Cracking the Code to Saving Lives: An Interview with Tony Surma


Tony Surma, CTO of Microsoft Disaster Response, explains how developers, testers, and others are contributing to disaster relief mobile apps that will save lives on a global scale. Learn how you, too, can help by using your powers for this fantastic organization.

Noel: Hello, this is Noel Wurst with TechWell, and I’m sitting here with Tony Surma who is the CTO of Microsoft Disaster Response. He works a lot with Humanitarian Toolbox, and we wanted to introduce Humanitarian Toolbox, talk about what it is, and what they’re hoping to accomplish. Tony, thanks for sitting down with me today.

Tony: Yep, thank you for the opportunity.

Noel: So can you talk about what Humanitarian Toolbox is, and the current state that you’re in and what you’re working towards?

Tony: Yes. Humanitarian Toolbox is an initiative across a number of technology folks as well as non-profit organizations. The goal is to create effectively a virtual software organization so that we can take the needs and the challenges that response organizations see.

So, think anything from American Red Cross to a bunch of non-profits which you’ve probably never heard of who show up at a disaster, understand their challenges and needs, turn them into real requirements, and then have that guide a process across building, maintaining, and making the solutions ready to deploy.

We focus on crowdsourcing and leveraging the generosity of folks who donate their time and technical expertise to build this solution—typically devs, UX, etc., as well as those who can sure we have high quality, so testers, and then those who know the industry, so those are the partners that we have, and make sure that what we built is actually what will be valuable and what will be used at a time of disaster.

Noel: Cool. I really like the science behind requirements gathering. I talked to developers sometimes who talk about helping that customer through these requirements, finding out what they want, and sometimes even just what they think they want, than even being able to offer like, “Well, here’s some other things we can do.” Do you ever find the thing where you meet with customers or potential clients who maybe don’t even know the capabilities that these kinds of tools can have?

Tony: Yeah, absolutely. It’s a combination of when you’re in the middle of a disaster what you need to do is very direct and immediate. Just like personally if you’re in a car accident, you don’t want to be fumbling with the UI or using a tool. So they think about the simplest way to get something done, and that’s actually pretty good because that helps us get requirements that are very clear.

The other thing we’ll do is, we also find that you, you, and you, multiple people might do things differently in a disaster and you can’t build a solution that works for everyone all different ways, so we find that common theme, find the common denominators so we can get the solution out there and get it used by the most people to have the biggest impact.

Noel: Something else I noticed on the website, which is that Humanitarian Toolbox came about from a collaboration with NetHope, CrisisCommons and GeeksWithoutBounds, and I’m sure than just a strengthen number. It’s a recipe. But as a whole, what are those groups able to accomplish that maybe individually would’ve been much more difficult?

Tony: Part of it is NetHope is a consortium of I want to say 39 response organizations worldwide. GeeksWithoutBounds focuses on how do you create something sustainable that actually gets built and deployed. CrisisCommons, much like their name, looks at what happens during a disaster and finds the common ways that people are solving problems. So each of those are a piece, or a part of the puzzle. Then adding in Humanitarian Toolbox, the support from Microsoft, the support from .NetRocks, the other folks that are found there, take that technical generosity, find all of those problems that they represent as connectors to other organizations, and then bring together that commonality. Because what we don’t want to do is design solutions no one needs or we don’t want to guess what it is.

Those are the experts. NetHope represents folks who get first responders on the ground. GeeksWithoutBounds has been doing this for a long time along with CrisisCommons, take that experience and knowledge, distill that down, couple it with the software and technical expertise and then get real solutions that can deliver real results.

Noel: That’s awesome. That sounds like a benefit for everybody. It’s not just something that Humanitarian Toolbox can do, it advances their causes as well.

Tony: Right. Ultimately, the idea is response organizations really like any company or organization or family or human being, never has enough time or enough resources to do everything they want to do. So if we can pull together those needs, we can build open source readily deployable solutions, then that can be something that amplifies their ability to deliver impact. Because they spend less time and less money building solutions, they can reuse what we’re able to build and provide for them, and then they can focus on literally saving lives, the types of things that they do during disasters.

