What's causing the Tesla fires?

With Tesla owners having reported 3 fires in 2 months, all the talk has been about the hazardous nature of driving around on a ton of electricity. However, I'm starting to think there might be more to these stories.

Tesla has been quick to point out that all 3 fires were a result of an accident. That's fair. Accidents happen and it's important to separate accidents from scaremongering talk of spontaneous combustion.

All three accidents involved the cars being driven over a hard object at speed, which in turn pierced the battery casing and started a fire. I'm curious to understand why Tesla owners are driving into things. Is it purely bad luck, or is it something more?

I have a theory.  It's completely speculative, but it is possible, nonetheless.  I wonder if these accidents may not have been entirely down to bad luck.  I wonder if they could have been worsened - perhaps even caused - by driver distraction. 

Could driver distraction be playing a role?

Never before has a driver been able to do so much behind the wheel of a car - over and above the primary task of driving! Never before have so many familiar, tactile controls been replaced with a cold smooth interface. Operating a Tesla Model S is unlike operating any other car.  And as brilliant as the technology is, it could be a lot more user-friendly.

Many of the car's functions - both simple and advanced - now require a visual check and a particular dexterity that was not needed in the chunky world of dials and buttons. In short, the interior of the Tesla Model S - as phenomenal as it may be - is potentially unsafe.

Lastly, we shouldn't underestimate the temptation presented by all that considerable computing power poised an arm's reach away. I'm tempted to use my iPhone all the time: when stopped at traffic, whilst waiting for a meeting, even during a meeting. It's not surprising that similar technology - super-sized and transplanted into a car - will sit there begging to be used, providing significant temptation, and inevitably distraction.

So are we perhaps missing something crucial with these fires? Rather than jumping to conclusions about the technology that propels these pioneering vehicles, perhaps we need to consider another possibility:

That indirectly, Teslas are catching fire because of a dangerous user experience.

Putting the touch into touchscreens (Part 1)

It is the big trend in in-car interfaces: replacing buttons and dials with functions in a touch screen.  But can the smooth, clean surface of a touchscreenoffer anything like the reassuring feedback of a button or a dial?

The Volvo Coupe Concept interior

Tesla started it all with the massive 17" screen in the Model S and it seems every manufacturer wants to follow in their footsteps. 
Volvo recently displayed their Coupe Concept with a beautiful interior dominated by a large central interactive display.  Even the new Peugeot 308 does away with many of the buttons you'd expect to find in a centre console, replacing them with functions housed within a touchscreen.

Whilst the interiors are looking minimalist and sleek as a result, hiding functions behind a screen means drivers now need to look and think about functions that were largely instinctive when there was a button or a dial to fumble for and tweak.  And if drivers are looking at and thinking about a screen, they're not looking at and thinking about the road in front of them.

So how do we make touchscreens safe to use whilst driving?  The answer lies in their name - the sense of touch - or haptic feedback to be more scientific.  Essentially, the answer is to segment functions within the car: functions that are accessed frequently and infrequently.

The interior of the Peugeot 308

Frequently actioned functions that should not require cognitive effort (like changing track or turnign up the heater) should be supported with an input device that offers tactile feedback.

The most obvious example of this is in a task like changing gear in a manual car.  Gear changes happen frequently in a journey, so the cognitive load around changing a gear should be kept as light as possible.  The thought of replacing a gear lever or flappy paddles with a touch screen that the driver needs to tap every time they wish to change up or down is absurd.

Infrequently actioned functions can generally afford a slightly higher cognitive load.  Opening and closing a roof, or even a window for that matter happens seldom on any journey and often when the vehicle is stationary, so using a touch screen to initiate these functions sound almost surprisingly sensible.

Of course this split isn't that simple - just becasue a function isn't needed frequently, doesn't mean it shouldn't be immediately and instinctively accessible.  A hooter is a good example of this, and there are many more similar functions to consider when identifying the most appropriate interface to action them.

So can we make a flat, glossy touchscreen work for functions that just need haptic feedback?  Continued in Part 2...