O’Reilly Solid 2015

‘Building connected products that help disabled people’

Given at the O’Reilly Solid conference on ‘Hardware, Software and the Internet of Things’ in San Franscisco on June 25th 2015


Talk overview

What's wrong with the Internet of Things

There is starting to be a lot of interest in the Internet of Things and people have finally stopped talking about fridges with questionable use cases. However we are still a long way from an Internet of Useful Things. Despite a huge potential for the technology to solve real problems for actual people it feels like most of the products that come onto the market don’t actually do that.


The problem is that most of those problems aren’t solved by products that work on Kickstarter or platforms that you could sell to Google.

solid_prez-006If we look at the first category. Works on Kickstarter. Checking on the demographics of the typical Kickstarter user:


Now I’m not saying that these people don’t experience problems, just that there are other groups who are more likely to experience more severe ones. This may be why we end up with stuff like this (I drew this cartoon last year but it still seems relevant):


Which brings us to the most peculiar strain of connected products – the bizarrely-shaped-WiFi-connected-glowing-plastic-boxes:


I was wondering why so many of these things kept appearing. Then I realised that they were the overlap in the Venn diagram.



Something that might work on Kictstarter that you could turn into a ‘platform’ and maybe sell to Google.



On the ‘platform’ front we see people piling in with offerings that seek to lock in thing makers for the long term.


The key issue with this strategy is that building the platform is not the difficult bit of getting a connected product to market.


The challenges in engineering, manufacturing, distributing and maintaing a physical product across the world are orders of magnitude greater than deploying and maintaining a time series database in the cloud. Any organisation with the that already has the capability to do this would be insane to end up in a situation where they were shipping products that were locked into a particular IoT platform.


So if the good use cases for IoT are not in Kickstareter hits and the players in the space are more likely to be thing manufacturers than platforms, how should IoT products be created? The answer is in the title of my talk at South by Southwest this year:


Check out that talk if you want the detail on why this is so important in general. Today I’m here to tell you why I think so many of the good use cases for the IoT are in helping disabled people one way or the other.


We are going to start with the Social Model of Disability. This is a very powerful idea that has strongly informed my work, since I found out about it. It asserts that people aren’t disabled by their impairments but by their environments and the attitudes of society at large. A way of stating it in a way that helps when thinking about design is that they are disabled by a poor fit between their capabilities and their environment. If designers can intervene to improve this fit then they are no longer disabled (or at least become less disabled). We shouldn’t be seeking to ‘fix’ the person – there isn’t anything ‘wrong’ with them, instead we should be fixing what is wrong with their environment. The IoT presents some really fantastic opportunities to do this.




Returning to more practical matters there are many straightforward reasons IoT is a good fit for disabled people:




Digital technology has already had a profound impact on the lives of many disabled people. This video is from some interviews I did for Scope where I people with a cross section of different impairments ‘How do you feel about technology’.

So we’ve got the why. Now onto the how. What strategies can we employ to create connected products that solve problems for disabled people.


I wanted to start with an analogy that occurred to me when I was chatting to an Uber driver yesterday who works mainly as a plumber.


I realised that good design of IoT products was a lot like good plumbing. It is about making sure you make the right connections. To do that you have to think about them and make them carefully. The right connections being in the right places is much, much more important than the total amount of water the system can carry.


This is not the approach most people are currently taking in the IoT. There is relatively little good design research going on working out the fine grain of the connections made possible by IoT products. Instead there is a belief that more data must be better, with relatively little concern where it has come from. This is a bit like optimising your plumbing system for pipe capacity rather than connections, which would be disastrous.

Anyway I promised you a process and here it is. I realise that this stuff is second nature to designers but in some engineering and technology contexts this is still not obvious so it is always worth emphasising it.


I used to work at the RCA Helen Hamlyn Centre for Design and Professor Jeremy Myerson who runs it is very keen on describing the design research process as a ‘double diamond’. Basically there is a research phase that has a both divergent and convergent phase and ends at the design brief (where a traditional design process might start). After the brief another pair of divergent and convergent phases get you to the actual design. The benefit of running a process like this is that you are much more likely to get the the right brief. Design is a hugely competitive profession and those that make it into jobs are mainly incredibly capable – there are few bad designers working. Yet many bad products make it to market every year. They are usually bad not because the designer failed to meet the brief. They are bad because the designer succeed in meeting a bad brief.



Before the case studies I wanted to introduce some of the tools I use at different phases of this process:




[Here in the talk I presented a case study of a project with Stannah but unfortunately I am not able to release it here yet.]

The next case study is Responsive Street Furniture that I’ve been developing with Marshalls.




I’ve been researching how disabled people move around in public space since 2009. I’ve done research projects for central government, local government and disability charities.



This work has used a common methodology, shadowing people on journeys they would regularly make unaccompanied. Observing how they move around and getting them to give a running commentary on how they are feeling, what is helping and what isn’t helping. I started shadowing only people with sight loss.



Then I got the opportunity to participate in an access audit for the City of York, where we shadowed people with different impairments navigating the same matrix of streets in the core of the city.





I made these drawings of all of the people we shadowed as part of the project. As we were putting together our recommendations to the council I realised that different disabled people often had conflicting needs when it came to the streetscape. For example a major issue in York was insufficient seating along popular routes. This disadvantaged a huge group of people who can walk, just not very far without a rest. Adding seating along routes would help these people but it would disadvantage people using wheeled mobility aids because it would reduce footway width. The actual street design ends up being a trade off between the needs of different disabled people.



Another trade off in street design was revealed to me by a video I found on Youtube. It combines three things I’m interested in independently but never thought would ever be combined together.



As this is America this may need some background. This is the Mayor of London, Boris Johnson


He was elected in 2008 on a platform of ‘Keep London Moving’ promising to make traffic flow better. To do this many pedestrian crosswalks, like this one in Kilburn, were re-phased to give pedestrians less time to cross.


This meant that local older people did not have enough time to get across, so some of them made this video.

When I saw this I thought to myself, this is 2013, how are we still giving everyone the same amount of time to cross, irrespective of how much time they need. Surely the Internet of Things can help here. I’d been working on this research project looking at digital accessibility and I realised that in digital a totally different paradigm was in place.


solid_prez-055Whereas in the streets we are trying to create a ‘one size fits all’ design that is the best compromise between different users needs, in digital different people are included by making devices, apps and websites adaptable. In this way different people experience different versions of the product that meet their particular needs. In many cases these digital products reconfigure themselves automatically in response to the user.



I realised that the Internet of Things can allow this principle to extend to the physical environment and that I could make a useful impact on the street. I pitched the idea to Marshalls, a major manufacturer of landscaping products and street furniture and they were up for collaborating on it.



Responsive Street Furniture is the result of this collaboration

Here’s an overview of the system. There are two ways of accessing it, using a key fob and soon using a smartphone.




If you want to know a bit more about where all of this fits into the Smart City discourse check out the Manifesto for the Clever City I published at South by South West this year.



Otherwise all it leaves me to say is get your design research hats on and…


That’s how you’ll build a useful IoT product. Thanks very much for listening!