QAL Podcast Todd Decapua featured image

Help Test COVID Safe Paths MIT Project (with Todd DeCapua from Splunk)

Related articles and podcasts:

We’re trying out transcribing our podcasts using a software program. Please forgive any typos as the bot isn’t correct 100% of the time.

Audio Transcription:

In this episode, I’m joined by global technology and business leader Todd DeCapua to share the exciting details of MIT’s COVID Safe Paths app that has been designed to help flatten the curve of the COVID-19 pandemic. Learn how you can get involved with testing the Safe Paths app and help fight the coronavirus outbreak.


In the digital reality, evolution over revolution prevails. The QA approaches and techniques that worked yesterday will fail you tomorrow. So free your mind. The automation cyborg has been sent back in time. TED speaker, Jonathon Wright’s mission is to help you save the future from bad software.

Jonathon Wright:

Hello and welcome to this very special COVID-19 edition of The QA Lead podcast.

This podcast is brought to you by Eggplant. Eggplant helps businesses to test, monitor, and analyze their end-to-end customer experience and continuously improve their business outcomes. And so I’ve got a massive thank you to say to the guys at Eggplant who’ve been helping with the COVID Safe Paths project run by the guys at MIT to help us all get out of the house again, to help reopen the economy and for us to start feeling safe again. 

So I’m going to give you a bit of a brief on this wonderful project because in this episode we’re going to reach out and ask for your help and how you can actually get involved. So there are two ways. The first is by going to which is the not-for-profit website that we’ve got set up to try and sign people up to help test this new application, which launched today in the app store and on Android. And then you can also go to to get involved today. So just go right to the bottom of that webpage and click join the team button. You also have a link there to the GitHub repository. And as mentioned during the podcast, you can also website and or email me directly at So there are three great sources there to get involved, which we’re going to be talking about today on the podcast. So I’m going to hand it over to my good friend Todd to kick off the show.

Todd DeCapua:

Hello and welcome to the show. My name is uh, Todd DeCapua. I’m a guest today, I guess you would say, uh, with Jonathon right here on the show. And it’s funny, he and I were just catching up a little bit, probably way too early on the East Coast of the U.S. and Jonathon based out of London of course, or I guess the greater London area. So, um, but we were having a little fun kind of spitballing on everything going on in the world regarding coronavirus and COVID-19 and we said, you know what, as kind of technology evangelists, technology I dare, I used the word maybe experts, I guess Jonathon. Um, but people that, you know, people can have like, listen, they want to know what’s going on. So we said, you know what, let’s go ahead and record a really quick intro of something that we’ve been speaking about that’s come out of MIT called Safe Paths.

And I know that Jonathon’s been a little bit more involved than I have, but again, we just want to take a couple of minutes, record a little bit about, you know, what is this, make you aware of it and some of the things that are coming up and how you can get involved and how you can help. And some of the thinking around, you know, some of the pieces that we’ve done in our careers, um, that we could now bring to something that would be, you know, for the better of the general community. And again, this is not just in either the greater London area or on the East Coast of the U.S. I mean, this is a global pandemic. So, you know, we’re looking to help share some information, have you be informed, and also enable you to understand how you can go ahead and help a little bit if you wouldn’t mind, would not mind or would want to do that. So, Jonathon, I’ll hand it over to you for a little bit more on exactly the topic, and then we’ll get into the little jive back and forth as we were starting to do already.

Jonathon Wright:

So it’s really interesting, you know, uh, I’ve literally, and this is hot off the press, you know, uh, I was approached by, or the test lead or the QA lead for an MIT project, uh, which they’re trying to launch this weekend. So they’re all working on it. And I’ve got behind me, I’ve got the Slack channel at the moment, which anyone could join, which is covidsafepaths.slack. Um, so you could join that channel now and start talking, you know, they’re looking for any help. They’re looking for UX designers, testers, developers, project managers, absolutely anything. And this is to help with the COVID-19, uh, challenge. So, um, Safe Paths. Let me just give you a little bit of background about what that is. It’s a, it’s a mobile app, uh, and it’s kind of look, looks at the kind of three kinds of main scenarios, right?

