Cycling UK Forum

There has been at least one crash every day this week on either the M27 or A27 and last week there were about three. One crash near Chichester proved fatal.

It seems like crashes are getting a lot more common especially around here. The press report the crashes, but don't seem to be asking their on behalf of their readership any questions about the prevalence of crashes. The general consensus seems to blame it on congestion, road design and volume of traffic. I do not see many reports that mention anything about general driving standards. Only when there is a notable court case does the subject of culpability come up.

One might conclude that crashes and death on the roads are an occupational hazard and can't be avoided. I would disagree with this and say they can, if we get our collective heads out of the sand, wake up and shed the denial. The media tend to leave the subject of bad driving alone and let consumers do the work for them on their comments pages. It's a pity that the reputation for tough journalism in the UK does not seem to have as much focus on people that could potentially kill others with cars as they do with the antics of minor celebrities.

Crashes are inevitable as long as we rely on humans to do the driving. Give it ten years, when presumably cars will have safety features that prevent them driving too close together, and things may well improve.

pwa wrote:Crashes are inevitable as long as we rely on humans to do the driving. Give it ten years, when presumably cars will have safety features that prevent them driving too close together, and things may well improve.

Until the Fancy Bears or ISAL hackers take over and smash them into each other ??????

landsurfer wrote:

pwa wrote:Crashes are inevitable as long as we rely on humans to do the driving. Give it ten years, when presumably cars will have safety features that prevent them driving too close together, and things may well improve.

Until the Fancy Bears or ISAL hackers take over and smash them into each other ??????

We will have to make sure they are offline, I suppose. I'm getting slightly stressed just thinking about the possibility of jumping into the car on an urgent mission only to find that it cannot move yet because it is configuring updates that I never wanted.

TBH, I don't think that driving standards have changed hugely, at least not over the last 20 years or so that I have had reason to observe them in the UK. What have changed are mainly three things: traffic density, the transference of bullying behaviours to the road envionment, and onboard distractions. The much higher traffic density means that small errors are much more likely to lead to accidents and more drivers are frustrated and taking their frustration out on other road users. This kind of bullying has gained acceptance among some road users. Mobile phones, satellite navigation systems, and the use of small computers other handheld devices all provide an abundance of distractions that didn't exist 20 years ago.

All of this is complicated by the fact that the government measure road safety largely by the number of people killed on the roads.

pwa wrote:Crashes are inevitable as long as we rely on humans to do the driving. Give it ten years, when presumably cars will have safety features that prevent them driving too close together, and things may well improve.

Absolute rubbish, I'm a software developer. No code is completely bug-free and no security is hack-proof. Disaster waiting to happen.

Vorpal wrote:All of this is complicated by the fact that the government measure road safety largely by the number of people killed on the roads.

And of course deaths have gone down because of safety features in cars (like airbags) while the number of crashes has inexorably gone up. Along with the spiralling costs (insurance, delays etc).

tykeboy2003 wrote:

pwa wrote:Crashes are inevitable as long as we rely on humans to do the driving. Give it ten years, when presumably cars will have safety features that prevent them driving too close together, and things may well improve.

Absolute rubbish, I'm a software developer. No code is completely bug-free and no security is hack-proof. Disaster waiting to happen.

Same is true of human drivers, except that they cannot all be upgraded to fix the issues found. Of course those are the upgrades that some won't want!

AlaninWales wrote:

tykeboy2003 wrote:Absolute rubbish, I'm a software developer. No code is completely bug-free and no security is hack-proof. Disaster waiting to happen.

Same is true of human drivers, except that they cannot all be upgraded to fix the issues found. Of course those are the upgrades that some won't want!

As a developer myself I can't wait until driving is taken from the poorly equipped, slow to respond, buggy and badly motivated wetware that we currently entrust it to.

Anyway, evidence is everything and so far computer driven cars are massively better than their human counterparts and this can only improve.

Back on topic, it's similar around here. Hardly a day goes by without huge tailbacks on one or more motorways or trunk roads. It's a lot worse than just a few years ago and there's no reason to assume it's going to get better.

When talking about standards of driving, it's important to be clear which skills are being discussed. The typical learner driver now after a dozen lessons will have interacted with the drivers of more vehicles than I did (test passed 1965) in the first couple of years of holding a driving licence. In that sense, they have vastly more experience than I ever had after a given number of years driving. OTOH, that in turn leads to many people showing more impatience, frustration and in some cases aggression. More risks are taken and sometimes things go wrong. Only the most serious crashes in terms of casualties are investigated. ."With so many improvements in vehicle design, the extent of the carnage had grown more slowly than the numbers of crashes.
============================================
PS

By "interacted with" I don't mean raising two fingers or playing Horatio. I just mean something like "had to take into account while driving."

tykeboy2003 wrote:

pwa wrote:Crashes are inevitable as long as we rely on humans to do the driving. Give it ten years, when presumably cars will have safety features that prevent them driving too close together, and things may well improve.

Absolute rubbish, I'm a software developer. No code is completely bug-free and no security is hack-proof. Disaster waiting to happen.

There is hack proof code. It just requires that the code has no external connectivity.

Of course that isn't the real issue. The target needs to be 'better than people' not perfection.

Of course, I'm merely a software developer, so what do I know...

[XAP]Bob wrote:The target needs to be 'better than people' not perfection.

A very difficult target to accurately specify (better than the average person?) - how would you properly verify that it meets the spec - trial and error based on the number of deaths before/after?

[XAP]Bob wrote:There is hack proof code. It just requires that the code has no external connectivity.

That certainly used to be possible, but these days even fridges have wi-fi. Besides which there would have to be some connectivity (even if it via a cable like engine management systems) for upgrades etc. Current vehicles have been successfully hacked while the car was actually being driven.

Vorpal wrote:All of this is complicated by the fact that the government measure road safety largely by the number of people killed on the roads.

Or to put it another way, they measure casualty numbers not usability.

tykeboy2003 wrote:

[XAP]Bob wrote:The target needs to be 'better than people' not perfection.

A very difficult target to accurately specify (better than the average person?) - how would you properly verify that it meets the spec - trial and error based on the number of deaths before/after?

This is currently how it's being done. That's the purpose of trials...