Published: 8 June 2017
IoT, Privacy, Smart Cities and Bikes
Gerry Gaffney
This is Gerry Gaffney with the User Experience podcast. Today’s guest is back in her native Australia as part of a UK trade delegation. She has a varied background, including in human resources, change management and organisation development at places such as Accenture, BT and Queensland Rail.
Most recently, she co-founded See.Sensein Newtownards, which is about 15 km from Belfast City on Strangford Lough. The company is a cycling technology and data company which is how it first caught my eye. Irene McAleese, welcome to the User Experience podcast.Irene McAleese
Thank you very much. Delighted to join you.
Gerry
The so-called… We’re sitting here under a picture of the Queen, which seems oddly appropriate… The so-called “Internet of Things” is a much-hyped development. Samsung has a family hub refrigerator and hipsters can have their single origin coffee begin to brew as their Uber approaches their loft. Do we really need to connect our bicycle lights into this Internet of Things?
Irene
Yeah, that’s an interesting example to use there and I think at the start of the Internet of Things there was a perception, is this all about a kind of gimmick. Is there anything useful that could come out of this? And I think there’s a huge opportunity for connecting up bicycles actually. I’ve been doing a lot of work with the European Cycling Federation and one of the things that they quite rightly pointed out is that, you know, cities at the moment are pretty much designed around cars and the roads are here for cars. That wasn’t always the way but as our cities evolved and become more and more congested, there’s a lot more work happening around the concept of a smart city; cities being able to use data to work efficiently, use their resources well and make data-driven decisions.
So if cycling is not part of that discussion simply because the data doesn’t exist, that’s a huge opportunity that we’re going to miss out. So cities of the future that are designed to be smart cities, if we’re not putting the data from cycling in as part of that equation, we could be, you know basically reinforcing the design of cities around cars.Gerry
Although I guess one argument, I mean I take your point about smart cities and it’s certainly one of the buzz words at the moment, isn’t it? A lot of the data that’s coming into certainly our Google maps and our Apple maps is coming from people’s already existing devices so our handhelds, our mobile phones are giving us all the traffic information that we need. What other sort of data do we need?
Irene
So the data that we collect is quite different to what you can get from a mobile phone for a few reasons, mostly around accuracy. So the sensors in our devices are actually dedicated sensors. They collect data; we’re able to send data back 800 times a second as opposed to an app which is sending data back about 1 to 2 times a second. So that’s hugely significant when you look at the kind of things that we’re monitoring, such as the road surface that the cyclist travels over. So we can look at that road surface and see whether it’s rough or smooth and over time how that road surface is changing. And that’s simply not something you can pick up via an app, for example
Other types of data that we’re looking at are accidents and near-miss events. So we can detect where a crash has happened but also quite interestingly where a near-miss event has happened. So that’s where the cyclist didn’t quite come off the bike but definitely wobbled and had some kind of reaction. So the data that we generate allows us to create a kind of heat map for the city, showing not just where the cyclist cycled and what speed they were at but actually layer that with really rich sensor data that can be very useful for city and town planners.Gerry
You’re currently running trials in Milton Keynes and Dublin, is that right?
Irene
Yes.
Gerry
Can you tell me a bit about each of those and what sort of data you are collecting and what you’re doing with it? What the city is doing with it?
Irene
So in both those cities we’re using the ICON light, it currently is, the ICON light is currently sold all around the world to more than 50 countries, but it’s only in these closed trials where we’re collecting the data. So we don’t collect data currently on the retail product but the purpose of these trials is to kind of help us understand the value of the data that we’re collecting, both from the point of view of the city but also from the cyclist’s point of view, because our ultimate sort of vision is we would like to be able to share back to the cyclist and to the city, this crowd sourced, really useful data that we’re collecting. But we want to sort of use these trials to scope that out and better understand that.
So what is the value of the data to the city? Our understanding the road surface, for example; how would a city use that accident and near-miss information? We can also collect temperature levels and light levels in the city as well; can they use that information? So in both of those cities they’ll be, the trials are actually just about to kick off. They’re handing out hundreds of lights to cyclists. Cyclists will basically just travel around, riding as they would normally do and the data is fed, goes from the light via Bluetooth back to the app and from the app into the Cloud where it’s anonymised and aggregated and where insights can be sort of extracted.Gerry
So that means the light has to be on at all times? I mean for you to gather useful data, the light has to be active for that to happen?
