Cyclists, The Law & What can be done to improve cyclist safety

sacamano

which is actually untrue both by the numbers and by the philosophy when you think about opportunity.

At least when I'm talking about it I'm basing it on my experience. Are you doing the same?

magicbastarder

sacamano wrote: »

At least when I'm talking about it I'm basing it on my experience.

what does that even mean? in the context of trying to justify your position, that is a weak ass argument to make. 'i'm basing it on my own observations which cannot be quantified or challenged'.

if you stand at any light controlled junction in dublin at rush hour, i can claim with an 80%+ chance of being correct, that at least one motorist will jump the lights on every single light change, opportunity allowing. i know of several junctions where motorists routinely run reds which may have been red for 10s or 20s or more.
sure, i've never seen anyone killed at these junctions, but by god does it create a hostile environment for pedestrians and cyclists.

sacamano

It sounded like he had some stats to back up his comments, that's what I was referring to.

And again, no argument as to which is more likely to cause serious injury as that is blatantly obvious and doesn't need to be even mentioned.

AndrewJRenko

Stark wrote: »

I just wish if they put in light controlled pedestrian crossings that the light actually goes green when the pedestrian presses the switch as opposed to 5 minutes later long after the pedestrian has jaywalked across and everyone has to sit at the red lights at an empty crossing like morons. Or change the law so it's fine to go through if there's no-one crossing (to account for kids messing with the lights and stuff).

I just wish they put in light controlled crossings where pedestrians have the right of way and motorists have to press the button to get permission to go through.

sacamano wrote: »

It sounded like he had some stats to back up his comments, that's what I was referring to.

And again, no argument as to which is more likely to cause serious injury as that is blatantly obvious and doesn't need to be even mentioned.

Stats like the Luas red light camera which found that 88% of red light jumpers were motorists?

sacamano wrote: »

Like to think I do obey every law. And no to breaking the speed limit. Those that do piss me off too but that's not relevant to this thread.

Good to know, because if you were one of the 82% of motorists that break speed limits (RSA Speed Survey), you'd be a thundering hypocrite to be moaning about cyclists breaking red lights. But good to know that definitely doesn't apply to you.

sacamano

Stats like the Luas red light camera which found that 88% of red light jumpers were motorists?

That stat is comparing motorists with cyclists, which is not what this is about. I was putting the cyclists who obey the rules of the road against the cyclists who don't.

Good to know, because if you were one of the 82% of motorists that break speed limits (RSA Speed Survey), you'd be a thundering hypocrite to be moaning about cyclists breaking red lights. But good to know that definitely doesn't apply to you.

Thanks, I'm part of the 18%. Great.

AndrewJRenko

sacamano wrote: »

That stat is comparing motorists with cyclists, which is not what this is about.

It does raise an interesting question though, right? Given that we have such endemic law-breaking by motorists that results in 2 or 3 people being killed on the road each week by motorists, why would anyone really give a toss about what cyclists do? Surely our top priority should be to get motorists to stop killing people?

sacamano wrote: »

Thanks, I'm part of the 18%. Great.

Gotcha. Presumably you're not one of the 88% of red light jumpers that were motorists, right? And you're not one of those drivers who dragged us down to 2nd worst place in the European league table for checking social media while driving, right?

sacamano

AndrewJRenko wrote: »

Surely our top priority should be to get motorists to stop killing people?

Agreed.

Gotcha. Presumably you're not one of the 88% of red light jumpers that were motorists, right? And you're not one of those drivers who dragged us down to 2nd worst place in the European league table for checking social media while driving, right?

Correct.

magicbastarder

slightly apt.

https://www.youtube.com/watch?v=yQO4Dnlwcg0

Stark

Really great ideas for improving safety for cyclists guys. Keep them coming

Unknown Soldier

Bus/Cycle/Taxi lanes should be called same, and be written on the road.

"bus lane" doesn't cut it.

magicbastarder

and they shouldn't allow taxis in them either. that was a retrograde move.

JMcL

smacl wrote: »

Question is more whether on average they'll be safer than human drivers which seems likely. I doubt they'll be doing any close passes, dangerous overtakes or breaking lights.

Unfortunately as evidenced by the Uber "trial" in Arizona, they'll just drive straight through instead.

As far as Tesla goes, neural nets for image classification are easily confused - a quick search throws up this rifle turtle https://www.youtube.com/watch?v=piYnd_wYlT8 Yes, it's advancing rapidly, but because of the nature of it there will always be edge cases, and I don't really want to be one of those. The other problem is of course that as drivers become more reliant on this stuff, they become even more de-skilled than they are at the minute.

smacl

JMcL wrote: »

Unfortunately as evidenced by the Uber "trial" in Arizona, they'll just drive straight through instead.

As far as Tesla goes, neural nets for image classification are easily confused - a quick search throws up this rifle turtle https://www.youtube.com/watch?v=piYnd_wYlT8 Yes, it's advancing rapidly, but because of the nature of it there will always be edge cases, and I don't really want to be one of those. The other problem is of course that as drivers become more reliant on this stuff, they become even more de-skilled than they are at the minute.

I agree, hence my point about LIDAR versus passive image analysis and opinion that Tesla have got this wrong. Even where cameras are used, they should be paired to allow distance resolution through stereo photogrammetry, which offers useful information that image classification will not. Image classification will tell you what something is with certain degree of confidence, LIDAR (and to a lesser extent photogrammetry) will tell you there is an unknown object of a certain size at a certain position and velocity relative to the vehicle. My opinion is that both pieces of information are valuable and Musk is downplaying the value of LIDAR primarily on the basis of cost.