the history of ACT-IAC from 1979 to the present. That’s both the history of the organization, but more importantly, think about the different presidential transitions that have occurred. What were state of the arts at the time?

If you really step back about it, when the last presidential transition happened in 2008, most people didn’t have a smartphone. There was no expectation that you could get things through an app and actually engage your public through an app or engage public servants. So what will be the new expectations that will occur with this transition as well? We need to think about those.

And you are absolutely right that we’re sort of on not just a pace of fast change, but actually exponentially accelerating change. So one of the things we want to underscore on Sunday evening is that we are really on that path in which this hockey stick of exponential change is going [up]. We have a choice.

Public service is not designed to move fast, partly because they wanted checks and balances. The last thing the founders wanted was a king-like individual. Public service either figures out new ways of delivering public service differently and better in the next 10 years and we managed to move forward as a nation or we continue to have the same conversations, we don't make progress and 10 years from now we've missed that opportunity to sort of ride that disruptive wave and figure out new ways of delivering the results for the nation. And that really depends on the conversations we have here.

So the goal is to really sort of put that forward as a proposition for the conversations on Monday and Tuesday so there are ripple effects in 2017 and 2018. The last thing I want is [us] having the same conversations a year from now or two years from now.

What I'd love to do is experiment. I mean how many of us with a smartphone, if we could, share data anonymously on air quality, water quality, transportation quality to improve our local environment we probably would, but right now we're not doing that. We're not realizing that in today's era, it's not just that things have to be pushed by government professionals, things can be delivered by the private sector and things can be delivered by the public themselves standing up and choosing to do it if they want.

And so that's why our theme is ‘Change Agents delivering results differently and better.’ Success for me is the 800 people at ELC, that's just a ripple in the pond that then go to 8,000 and 80,000 change agents in public service.

In talking about Change Agents and also Edge Leaders. If you had to describe the skill set you would think would be crucial for those two components, what would you want to see in those?

When I look for a good change agent or a good edge leader, the first thing I always look for is the ability to first listen before they seek to be listened to or to first understand the landscape that they seek to disrupt before they disrupt it.

I've seen great cases. In fact, here at the FCC, most of our successful change agents, whether they come from private sector or other government agencies, the key to their success is they first identify what's going right, what's not going right and then they plan their disruption. It’s when you get someone that comes maybe from the West Coast or something like that and the first thing you try to do is try and make it as if it was whatever they came from. And it doesn't usually work out.

So most successful change agents, it’s when they come in and they take the time to understand the landscape they seek to actually change.

And so I've seen cases where people come in and they think they want to recreate maybe that start-up they came from or that dot-com. And while I applaud that sort of hunger to bring disruption, I myself back in 2009, I was actually trying to push agile development. The Agile Manifesto came in in March 2001, and I was being told by my agency, ‘No, buy into the five-year plan,’ and I was like, ‘Well no, I was actually with the Bioterrorism Preparedness Response program. We don't have a deal with the terrorists not to strike until we have something out,’ and it was literally on Sept. 11 when I was actually supposed to brief the CIA and the FBI as to what we would do, and thanks to agile development, we had something.

But I share that because if you're going to change a system, you have to recognize that there's probably decades of things behind the scenes, there's reasons why existing processes [are there.] It doesn't mean we have to repeat those processes, but we can change that. And so change agents seek to understand what you're going to change.

The other thing I look for it with a good change agent or a good edge leader, you're going to incur friction. All I ask is make sure that most of the friction that you're going to create is about moving the car forward versus heat and light. You need friction to make cars go forward, but there are some times when friction is just spinning its wheels and producing a whole lot of heat. That's when I'm really happy to work with a change agent to try and figure out how can we actually translate that into forward momentum.

So if you're a good listener, you understand how to make things change and you understand how to empathize with the people you're trying to disrupt and then your disruption actually gets to somewhere, that to me is the sign of success.

With disruption in the private sector, you've seen often times a development and whatever society turns around to catch up to the development. It seems like with government itself, you can't have that 'blown-up' disruption. So is it a conversation between both parties that is going to make those conversations happen?

Yes, I think so much of what we do in public service really does depend on engaging the industry and engaging the outside public. One of things I always say is the last thing we should be doing is writing custom code, that's not our skill set. Our skill set is about delivering services to the public.

So how can we use micro cloud solutions that are already out there, and we mix them in interesting ways. Whether they be processing a form in a more effective way, whether they be machine learning, if you actually get a response faster and in better time — whether it be showing things on a map better — we shouldn’t write that from scratch. I mean, that's not our forte. And so it does require engaging the private sector and saying, ‘Here's our challenges. How can we best work on them together,’ and it's a conversation.

I was recently at an event where a large brick and mortar company was talking about how they realized that if they didn't change their business model in the next three or four years they're going to go bankrupt. They also had challenges of culture change, they were very hierarchical. It was going to be a slow process. And so their solution was essentially to create a second organization and have that run as a competitor. When the second organization succeeded, that would sort of consume the old model. Or if it didn't work they'd write it off and try again.