Irene
Yes, we have the light on, when you have the light on it’s collecting data. So there’s a big shift actually, a trend towards using bike lights during the day. There’s a lot of research, I mean most, well the research shows clearly now that 80% of accidents on bikes actually happen during daylight hours and it’s the commuter periods, 6 to 9am and 3 to 6pm where you see these big peaks. And there is research coming out that shows that having daylight visible lights can help to reduce accidents even in daylight hours. So, for instance, in 2011 in Europe all cars are required to have daylight visible running lights and that’s based on research which shows a reduction in accidents. So you kind of figure, well you know, if something as big as a car needs it, then we would advocate as well for cyclists to use it during the day.
Gerry
So the data that you gather, do you, like when you pass on to Milton Keynes or Dublin or whatever city you’re working with, will that be just on a big spreadsheet or are you doing a value-add to it prior to that to try and interpret it?
Irene
Absolutely yeah. It’s about, I suppose the technology is really an enabler and it’s very much trying to position this in terms of being, helping a city solve their problems. So it’s not just sort of handing over raw data but it’s working with the city to understand what are the particular problems you are facing in your city and how can we help use our data to solve that? So in fact in in each of those cities we’re working with, we’re particularly tailoring the data and the insights that we gather to answer particular needs.
So, for example, in Milton Keynes, they’re actually about to do a large investment in their infrastructure and they want to build some sort of cycle superhighways, a bit similar to what London has. But as their first stage, they want to understand where are the cyclists actually travelling in Milton Keynes and what are the conditions of our existing cycle pathways. They want to sort of map that and know what state they’re in. So that’s actually really important for them so then they can say okay this is where we should build our cycle superhighways based on where cyclists actually want to travel, rather than where we think they might want it. So that’s something useful for them so we’ll prioritise our analysis and insights around that in the first stage for them. Dublin has… we’re working with them actually at the moment to scope out their needs but the early sense that I have is they are very focused on increasing cycling participation and using our app and our tool and our technology as a tool to engage people to have a conversation with the city and to feel part of the momentum that the city’s creating as they try to promote cycling.Gerry
I guess it’s interesting because a lot of you know there are apps out there that invite you, and websites that invite you to report incidents or report places where you’ve had an issue or something but they’re all effortful from the cyclist’s point of view. So I guess the idea with this is once you sign up the effort is offloaded onto somebody else.
But what about the privacy concerns? I mean I know you said the data’s anonymised before it’s put on the Cloud but are you, are people that you are talking to, cyclists that you’re talking to, concerned that about their privacy?Irene
So, at the moment, as I mentioned, we’re doing the closed data trials, although we have done some surveys of our existing customers to ask them about how they feel about the data and get a sense of what their privacy concerns are. So I asked a really open question like, “How do you feel about us using your data for you know sharing your data with the city?”
Over 90% actually came back and said “yes, have it”.Gerry
Wow.
Irene
And it was really, it was great to hear that level of trust I think in our company. We are a company that set out from the early stages to be by cyclists and for cyclists and I think that being a smaller company with those values obviously helps generate that trust.
So the 90% said yes straight out, 5% said “Yeah, I’m kind of interested. I just want to know a little bit more information,” which is fine because it was a very open question. And the other 5% said “no, I just don’t want to”. And that’s fine and I think that our model would be an opt, you know we will clearly and be very transparent about you know what we’re doing and allow people the ability to opt-in, opt-out as required.Gerry
I’m always a little bit cynical about that, not from the point of view of your product and your company, but you know we’re always talking about yeah it’s always optional but time and time again we see people opt-in, opt-in, opt-in, opt-in and have no real idea of the amount of data that they’re sharing on their loyalty cards and their transactions and their transit usage and everything in general.