The first one is, you know, you’ve been diagnosed with COVID and you know, they’re asking you, you know, have you been out anywhere? Do you think you might have, you know, potentially, um, infected anybody without you trying to have to remember where you got it? We’ve already got this data. You know, we’ve got like, I’ve got four different Android phones behind me, but you know, each one of these Android devices will use Google location services, which runs in the background anyway, by default, and tells you where you’ve been. And you can see on Google maps if you go back and look at your history. So that kind of straight away is a freebie for kind of data to say where you’ve been. And then what the second scenario is, is that if you’re, you know, at home and you know, you’ve had to go out shopping and you’ve crossed paths with one of these people, you know, there’s a chance you might have got infected.

So part of it is, this is the idea with Safe Paths is it actually then uses that data to say, well, there was somebody who was infected, um, and you’ve spent time with them because your GPS location coincides so you’re potentially at risk. So again, it’s helping prevention because the kind of, you know, part of it is you want to make sure your, you know, you’re not going out, you’re not putting yourself at risk, but also if you are, unfortunately, do catch the virus, it also helps you to tell other people to isolate. So it’s a really useful idea and it’s a great concept. We’re trying to get the mobile app out by the end of the weekend onto TestFlight. It’s already an APK file. If you’re an Android user, you could download that from the, uh, from the GitHub repository, which we’re going to share. But part of it is this is a call to action, which is kind of what Todd was just saying about getting involved. Right? But, you know, they’re looking for testers to come in and help us because your data can help, um, you know, make this real. So, you know, Todd, you know, your background, doing this kind of stuff, you know, this is so important. The data.

Todd DeCapua:

Absolutely. And, you know, I know that, um, between the work that, you know, some of the bits that you’ve done and some of the smart cities space as well as, you know, the work that I’ve had the opportunity to do in many large enterprise organizations around the world. Um, whether it be financial services, insurance, pharmaceutical, medical, you know, every one of these industries you kinda run into some of the same challenges. And especially as we’re talking about data gathering, we’re data to get the gathering this data off of a mobile device of some sort. Um, I know you and I were chatting, you know, whether you’re on a train or you’re in a bus or you know, on the underground, um, or the Metro or something like that, you know, again, how is it that you’re able to capture this information? How is it that then with that data, uh, we’d be able to interrogate it and say, Oh yes, it looks like you were in a vehicle.

You know, that could have been a car or it could’ve been a bus. You know, can we overlay, you know, bus schedules? Could we overlay train schedules to understand and you know, maybe, more importantly, you know, and looking specifically at New York City and the pandemic there and you know, retrospectively like why is it that it just kind of exploded in New York? Well, a lot of it was so many people on public transport, you know, putting their hands on the rails as they’re riding. Um, you know, the subway there. Again, this was one of those main reasons. So again, how is it that being able to capture this kind of data, um, and other geographic data along with network, you know, how much of this network that we need to think about, what type of connections are people on? Again, if we drop for how long do we drop, can we still capture it on that end client and then be able to send that data in?

Putting in all of these elements become important as we think about collecting data. There are a couple of other pieces that I’m sure we’ll get into, but I guess first and foremost, I think of collecting the data and all of that kind of diverse scenarios, but really bringing that value back to, you know, what is Safe Paths at MIT and that project really looking for, it’s gathering all of that type of GPS and Bluetooth data for the citizens, for the public as well as various health officials. So again, how is it we can do that within this app trains the underground geographic location as, as the network and potentially doing some pretty significant overlays with that dataset.

And yeah, I think part of, um, you know, visualizations are absolute key finish, right? You know, tools. Uh, you know, I know a lot of the APM tools have these kinds of visualization of data and you know, we’ve talked about this in other shows, but it is so important to be able to see that data because like you just said, you know, the network drops because you’re on the Metro, right? And then you pop out at another tube station. So the point is they need to correlate that as a journey. The journey is you’re on the underground. Right? Um, so part of it is, is quite a complex problem. And you know, the guys at MIT are working on a kind of building algos to identify, yeah, he was going down the M 25.