We really can't do that in public service. I can't imagine going to Congress and say, ‘Look, we need funding for a second agency that's doing the same exact thing. We are just going to see if it works. If it doesn't, you write it off and we’ll try again.’ Congress would not allow that. We have to be good stewards of taxpayers' dollars. So how you do culture change is a little bit different in public service because you don't have seed funding, you don't have an investor capital fund. The Department of Defense cannot go bankrupt and say, ‘Let's do another round of [venture capital] funds.’

What we can do, however, is we can find safe spaces to experiment on those things that are possible, and if it succeeds, we show that and we say, ‘Look, this may fit your context. It might not, it depends on what you need in your mission edge, but this is what's working over here. If you want to copy it, great. If it doesn't fit exactly, try some other experiment and if that works, let us know.’ So we can begin to cross-pollinate across different groups.

The most important thing in my opinion, we need to stop thinking about hierarchies and we really need to think about how we do network effects. Because nowadays, no one problem fits into one department or agency. The challenges that we face as a nation, and as a world, really are sort of horizontal in nature. Yet how do we actually break out of these department and agency models and think about teams of change agents that span across sector?

Now heading to the presidential transition, you have very much been a spokesman for shifting off of legacy systems and updating IT. If you were to have a successor, what advice would you provide them to step into that role?

The good news is one, I’ve done succession planning. I think that's incumbent upon any good leader. Not that I'm planning to change, but if I did, the most important thing you can do as an effective champion of change agents is give people autonomy. If you take away people's autonomy, then you take away their sort of intentional intrinsic motivation to do a great job. And then just becomes a nine to five job and they're checking a box.

We've had great success where I say to people, ‘Look I'm here to be a digital diplomat and your human flak jacket. I know this is going to be a bumpy ride. I know there's going to be friction, however, I trust you. If you come to me and make a good pitch as to what we should do and why, I'm going to invest in you and give you autonomy. Run as fast as you can and learn from it. You're going to maybe hit some roadblocks and are going to have to adjust. Just run as fast you can, involve as many stakeholders as you can that you can bring along with you, but any flak is in my way.’

And I think that's something that you don't see enough, not just in the public sector, I think in a private sector, too there are jobs where people's autonomy is taken away and they’re no longer motivated. And so if there was one message I could share about public service to my future successor, people can do amazing things if you give them trust and an environment in which they can learn.

You know that's where I often say I don't really feel like a CIO. Sometimes I say therapist, but other times I just say I'm really just a champion of change and it just happens to be so much to change right now as digital.

Wherever I go next, I like to sort of understand and help people understand how can we move quickly and how can we deliver results and experiment. Because the only way we're going to figure out how to deal with this exponential change is do things differently and better. There's no immediate textbook, but let’s experiment as quickly as we can, let's involve our stakeholders on good ideas and let’s deliver on those results as we move forward.

Recently there has been a lot of talk about not only open data and its possibilities — which seem almost limitless — but things like transitioning off about legacy systems to facilitate open data. What do you see is the future of those two sort of conjoined movements?

Actually, that’s a really great question. I recently was invited to give a short, almost TED-like talk at the Executive office of the President on ‘Beyond open data.’ Because in my opinion, I've actually put forth two things: One, ‘federal government’ is a 20th-century word that we need to actually stop using. Because, again, if you think about it, in 1776 [with] packet latency, the time it took to send a message between New York and D.C. one-way was three days. Nowadays, it's milliseconds. We don't have to centralize everything.

Similarly, I think so much of what makes this nation succeed is really all of us. It’s when industry, plus local communities, plus government professionals are in alignment, we do great things. The challenge is often we're not in alignment. Maybe that's, in some respects, good because there’s checks and balances. However, I think the art of a good leader is having that conversation about how do I get everyone on board at the local level, with industry and the government officials to deliver results.

And so with open data. Nowadays, with smartphones, we have the equivalent of a supercomputer back in the 1980s in our pocket and we've got it for an amazingly low price. That could inform transportation, that could inform water quality if we had something that was anonymous and actually people had trust.

We've done at the FCC, back in 2013, we did a little experiment which said, ‘If you want to inform how fast your mobile broadband connection is, you can get the results yourself and you can share that anonymously with the FCC to make sure you know where there’s broadband spots, where there are gaps.’ That actually ended up where one, we saved money because we crowdsourced it, but even more importantly, it was about building trust.

You can imagine in 2013 saying, ‘Hi, I'm with the U.S. government. Would you like to actually download an app that I will monitor your broadband connection speed?’ You may not succeed. But what we did do is we made the app open source. By making open source, enough people took a look at the app. It was on GitHub and they said, ‘Well look, they're not collecting your IP address. They don't know who you're in a five-mile radius. It's privacy by design.’ And so that actually as a result made us the fourth-most downloaded app, right behind Google Chrome for a while on the iOS store.