Irene
Yeah. Yeah, people sign up to Facebook and all sorts of things and there’s long, long terms and conditions that nobody reads and they’ve kind of signed into. I think my, I think the trend seems to be you know it’s a balance between privacy and benefits gained and, I mean, the cyclists that I’m speaking to anyway are really motivated to see better cycle infrastructure, you know the roads improved, you know, safety to be put into the design and I guess that that has to be balanced with the sharing of data.
Yeah, we can anonymise the data and aggregate it so we try and build in privacy.Gerry
In regard to your… you were talking about collecting sensor data about essentially about cycling conditions but what about things like pollutants and other things that one might measure in the city? Are you looking at any of that sort of stuff?
Irene
Yeah, we are actually. That’s definitely on our road map. One of the… we have a trial, or a collaboration starting actually with Belfast City Council. That’s really exciting because we’re actually going to integrate our tech into their city bike scheme.
Gerry
Oh cool.
Irene
And the city bikes are a nice piece of equipment to work with because they’re kind of bigger and there’s more space for us to do stuff so...
Gerry
And they’ve got electricity built into them as well.
Irene
Yeah through the dynamo and all of that. So just because the cavity is larger it actually allows us to you know it gives us the opportunity to do some more stuff with the sensors.
So yeah, air pollution monitoring is something we’re looking at. It won’t go into the first phase of the bikes but we are looking at maybe a second phase to do that. And there’s some really interesting work going on in the UK at the moment around looking at the types of sensors that you have. It’s not easy. They have to be kind of miniaturised and you have to be measuring the right things and to measure them with the accuracy that’s needed to actually be meaningful. But there’s a huge advantage of being able to measure air pollution on bikes. The fact that they’re mobile, one, you know you don’t have a static monitoring system and bikes go places that you don’t currently get information. They’re at kind of a nice height as well. Some of the particulates can, if your air monitoring is up too high, you’re not capturing some of the particulates which are that lower level where the bikes are travelling around. Of course bikes don’t produce any pollution so it’s a little different to sticking it on a bus or a car. So yeah, there are a number of different advantages to be able to do it through a mobile sensor network platform on a city bike scheme.Gerry
Yeah, I guess one thinks of China as being the obvious market for such a measuring system, you’ve got somewhere like Hangzhou which has got a massive number of bike share stations.
Irene
We’ll have to get to that next.
Gerry
You designed, the company designed, the company See.Sense designed its own software and hardware. Why did you take that particular approach?
Irene
Because we can. (Laughter.) I mean, Philip, my husband, you know he originally trained as an electronics and software engineer. It’s his passion, it’s what he loves and we’re good at it. And I think that’s actually a real strength for See.Sense that we don’t outsource our tech. We can design it in-house. We have that expertise. It makes us really nimble, it makes us really innovative and yeah that is definitely something that we will continue to do.
Gerry
I must admit one thing that disappointed me a bit, originally I know that you had, you kind of had an open source approach and you were going to be able to program the firmware on the on-board chip using you know ATMEL in-circuit programmers, or whatever they call them, but then you removed that capability; what was the reasoning behind that?
Irene
So that was really early on. That was our first product we launched on Kickstarter and it was one of the reward options in the tiers. To be honest, the only reason we removed it was lack of demand. We had literally, I think it was five, I think five people out of everybody who had backed it on Kickstarter selected that option.
Gerry
Is that right?
Irene
Yes, and then of the ones that did purchase it, they never came back for any instructions or engage with us at all.
Gerry
SDK information or anything like that…
Irene
No, so it suggested either they didn’t probably do it or they were so expert at it they didn’t need to talk with us but yes clearly there was no demand. We’ve never had any interest or enquiries about it at all. You’re the only one.
Gerry
I opened up one of my ICONs because I wanted to connect my ATMEL program to it and lo and behold there was no way there to do it.
Irene
Ah.
Gerry
So there you go.
Irene
On the ICON version?
Gerry
Yeah.
Irene
Yeah, just no interest.
Gerry
Yeah, that’s fair though.
