Rusty Pickens, the former acting director of new media technologies at the White House, recently moved to the State Department's International Information Programs (IIP) Bureau, where he serves as senior adviser for digital platforms. Pickens sat down with Federal Times Senior Reporter Aaron Boyd to talk about digitizing citizen engagement with the White House and how he plans to apply the lessons-learned to communicating American ideals abroad.
What did the IT architecture look like when you got to government in 2009?
Very big, heavy systems on premise. Not a lot of laptops, even. A government standard issued BlackBerry -- that was mobile then. Lots of big desktops, lots of big servers, lots of big data centers.
From what I hear and also from when I arrived at the White House in 2011, it was very similar. Lots of on-premise systems for messaging and SharePoint for collaboration and stuff like that. Not to the cloud yet.
We were using some of that stuff on the campaign back in those days for email messaging and text and things, a lot of online fundraising. So we were sort of there a little earlier on some of that stuff.
Then coming into government was a very, very stark contrast. So much old-school enterprise IT at that point. It's taken years for the government to adapt. So that was sort of an awakening for all of us tech folks, as there are now many more constraints that you didn't have to deal with back on the campaign days.
We run a tighter ship, but that means things move slower and the rules need to catch up. You're not going to have the ability to just stand up systems overnight like you might have been able to do in the past.
Why not? For people who came in knowing what needed to happen tech wise, what slowed you down?
I would say, first of all, the infrastructure just wasn't there. So even things such as server virtualization, really standard things that, now in 2015 you think about are just standard practice for an IT shop, weren't available at that time. They were bare-metal servers with operating systems installed on them and it was older coding technology -- ColdFusion and Java and .Net a few versions back. There wasn't PHP and HX and Ruby and all of this stuff that is now just part of the standard wheelhouse of Web 2.0, so we didn't have the tools at our disposal.
Even if we had known how to use them at that point, you really couldn't access them or get to them, so you had to start with exactly where you could in the moment of joining those agencies and build out the infrastructure and create the world in which you wanted to live.
What did the tech enterprise at the White House look like when you left in October?
When I left in October, it was leaps and bounds beyond where it was when we came in. I joined in November of 2011 and left in October 2015, so, a good almost four years there. Lots of improvements on the cloud side.
Our team -- my team in specific -- was responsible for all of the cloud infrastructure properties at the White House in development capacity and infrastructure to support the digital strategies teams and the forward faces of the White House. Those things had become vastly improved. When we came in, if there were things that weren't on premise at that time, they were legacy-managed services you might call them today. We were paying a contractor to maintain the servers for us, maybe they were on site and government hands on them -- that was about as close to cloud as it got at that point.
One of our larger accomplishments was moving the entire backend of WhiteHouse.gov over onto actual web services that are cloud based, then establishing things that we just didn't have the option for before like fail-over sites. So in the event that this instance goes down, you can actually spin up the second site in very short orders and there are no outages. So service levels vastly improved right away just from moving over onto industry-standard cloud service providers.
[We did] other things like improving the security posture. We were one of the first agencies to participate in the HTTPS by default initiative alongside GSA and the Federal Trade Commission.
Some of those things are vast improvements over what they were before. So even though content on WhiteHouse.gov and most of your agency websites is unclassified, low-side, publicly accessible information, it's just a better best practice to secure that out of the gate from day one and respect the citizens privacy. You talk about user-centered design, this is almost privacy-centered security as well, just making sure that HTTPS, by default, is the way we handle all web traffic, period. Those are two marked improvements I would say.
One of the big projects on which I worked was the president's correspondence system. It was in a similar state: A legacy-managed service that was hosted by a vendor and we had really just outgrown it. It was designed for older approaches to correspondence and wasn't able to quite handle the volume.
President Obama was really keen on leaning forward into the digital age, so we stood up websites and drove our traffic for communicating with the president: Send your message in through the website and we will respond to you digitally. That lowered the barrier to entry for people to be able to submit a message to the president; therefore increasing the output and the delivery that we were pushing through the system.