Noel: So what are some of the things that a single tester or a single developer can do either at a conference like STARWEST or even someone who just comes across the website? Again is that What’s something that one tester, that one developer can really contribute?

Tony: The process we go through is we take those needs, we document them, we do that on both GitHub and Team Foundation Services. Simply, one can take what’s currently in a repository if you’re a dev, look at one of the issues of the requirements that isn’t addressed, make a simple pool request, check in whatever process you want to use, add that bit of a feature and then contribute.

And what we’ll do, is when people do that, we’re tracking who they are and the fact that they’re contributing. Because our biggest thing is we want to see when that later has impact, it could be three weeks later, it could be years later, to be able to go back and let people know that they contribution they made actually delivered impact in a disaster. So if a hurricane comes, you contributed to the crisis check in solution which is what the testing is happening here on, then you’d be able to know and get that reward, if you will, of knowing that your work contributed to something. Same thing on the testing side of things.

Obviously as we all know developers deliver a lot of code, build a lot of stuff. We have to have folks who can really look at the quality and take that step back and understand is this easy and simple to use? Is this the right functionality? Because when you’re in the middle of a disaster and you’re doing some work, you don’t want a bug, you don’t want something not to work, and frankly, developers are going to be focused on building it, not necessarily focusing on that end quality. And that’s where we really see a value that a tester can go in, download it, or we usually have most of them alive on the website, and hit that live website, test through it, find a bug, log an issue in TFS or GitHub and then another dev can maybe come along and fix it.

Noel: Awesome. That goes right into the last thing I want to talk about, which was I where saw on the website there was a call for developers and testers, but also for UX masters. And I like writing about UX, learning about how the UX changes overtime, and was curious as to what kind of UX goes into a tool that is more than likely being used by someone under great duress, whether they’re missing a loved one, something happened to them directly and they’re in need of shelter, or help, or a first responders? How do you add an engaging UX into a tool that is … I’m sure it goes more than just it needs to work, and not cause that person more stress.

Tony: Yeah, I think the way we tend to talk about this is think about any time-critical situation or life threatening situation you’ve been in, you are at that point allergic to complexity. You want the simplest, most straightforward, easiest to use. Now, that’s also true. No one wants to have a complicated process to buy something online. So it applies in all scenarios, but this is really truly a case where as you said I tend to use the word literally in these cases to be clear, literally the earth may be shaking, you may have lost a loved one, you may not know where your loved ones are, or you’re a response organization and you know every moment you can spent helping somebody is that much better you can do for the world.

So when you think about UX we think about what are the simplest most straightforward ways for this to be accomplished? What is the most intuitive, and intuitive means following the patterns that are native to whatever platform they’re using. It also means thinking about, not so much what’s the UX of a screen, but what’s the UX of a how I’m using this as a whole? If I’m someone who is going to a disaster site and doing damage assessments of buildings, what is my mindset, what is my thought process? And it’s less about engaging them like you would with a brand, and it’s more about how do you engage and the way they’re thinking and reflect that on the screen. You want it to be set with options and different approaches because different people work differently, but you don’t want it to be so complex that it’s hard to use. So there’s an important balance there that we’ve seen UX folks can really help in disaster situations.

Noel: Awesome. That’s very cool. Who wants to learn more about Humanitarian Toolbox can go to their website which is I know that on Twitter it’s @htbox. Any other places they can go for more information or to being able to start helping immediately?

Tony: That’s great, go to those two, off to the website there’s link to the bottom to our GitHub repository as well what we have in TFS. The website is going to be evolved in a bit, it’s actually one of our open source projects, so you can help us do that. And we’re going to add information about the projects who are project leads and things that are going on. So we’ll get more and more information over time.

About the author

Upcoming Events

Jun 02
Sep 22
Oct 13
Apr 27