Now since this is the User Experience podcast, I have another question I just have to ask; the original version, or the first one that I saw, in order to turn it on you had to you know spin around three times and you know rotate it in a certain direction, run through these amazing hoops to turn it on and off and I remember not only I had trouble but my kids having trouble with it and sort of berating me for having such a device in the house. It was an engineer-led solution, and you talk about being, and you know for cyclists and doing research with cyclists: How on earth did that get past anyone and into production?Irene
[Laughter.] Yes, well good point and I think that, the first product we’ve learnt a lot since then. We did test the first product with cyclists and that was to be fair mostly Philip coming at it from definitely from an engineering point of view. The light was tested with cyclists but he would actually go out and mount it on the bike for them and he was thinking more about the intelligence of the light and the algorithms and how it would react to the environment and all of that good stuff and thought, yeah gestures are cool, we like intelligence so I’ll make it turn on with gestures but of course that’s not that user-friendly and so we pretty much quickly realised that once we got it out, the first Kickstarter.
Gerry
Now you told me before we started talking that coincidentally Ian Fenn, who was on the podcast, a couple of weeks ago talking about UX portfolios, is an ICON See. Sense user. Did he complain to you about that by any chance?
Irene
I can’t remember if he mentioned about that.
Gerry
I would hope he did. [Laughter.]
Irene
But there was, yeah, I mean part of the great thing actually about Kickstarter is that people love to give you feedback and we have welcomed that. In fact we, from the first product, the 1.0, we eventually launched a 2.0 version which was based on a working group set up of some really early adopters from that first Kickstarter. There were probably about 20 people who were like super feedback providers that we… and we thought you know what? They’re doing this, they’re giving us the feedback because they love what we’re doing, they want to see this succeed and they want to see it work. And so we kind of brought all of these super feedback people together into a group and we really worked with them really closely and the insights that we got from them, I don’t think you could have paid for if you’d gone to some sort of marketing consultancy or something. We had a great group of people. Everyone from you know local club riders, commuters, through to people whose day jobs were working in space technology and things like that, a really amazing group. So, their insights were really useful. We released a 2.0 version pretty quickly.
Gerry
With a proper “On/Off” button on it?
Irene
Well ICON is now the one with the On/Off button, yeah.
So we moved to that in under a year. So it’s great, I mean, we basically took all of the great stuff about the first version of the product and then worked much closer with real cyclists throughout all of that so I think the version we have with ICON is the result of hundreds of cyclists in putting in feedback through those sessions, so I think we’re pretty happy with where we’ve landed now.Gerry
Great. The IoT holds great promise but also the opportunity to make things disjointed and difficult to understand and just plain frustrating. Do you have any thoughts on how companies like See Sense can help avoid that disjoint or that disconnect?
Irene
Absolutely. I think from the customer… I mean, our model is really two sided so one is the customer, and we need to make sure first and foremost that our products work as products. And I think that is really important and we can’t lose sight of that. So basically, our model has been, you know first and foremost it’s a great bike light. Forget about the, you can almost forget about the IoT stuff. It works on its own. In fact it’s sold around the world at the moment as a bike light. Yes, it has a connection to the app and there’s some really compelling features through the app such as the crash detection alert which would notify your nominated contact if you’ve had a crash with a GPS location, theft detection alerts to notify you if your bike moves and you’re not on it, and people are finding that quite useful if they’re just popping into a cafe or something and they haven’t taken a big heavy lock with them. It’s just a bit of peace of mind. It’s a really practical thing. It’s like being able to check your battery level, see the percentage of that and get a notification if your battery gets quite low so you’re not caught out while you’re cycling.
So there’s some interesting things in there that come with the app, but we haven’t lost sight of the fact that, yeah, first and foremost it’s a great product. The bike light works as a bike light and it’s not a gimmick from that point of view. In fact just a few weeks ago, I’m really proud that the See Sense ICON was voted by road.cc, which is the largest online cycle website in the UK, by their readers as the best bike gadget. And that goes to show the sense of, you know, excitement about the product and loyalty so then I think it’s balancing that with you know what we deliver to people in terms of the benefits but also then the other side, to the city and how we can use the crowdsourced data from that to help you know enable cities to do some really interesting things for cyclists as well.Gerry
Okay. Irene McAleese, thanks for joining me today on the User Experience podcast.
Irene
Thank you, my pleasure.