The older system was sort of crumbling and it wasn't able to keep up with the load. So the project [manager] was building that system from scratch on an industry standard best-of-breed customer-relationship management system that the rest of the private sector is using to much acclaim and is standard fare these days. But we rebranded it and put a citizen-management spin on it and really tried to focus that 360 view around how we are interacting with a person. Have they written us a letter? Have they attended a speech that the president's been? Has he written them a letter back? Just simple data points along the way to maintain those relationships with people.
That was a very hefty challenge because you can't ever have that system go offline, so production cutovers and things like that were really important.
We had to deal with a government shutdown in the middle of the migration there, and that was a bit of a challenge. But we saw through all of that and really established our second major cloud service provider that we had at White House. Immediately we started to realize great gains on how those platforms could partner together and coexist and complement one another. Lots and lots of really great software product were shipped and grew out of that partnership between those cloud system.
How do you get there? What did it take to go from A to B?
First of all, you have to have the institutional drive to want to get it done. Every agency is different, every agency is subjective, and you're not going to be successful if you don't have buy-in all the way up and down the stack there.
Leadership has to be behind you. They may not necessarily need to know all of the details of how you're going to execute that technology mission, but they can easily see the value in why you should. That has to permeate down through the IT teams and digital teams so that the folks with the hands on the keyboard really know why this is important because they're the folks who are going to have to do the hard work of slogging through on a daily basis to make sure their processes and stuff catch up. Organizational buy-in is key.
The second one is we are getting much, much better at two different things. How we acquire technology -- there are still great gains to be made on that side, but we've done a lot. We now have a cloud computing policy that has been released as of a few years ago. Then the acquisition has been following along behind that rather slowly, but that's also very important.
We're going to have to confront that as the first half of the problem. In my view, the technology that you're using and the project outcome are destined for failure if the acquisition wasn't structured properly up front. Another hard lesson learned was that just buying it and acquiring it and knowing how to work the contracting side of it has been half of the challenge, and that's before you ever even start to write one line of code or actually get the product together.
I had a really good partnership team at the White House who was keen to try new things on the acquisition side there and we were able to craft it with an eye toward how making this contract look like agile software. The contracting shop did a great job there of crafting that and testing it with us.
What's the project that you think was the most interesting, the biggest shift, the one that stands out for you?
My favorite project to work on at the White House was definitely the CRM project. It began as really tactical and really necessary for the agency to move all of the old systems. Operationally, it was failing. It wasn't passing muster anymore. We weren't able to get our work done.
But then it morphed into new things. By putting it on that cloud platform, a lot more product grew out of that. Some very cool aspects of that project were sort of unexpected in the beginning. We didn't go into them thinking this is part of the value proposition or this is an area about which people are going to be really excited. It came out of collaboration and partnership later.
Things like being able to track anytime a citizen writes to the president and they have an actual issue with an agency. The correspondence team there at White House does a great job of capturing that stuff and assigning it to the agency for completion. We radically modernize that process, so we brought it all into this cloud-based system securely and then wrote new processes with them where the folks on the agency side can just login to the same system, pick that case up, work it to completion, attach case files or whatever might be necessary to follow up and then close it out.
The security on it was tremendously higher than it was before over email or faxing or sometimes just shipping things around and mailing them in envelopes or on USB keys and things like that -- things that you would feel were much older technology. Now it's all online, it's under authentication, it's behind secure sockets and it's protected.
In terms of the service level, for the first time the White House was actually able to see the turnaround time for some of those cases per agency and track them from start to finish, and that just became wildly powerful out of a data-driven analysis perspective. We didn't even know we needed to track that at that point, much less that we can improve it. So those things grew organically out of the system as a better way to do things. Lots and lots of time savings there and lots of improvements.
Those kinds of mechanisms are really revolutionizing the way the agency there does work because it's just so much better than it was in the past with spreadsheets and email and manual process. I'm sure as they keep using it and the teams keep enhancing it, they're going to discover more of that stuff that grows organically once you have the tools in place to do it quickly whenever the opportunity strikes.
