Cycle paths

[mattheus]

But is that post in the middle of the cycle track safe or necessary?

Not safe, although risk looks minimised with white paint and reflective band.

While I hesitate to do so,

Clearly not for long enough.

[PDQ Mobile]

Now you know why I hesitated!

And your answer to the (what I see as) a polite and valid point is?
I guess the edited quote means you don't have one?

[mattheus]

:shock: Now you know why I hesitated!

And your answer to the (what I see as) a polite and valid point is?
I guess the edited quote means you don't have one?

My answer is that a more useful comparison would be with other furniture on rights-of-way e.g. bollards, signposts, arm-co, manhole covers whatever.

Comparing with human beings seems … less valuable. IMO.

[Bmblbzzz]

I don't think I'd call the post unsafe and for certain it's far preferable to a slalom or chicane as usually installed on UK cycle paths.

It's a shame to lose priority. With the square corners reducing car speed and ensuring decent angles, it should be possible to maintain priority there. Even the UK is catching up with this one! Of course, if -- as it looks -- it's a rural road with low traffic and the side roads are not frequent, then it's less of a problem.

[Morzedec]

Suggest that too many people are quietly calling what others post a load of 'bollards'; calm down folks, an be nice to one another - this is cycling forum, not a soccer one.

Happy days,

[PDQ Mobile]

:shock: Now you know why I hesitated!

And your answer to the (what I see as) a polite and valid point is?
I guess the edited quote means you don't have one?

My answer is that a more useful comparison would be with other furniture on rights-of-way e.g. bollards, signposts, arm-co, manhole covers whatever.

Comparing with human beings seems … less valuable. IMO.

Well I fail to see the logic of that at all.

If it makes something more visible then it is surely more useful to all road users?

To me that seems a reasonable conclusion to draw.

[sjs]

Do you sink in water mjr? That's good news - at least you're not a witch

I imagine his bath isn't deep enough to float in. Whether it is or not, he'll displace a volume of water equal to the volume of MJR under it. If he's floating the weight of that water will equal his own weight.

[mattheus]

:shock: Now you know why I hesitated!

And your answer to the (what I see as) a polite and valid point is?
I guess the edited quote means you don't have one?

My answer is that a more useful comparison would be with other furniture on rights-of-way e.g. bollards, signposts, arm-co, manhole covers whatever.

Comparing with human beings seems … less valuable. IMO.

Well I fail to see the logic of that at all.

If it makes something more visible then it is surely more useful to all road users?

To me that seems a reasonable conclusion to draw.

OK.

[mjr]

Do you sink in water mjr? That's good news - at least you're not a witch

I imagine his bath isn't deep enough to float in. Whether it is or not, he'll displace a volume of water equal to the volume of MJR under it. If he's floating the weight of that water will equal his own weight.

Deep enough. But not long enough. I'm rather taller than the average bath.

But returning to the displacement: Archimedes's bath from the fable (submerging an object to measure its volume, to enable calculating its density, to decide whether the crown-maker has conned the king by subbing cheaper metals for the gold) is not exactly the same as the Archimedes principle (which is to do with floating and upthrust).

How easily tales of good design can get muddled and confused and you end up with a bogus claim that all designs are ten times worse and chamging the design makes no difference...

[mjr]

I don't think I'd call the post unsafe and for certain it's far preferable to a slalom or chicane as usually installed on UK cycle paths.

It's a shame to lose priority. With the square corners reducing car speed and ensuring decent angles, it should be possible to maintain priority there. Even the UK is catching up with this one! Of course, if -- as it looks -- it's a rural road with low traffic and the side roads are not frequent, then it's less of a problem.

The post is preferable to alternatives (including cars on it) but still unsafe compared to a flat open track.