Jonathon Wright:

It wasn’t a coat or a bus, you know, there was no other public transport there. You know you were in isolation, you’re a key worker, you’re going to work, right? So, you know, there’s that kind of scenario. But I think from a testers perspective, which is kind of where we’re doing this call for crowd testers is, you know, it’s your data, right? And you’ve got your data has got all these historical actions and your first thing, I guess a lot of people who are listening or say, well, what about my privacy right now? There’s a couple of things. The first one is, um, you know, GPX is, is quite an, it’s like an extendable JSON format, which is an international standard, right? So you could load that up into a PARS app, which the MIT guys have got. And what you could do is it will visualize, but like Google maps, right?

It’ll visualize where you’ve been. And then what you could do is you can use geofencing, so you could literally highlight your home or you know,

Todd DeCapua:

Your work, or—

Jonathon Wright:

Wherever is sensitive data, you don’t want to share.

Todd DeCapua:

Some pub you might frequent.

Jonathon Wright:

A pub I’m frequently when I shouldn’t be doing right. So you get rid of that data, you just highlight it and delete it. But the other stuff you’re happy to share, right? You’re going, you know, I’m commuting to load there and I’m getting on the Jubilee line. I’m getting off at Canary Wharf. You know, I’m then going on the DLR over somewhere else. Each one of those journey bonds is incredibly important because we need to be able to create realistic data sets, right? And so we’ve, we’ve done this problem before. So I think as you said, Todd, you know, I did, uh, the Copenhagen smart city project and that was really interesting because the challenge was exactly like this challenge is today is, you know, they said to us, look, when it launches, we’re expecting about 40,000, 50,000 people to be wanting to use this app, which is obviously a lot of data, right?

But the second thing, which was a bit of a challenge was obviously, you know, people download the app, sometimes they enable location sensor, uh, you know, thing pops up and says you share, you want to share your location. Yes, you press that, then you’re starting to track, but you will, how frequently do you track? Because the more sensors you’re sending, the more battery you’re going to drain, right? And nobody likes an app which is going to literally run their phone flat completely. Now you know this about 650 data points that Google takes in the background, right? But that kind of would put you quite sporadically around the really what you need to do to complete the picture is kind of balance of that. It is kind of saying, well actually I want to share my information. I want to share, you know, the frequency of my information.

And what we did is we actually got a GPX logo, which you can download off iTunes or Apple. Um, and literally said to them—we paid them in beer, right? The university students—and said, go off, record it, share it, and you’ll get a voucher. So they went off and you know, shared their data, gave us all this reference data, which was, a Delta. It was like a base for us to say, here’s the city of Copenhagen, here’s people on bikes, here’s people on the, on the tube, here are people on the tram. You know, we’ve got those different modes. We’ve identified those modes. Now we understand the different types of transportation we can then use synthetically generate historical transactions. So we then took that data and created 5 billion historical transactions. What had happened over a six month period. Cause we needed reference data to be able to train the algorithms to make sure that, you know, the false positives aren’t going to go there.

So what happens if you go slightly past somebody and they’re in Starbucks and you’re somewhere else, you know, you don’t want them to come up with a red alert. You need to have a certain amount of tolerance, right? So we need that reference data. We need to solve this problem. So, you know, sharing your data is going to be a big help, right? But also there are other people out there who still can download the app and use every day. There are key workers, in the UK, there’s NHS staff. I’m sure you’ve got a lot of people who are still having to be first-line and they’re out and about. They could register for this app and download it and enable it and start sharing their data.

Todd DeCapua:

Yeah. Believe it or not, my, my wife Shannon, who, you know, well, I mean, you know, she’s, she’s one of those essential workers. So yeah, absolutely required to go into work. And I know this would be something that she, amongst the other essential personnel would absolutely leverage and benefit from. But I guess Jonathon, I’m just summarizing here, like some of the things I heard you just say like, we are, you know, bringing in an APM solution to be able to go ahead and capture this type of data that we need. Um, clearly, you know, historical things like logging are going to be important in how is it we can bring that in. You talked about the, you know, Jason, um, a file that we can bring in. That’s, that’s easy. Um, there are other security elements, but we talked about how do we want to anonymize this data.