Let's move onto the State Department. What is IIP, what do you do here and how does that fit into the overall structure of the State Department?
The Bureau of International Information Programs is my bureau. They grew out of an effort in the '50s under President Eisenhower. They were established as the U.S. Information Agency. They had a pretty narrowly tailored mission of improving the relationship with foreign audiences and educating the foreign public about American values and what we as Americans are doing globally.
In 1999, it was re-orged into the State Department. But the mission remains the same: Engaging with foreign audiences and teaching them about American values and helping them plug in where they're interested to learn about America and what we're doing abroad.
There are really three areas of the bureau on which we're focusing. We are all about empowering the rest of the agency to have the technological platforms to execute on that mission, and then programs and products: the content that actually goes on top of those platforms that gets delivered. [It's] a three-tiered approach to how do we improve the media engagement with foreign audiences worldwide.
Where are people finding it? Are they on their phones? Are they logging on? Is this all digital media?
It's trending more toward digital. Across foreign segments, it's wildly disparate. We're really used to having broadband and cable in the homes here in the American populous, but overseas that's not necessarily the case. What we're seeing is that a lot of folks are only able to get onto 3G cell phone service, but they don't have a desktop or a laptop at home, so it's all mobile. We're trying to offer a cross section across all of those channels.
Probably our biggest properties at this point on the web side are all of the embassy websites. If you go to USEmbassy.State.gov, there's a comprehensive list of every embassy and you can drill down into those home pages per mission.
We are currently undergoing initiatives to move all of those over to centralized cloud systems and brand them uniformly [to give] it a consistent look and feel no matter which site you're visiting. Then the project on which I'm working is how do we handle emails subscriptions and make sure that the messaging campaigns bolted on and around that stuff are as robust as possible as well.
What kind of tools can you bring to bear here at the State Department to help meet this mission?
The legacy infrastructure around the government makes it tough, so you have to be a little more risk tolerant of some of this and I think that IIP is helping lead the department in this area in many ways. Specific tools for collaboration -- things such as Google Docs and Gmail that are accessible, mobile first, and they're available outside the four walls of an agency building are very, very helpful. Allowing people to talk to one another and collaborate around documents wherever they may be is very, very key.
That's part of one of our initiatives, the Digital Productivity Tools Suite. We're also using Slack for the instant messaging and collaboration piece of that. That has been tremendous, a huge benefit to the agency and the bureau, the ability to have folks collaborate in real time in their own channels, right where they live and do their work, and asynchronously as well.
Collaboration was key. That was one of the very first things.
Closer to home for me on my projects was agile software management tools. We are using Salesforce here and we are building atop that platform. There's a really nice free tool that comes with Salesforce called agile accelerator. It's very, very simple. You set up your sprints and your user stores it in that tool and just manages it right there in the platform that you're working on. That has been tremendous.
We opened that up to the project team and other digital folks who want to use it. Eventually I'm hopeful that we'll actually have users and customers who are out there in embassies and post in that system, writing their own user stories and providing feedback right there. Cloud enables you to have a much more robust toolbox of things that you can bring to bear on these challenges that we just didn't have before.
What do you hope to accomplish here at IIP and the State Department during the last year of the administration?
One of the biggest reasons I came over from the White House was all the tremendous work we had done there on the citizen engagement side was focused at an American audience and that was amazing and so much fun. But what enticed me to State Department was, let's really try to do this globally across the foreign audiences as well. Refactoring my efforts on a global scale and pivoting to a much larger project.
I don't think we'd be able to offer as featureful a product that we were able to ship at the Whitehouse, but here IIP, we're going to be able to do the same things on a smaller level for a lot more people.
Success for me would be getting a little more institutionalization around the CRM system and having that take hold on some of the agency and really seeing them feel empowered to use this in a better manner than they were before with the old manual processes. Then really driving the citizen engagement on the foreign audience side as high as we can get it and really promoting American values abroad in a nice centralized, cloud-based manner that is flexible enough that it can grow and live with the agency as it evolves beyond inauguration day next year.