Priority is irrelevant, to a point. It still hurts the cyclist more if there's a collision, so most will naturally time their crossing into gaps. Perpendicular approaches, visibility and an ability for most cars emerging to wait ahead of the crossing without blocking it are all more important. The UK still routinely gets this wrong even at priority and light-controlled crossings, encouraging people to stop with their back to likely RLJing drivers whose paths they will cross.

It's probably even less of a problem at quiet crossings, as you note.

[[XAP]Bob]

Archimedes used volume displacement of a denser than water object to measure its volume. I can’t recall if he figured on that basis why some things float.

[Pete Owens]

Ah, the classic old chestnut of an ancient dodgy risk calculation from a different design being applied to everything!

No the compilation of a large body of research over many years and different countries on the relative danger of cycle paths of various designs crossing priority junctions. It is no "dodgy" because it happens not to confirm your bias.

It's dodgy for many reasons, including: 1. it confounds a wide variety of designs

Wrong - while we see the results from many designs - that doesn't mean the data is confounded. Each individual study was looking at an individual set of circumstances - comparing like with like.

- some are good, some are bad - from a wide range of countries - each with different reporting practices - and a wide variety of collision types,

That is a strength - not a weakness. The fact that we see the same pattern repeatedly for different designs means that we can be confident that what we are seeing is a fundamental flaw in the concept of an at grade priority side road crossing (rather obvious when you consider that you are arranging for two parallel streams of traffic to cross each other from out of sight of each other), and can rule out design tweeks in the junction layout.

Actually, since most of these studies were undertaken by traffic engineers ideologically committed to segregation, they tended to see their own results as aberrant and a feature of the particular design they were studying. Rather than recommend this sort of crossing not be used they recommended minor design changes - without any evidence to support those designs. Some would recommend bending the path towards the carriageway - some would recommend bending away from the carriageway - some would recommend merging onto the carriageway and so on.

The one key difference that was observed was between 1-way and 2-way cycle paths. The former increasing the risk of collision by a factor of 3 and the later (such as the example posted by the OP) by a factor of 10.

2. even if you don't accept that, 1993's Pete Owens claimed it was "a factor of 3 to 10" - what's changed?

Nothing - the factor of three is for one-way cycle paths the factor of ten for two-way paths. I am sure I have pointed this out to you before

At least it's not the notorious "11.3 times more dangerous" used by 1995's Pete Owens.

You seem to have an unhealthy obsession with me to be able to pluck an out of context quote of mine from a quarter of a century ago!

A factor of 11.3 will have from one of the individual studies and is entirely consistent with an overall figure of 10. We are taking about an order of magnitude here - not trying to tie down exactly how much more dangerous cycle path junctions are to 15 decimal places.

But we've had this discussion about this exact dodgy claim many times before, always inconclusively and usually with you stopping replying, like at viewtopic.php?p=870054#p870054 and viewtopic.php?f=7&t=119879&p=1206380#p1206380

And I'm sure that within minutes you will have replied to this one. I have better things to do with my time than engage in an endless oh-yes-it-is-oh-no-it-isn't exchange.

[[XAP]Bob]

:shock: Now you know why I hesitated!

And your answer to the (what I see as) a polite and valid point is?
I guess the edited quote means you don't have one?

My answer is that a more useful comparison would be with other furniture on rights-of-way e.g. bollards, signposts, arm-co, manhole covers whatever.

Comparing with human beings seems … less valuable. IMO.

There are vanishingly few other items of street furniture which are substantial enough to stop a vehicle and placed in a location where a vehicle would be expected to be travelling in normal situations.

I'm yet to see a council put a road around a telegraph pole such that it obstructs one lane, or install some 'path' lighting by putting a lamppost in the middle of a carriageway.

Bollards and armco by design won't be on a right of way (they might serve to remark the extend of said ROW.
Manhole covers (unless badly damaged) don't pose a risk to a vehicle in normal usage (although they might be slippery).
Sign and lamp-posts are only ever put in the path of very specific ROW users (the more vulnerable ones).