So again, not to have concerns about that piece. Um, but all this data, again, that we really want to be able to visualize this in a world-class way and make that visualization available again to all the appropriate parties that we’ve spoken about so that, you know, whether it be for citizens or public health officials, you know, they’re able to reference this data and really start making some informed decisions. Um, we talked into how to like some of the bits that you and I have done before and captured and replay, um, especially around the mobile site and what would be required there. And we’re going to be reaching out to some partners to see if we can get them to participate in this type of activity as well. Um, and kind of the next stage. Um, this is all been moving very quickly this weekend. Um, again, I know you mentioned the Slack channel.

Um, I would like to mention, um, your website, which you just put up, which is Um, again, out there you’ve made all of this widely available, all the key links for the are out there. But I mean, really this call to action is, you know if you’re into things like testing if you want to make a difference in the world for COVID-19, or you know, coronavirus, whatever you want to call it, you know, this is an opportunity where you can sign up and participate in that testing activity. We’re also looking at performance. We’re looking at security, we’re looking at functional, we’re looking at all the different devices that are out there in the world. Um, and again, we’re not gonna pay you in beer, but you know, this is something that is a world pandemic. And I think for people like Jonathon and me, we just want to know, you know, what is it that we can do to help make a little bit of a difference and help to enable people like you, citizens in the world to be informed about what’s going on and maybe where you may or may not be at risk.

Um, and then also enabling our public health officials to make these well-informed decisions. So again, that was just kind of a quick summary of all the things that I know we’ve been talking about. Jonathon, I don’t know if there’s anything more that I might’ve missed, but if you want to wrap this up, I think that’d be perfect, then.

Jonathon Wright:

Yeah. So you know, thanks for the intro there. So literally this morning I put up a So Foxtrot, ox Oscar, um, you know, and part of that is a domain anyone could go to, it’s a call of action. You know, it’s the same from MIT it’s the same information. Uh, it’s about people volunteering, you know, uh, as a global community, you’ve got engineers, you’ve got data scientists, you’ve got professors at Harvard and MIT who are also involved in this project. Um, and all you need to do is if you go to that website, there’s a register, there’s a literally a help—how do I help? And, or you could just email me, which is that’ll come to me and I’ll pass you onto the right kind of people if you need it. If you’re interested in doing that, I can send you the links.

I’ll make sure that you know, we can try and get as many people as possible to help with this crowd testing. You know, if you’re not going out, that’s fine. You know, we’re going to get mobile labs that we’re going to set up into the cloud. You’ve got to access them. We’ll give you the payload and historical data to upload onto it. So you can literally just test it. You can use it, you can give us suggestions, feedback, bug fixing, absolutely anything, right? But you’ll be helping save lives because like Todd was just saying, you know, part of it is the information that’s used there we can use to help prevent, you know, the infection growing. And I think that’s really important that you can make a real difference. And you know, at the same time you’ll be able to start, you know, you’re well to use some of the technologies and get yourself up to speed with, uh, you know, some of the great work that the guys at MIT will be doing. Really interesting project, great community, you know, come and help us.

Todd DeCapua:

Outstanding. Jonathon, well, I don’t think we could have covered it any better, but this is a start. I’m looking forward to our next recording where we’ll be able to show you some of the data that is coming in. And you know, once we get past the MVP, which I was shocked to learn that this has been a huge push through this weekend’s and the MVP is going to be going live this week, um, as soon as tomorrow, is what I understand. So, uh, again, Jonathon, just again, kudos to you on the team. Uh, appreciate the opportunity getting pulled in and being able to help make a difference with you.

Jonathon Wright:

Wonderful. Thanks Todd.

Easily build reliable tests the evolve with your application's UI.