Think about what recently happened in New York [with the capture of alleged bomber Ahmad Khan Rahami], essentially that was crowdsourcing, basically a community watch. So it's not just government making its data available, but the if public wants to share data and they volunteer it, they can also share their data and it can make our community safer and more resilient. That, to me, is actually what's needed beyond open data.

However, you're absolutely right. Right now, as long as we're using legacy systems we can’t seize on that. That's partly why at FCC, we've said we're going to 100 percent public cloud.

The other sort of disruption that's right on the horizon: Back in 2013, there was a competition to see if anyone could write an algorithm that would grade papers as good as a third-grade teacher and for $60,000, someone actually won that competition. In 2014, actually, there is a copier that came out that could actually take a handwritten test, handwritten answers and actually grade it for the teacher.

Now it’s at the third-grade level, so we're talking about sentences and math and things like that, but it's pretty impressive. In 2015, a company elects to its board of directors an algorithm and gives it voting rights. Also in 2015, [artificial intelligence] has been tested to say which meetings you should go to, which meetings you should skip and if you do go to meetings, who should go along with you to be the most productive. In 2016, I've actually seen competitions where they're actually trying to write an AI that can actually answer real estate law questions as good as real estate lawyers and are right now at about 80 percent accuracy. There're actually other cases in California where the jury setting bail is actually being done by an AI, which doesn't discriminate, it just looks at the facts.

So how much of what we do in public service right now is either prolonged in terms of it takes a very long time because we want to make it fair, so we have so many different people involved to reduce bias, or it's just a very cumbersome process, again, because we don't want any one person to bias it. If you had an artificial intelligence that was open source, you’d get answers in milliseconds and it could actually be a way that people could see that it can't be biased because here's what the algorithm is doing.

So I think the next major disruption, in my opinion, for public service is not just open data, but actually how do we get AI to take those rote things that we need to be made equitable, make them faster, deliver the results faster to the public and then that allows us to focus on what are those things that are so hard that we can't automate them at the moment.

I was reading an interview you had done about Operation Server Lift to shift the FCC data solutions to third party systems. You said that FCC was able to set up a cloud prototype within 48 hours. Now, that seems like the level of speed needed to develop agile government. What kinds of things would you advise other federal agency leaders to set up that kind of infrastructure?

So what you touched upon, Operation Server Lift, this is precisely when we made this intentional shift to go to a public cloud and commercial environment. Having things on premise generally is very expensive to do, but even more importantly, very slow when you need to actually set something up new. And so by going to public cloud, the gain that we've got first and foremost is agility such that we've actually had cases where bureaus and offices come in with a new idea, they give us requirements and two days later, we've actually got something for them.

That is why you have to get off doing things on premise. This is why you have to stop coding applications because that is just going to slow you down and you can't move with speed. The second thing that going to a cloud infrastructure gets you is that resiliency.

Just yesterday we had a system that involved with emergency alerts. There was huge interest, especially in the aftermath of the events that happened in New York two weeks ago, and so had that been on premise, we may not have had additional capacity. But because it was cloud, we could actually have some elasticities and expand to address the additional users.

Then finally, what cloud gets you is you're not really spending a lot of money to maintain those systems, that's what the vendor does. That's economy of scale. They also have way more cybersecurity people than we could ever possibly hire because they're a large firm. And so you get that sort of economy of scale and efficiency, so that instead of spending your money on trying to maintain the system you're focusing it on a new development, you're making changes as the mission changes and whenever a new requirement comes in from the Congress or the administration, you can actually almost literally turn on a dime as opposed to having all your money locked up and trying to maintain old systems.

Beyond the modernization of IT systems, why do you think it's important for the government to develop faster solutions?

If there was one thing that I really wish could be emphasized in the next transition, I think the public, in general, doesn’t recognize the difference between people that are political appointees versus career, and in some respects, they don't need to.

One of the things I celebrate in the United States is that we do have both people that are political appointees and career. Generally, the political appointees set the policy and set the political objectives and then there are those of us that are career senior executives.

Whether or not I'm doing public service 10 or 15 years from now, I don't know, but my main focus is less the politics and much more about just how do we get things done with speed. And I share that because so much of what I am greatly concerned with as I look at what's going on is if we don't figure out a way to continue to deliver public service with speed that meets the needs of the different stakeholders in the public, we may actually undermine trust in representative government.

And that's why again I'm trying to break the word and not say ‘federal government’ anymore. I'm trying to say intentionally ‘public service.’ Because otherwise, people might start asking for something that looks like a dictatorship or an oligarchy, which can move very fast. However, again, the founders wanted checks and balances.

I go back to James Madison who said he wanted ambition to check ambition. ‘What is government but the greatest reflection of humanity? If all men and women were angels, no government would be necessary.’ I share that because I think this is a conversation we need to have as a public, which is where we can we actually begin to experiment with new ways of doing public service and involving the public, not all of it just technology. Maybe some cases the public's helping make decisions.

Because otherwise, if we don't show that representative government, representative public service can move with speed, I am concerned about what the future has in store.

Recommended for you
Around The Web
Comments