[mjr]

- some are good, some are bad - from a wide range of countries - each with different reporting practices - and a wide variety of collision types,

That is a strength - not a weakness. The fact that we see the same pattern repeatedly for different designs means that we can be confident that what we are seeing is a fundamental flaw in the concept of an at grade priority side road crossing (rather obvious when you consider that you are arranging for two parallel streams of traffic to cross each other from out of sight of each other), and can rule out design tweeks in the junction layout.

I can see why you would think never discarding old data is a strength. It means that the best designs get lost in the noise of British-style 1980s crap almost forever because bad designs and outdated designs never leave the dataset (even if the junctions themselves are rebuilt!), so as long as you get some big numbers of bad junctions in there early, you're set for life and can quote the same 1990s comparison forever.

However, for anyone interested in improvement instead of unprincipled blanket opposition, parrotting the same never-changing figure for all designs is definitely a weakness.

I'm also unsure how it's not obvious to you that two parallel streams of traffic crossing each other with good visibility at a crossroads is far more desirable than two streams of different-width-and-top-speed traffic conflicting with each other the whole route on the same desire line. Merging two different traffic streams into one internally-conflicting flow may be safer through junctions (even then, left-hooks and right crosses consistently appearing in the top ten collision types suggest it's unsafe), but it doesn't get and keep bums on bikes like more harmonious flows can.

A factor of 11.3 will have from one of the individual studies and is entirely consistent with an overall figure of 10. We are taking about an order of magnitude here - not trying to tie down exactly how much more dangerous cycle path junctions are to 15 decimal places.

Indeed, but it's also worth remembering that we're talking about factors here and even with the abysmal old designs lurking in those studies, the difference is essentially between naff-all and ten times naff-all. Sucks if you're one of the extra ones, but studies suggests the whole-trip risk comparison wouldn't be anything like as high as that overall because what % of your major road journey is going through such junctions anyway?

But we've had this discussion about this exact dodgy claim many times before, always inconclusively and usually with you stopping replying, like at viewtopic.php?p=870054#p870054 and viewtopic.php?f=7&t=119879&p=1206380#p1206380

And I'm sure that within minutes you will have replied to this one. I have better things to do with my time than engage in an endless oh-yes-it-is-oh-no-it-isn't exchange.

The reason why this discussion is endless is that, as hinted at above, we have different aims: I want to find the best road layouts to encourage cycling and that's broader than reducing the risk for the tiny faction like us who are willing to cycle along stubbornly with an HGV at their back wheel; whereas I think you want to find any justification to oppose cycleways forever, so you will never accept that a comparison including 1980s crap may not apply to all current designs and you walk away from any attempt at an updated discussion.

How about an attempt to reconcile your supposed increase in junction risk with the reported decrease in collision rates on streets with cycle tracks or lanes cited in articles like http://ajph.aphapublications.org/doi/pd ... 016.303507 - possibly an interplay with "Safety in Numbers" because cycleways attract riders?

[mjr]

I'm yet to see a council put a road around a telegraph pole such that it obstructs one lane, or install some 'path' lighting by putting a lamppost in the middle of a carriageway.

I can think of several examples of both, but they are far rarer than in cycleways, they're gradually getting kerb islands and other protection built up around them as motorists demolish them and they tend to be middle of a carriageway but between lanes, rather than middle of a lane - whereas I feel that we're seeing more of the blasted things getting moved from separation verges into the outside edge of cycleways just so they're further from potential motorist-demolishers. I agree with you completely that this seems unfair and wrongheaded.

Bollards and armco by design won't be on a right of way (they might serve to remark the extend of said ROW.

I'm not sure that's right. Armco is often set on posts set into the road surface (which OK you then can't drive on, but you used to be able to before the Armco went in - you can still walk on it if you're determined, stepping round/over the posts) and bollards are often used to filter ROWs, to allow small vehicles through but not prohibited large ones, and they have to be on the ROW to do that.