Great goals. How are you going to get there?
It's going to be a lot of work. We are working right now, so all eyes have been focused on the minimum viable product. We have been making sure that what we're offering to these posts is as streamlined and stripped down and basic as what we think it needs to be for them. The second step is going to be to take it to some of them and roll it out and pilot it with them for a while and get their feedback on it.
I as an IT guy can't tell a person whose role is public diplomacy and public engagement how to do their job. I would rather partner with them and get their feedback on the system and then improve it to meet their needs instead of mine.
Then I think the third step is scale it. Enhance and augment with contractors or civil servants where we can and then really scale this thing out to do more trainings and more onboardings. It's not going to be enough for me and a small project team to get on a plane and travel around to a post or three in a month. We can't train enough people fast enough once the product is ready. I'm hoping that we can build the product in a way that is going to make it easy enough for people to access and that really they can start to train each other.
Once we have an embassy on board that's really good at it -- or a person or two at post who's really good -- they can teach each other and help each other and compare notes among themselves. At that point the IT folks aren't driving it as much anymore. We get to take half a step back and work with them on how we improve and enhance the system, but it becomes their own. It becomes the public diplomacy engagement system, really, and that's where I think we'd be wildly successful in how we'd be able to scale to be where I think we should be by next January.
What is the one thing that government has to get right and what happens if it fails?
This may sound way too simplistic, but embrace and employ [cloud] properly. We're coming up on the fifth anniversary of the cloud computing policy the first federal CIO, Vivek Kundra, published. But lots of agencies are still struggling with even getting started. There are some early adopters out there in the space; there have been a lot of people using it. But I think we, as a federal government, are going to have to get much better at that and commit the institutional willpower to get it done.
People are still stuck in the market-research phase in perpetuity -- analysis paralysis as they call it -- and never making a decision to move. Others are just too risk-averse and scared to get started.
But for me, what I look at is what we saw coming into SBA or the White House, those big data centers, those big systems that are managed on premise by government staff. We're just not good at doing that, I think, frankly, and we've got to get ourselves out of that business.
Contrast the amount of time that we spend and the money that we spend on managing our own systems and getting our hands on it versus really big players such as Amazon Web Service or Microsoft or Salesforce. Their company livelihood is based on security of the system and keeping the system online. They have a built-in incentive to keep the stuff running that I just don't think the government has -- right, wrong, or indifferent. So we're going to have to start standing that down and really lean in to the cloud a lot more.
That's going to be a vast change from where we are right now. So those tools coming in place and the agencies being willing to commit to them is going to empower us to do so much more. And we're going to have a lot more money to do it with because we're not spending that money on servers and operations that we're just frankly not good at.
What are the ramifications if the government can't make that jump?
If we don't embrace that, it's going to be more mediocrity like we've seen the past. You're going to have really big projects that have multiple contractors working on them. They weren't built in a modular fashion and they're not running on cloud services. They will fail or citizens will be frustrated when they try to use them. There's really no need for this.
I think that's the overriding point that all of us as modern technologists in the government trying to lead the way here: We need to be building and providing services and websites and apps that delight the user, who is the citizen in this case. If they go to look at their tax return or go online to file a case with the Veterans Administration or they need to go find their visa at an embassy website, all of those things should be very easy and accessible and really clean and easy to get into and maintain and understand.
If we don't do that, you're going to see more of what we have today, which is very large systems that maybe the page is not online or maybe it doesn't work properly or 404 errors a lot, all those sort of things that we have come to tolerate but probably shouldn't from government.
We need to expect a better service level from government-provided technology than what we have today. We can do that absolutely, if we embrace the cloud and employ it properly.
Aaron Boyd is an awarding-winning journalist currently serving as editor of Federal Times — a Washington, D.C. institution covering federal workforce and contracting for more than 50 years — and Fifth Domain — a news and information hub focused on cybersecurity and cyberwar from a civilian, military and international perspective.