Intersesting term "hierarchical topology." I can't help but wonder if bicycles are dismissed in the hierarchy, with the end result being the urban freeway like road structures that are not all that pleasent to bike upon (or walk upon for that matter). Often the result of hierarchical topology is a walled canyon like mulilaned high speed arterial that is just short of being a freeway.

There is a hierarchy of performance; for the sake of argument let's assume it is a combined measure of design capacity for traffic speed and volume.

Then there is topology, the shape of the network. If the shape of the network follows the performance hierarchy, then any trip of any significant distance requires travel from the local street to the collector, and then to the arterials, minor then major. If the network topology is instead well-connected and independent of the hierarchy, then longer trips can be made on local roads and collectors.

I think it's valuable to cyclists to promote both (1) better cycling conditions on the biggest roads in the performance hierarchy, and (2) better lateral connectivity across the hierarchy.

I think traffic engineers have ignored cyclists in terms of how both of these issues affect our travel. However, I think some of them are coming around, understanding the reality of having to use arterials for useful medium and long-distance utility cycling in many places, and understanding the desirability of having alternate, lower-hierarchy streets available for useful trips. I believe this understanding has supported new design standards to include wider pavement designs (WOLs, bike lanes) on more important roads, better planning of collector roads, and better connectivity of local streets.

There is a hierarchy of performance; for the sake of argument let's assume it is a combined measure of design capacity for traffic speed and volume.

Then there is topology, the shape of the network. If the shape of the network follows the performance hierarchy, then any trip of any significant distance requires travel from the local street to the collector, and then to the arterials, minor then major. If the network topology is instead well-connected and independent of the hierarchy, then longer trips can be made on local roads and collectors.

I think it's valuable to cyclists to promote both (1) better cycling conditions on the biggest roads in the performance hierarchy, and (2) better lateral connectivity across the hierarchy.

I think traffic engineers have ignored cyclists in terms of how both of these issues affect our travel. However, I think some of them are coming around, understanding the reality of having to use arterials for useful medium and long-distance utility cycling in many places, and understanding the desirability of having alternate, lower-hierarchy streets available for useful trips. I believe this understanding has supported new design standards to include wider pavement designs (WOLs, bike lanes) on more important roads, better planning of collector roads, and better connectivity of local streets.

Why not, in such a hierarchical environment, design in (these are usually newer neighborhoods) as a mandate, a parallel wide bike path that provides similar access to the same areas served by the high speed high volume arterial; a low volume set of streets for instance that will not readily serve the capacity for thru traffic speed and volume.

Such a design could be discontinuous in such a way as to discourage heavy thru motor traffic, yet encourage skinny fast** vehicles. Bike boulevards so to speak. ** I mention fast here on purpose... anything that is going to serve commuting cyclists should be designed to allow 25-30MPH traffic. This could readily serve the cycling community while discouraging motorists looking for shortcuts.

One way to do this is just a set of parallel streets that dead end with a thru connecting narrow path. Often this can be done quite easily by using the same "bulb out" neighborhood streets and a connecting bike path between "bulbs." This type of configuration also serves the dual purpose of allowing emergency access for fire departments by allowing alternate access (beyond the main arterial) for fire equipment, where barriers that restrict access to narrow vehicles only, are removable by fire crew. This ultimately becomes a win-win for the community. Quiet fast bike paths for commuting cyclists, fast freeway like arterials to serve the driving community, and alternate emergency access for the community.

I should have been more clear. My apologies.

I was not referencing the notion contained in the statement above; but the reference to an outside source with an extra statement suggesting that it would be difficult to understand ("best wishes ..."). But I might have taken that the wrong way too.

It is true that failing to reject the null hypothesis that the parameter is zero is not equivalent to "no conclusion about the cause/effect relationship between variables can be drawn from the data collected". For instance, one could still create a confidence interval to determine the likely range of that parameter. Using the standard 5% rejection rate, it would include zero in the discussed example; but one certainly learned something (made a statistical conclusion) about the parameter.

More generally, I would say that different disciplines have different standards with regards to statistics. If a sociologist, statistician, economist, biometrician, engineer and so on are in the same room talking about a particular subject, one would find a fairly wide range of practices and norms. I understand your point regarding sample size and validity of the study. Personally, I would say that the study contains information and is valid. As an aside, the word valid is yet another loaded word. But in the aforementioned room, one would find people using identical words with different implied meanings since they are based on different acceptable standards within the disciplines.

In this sense, I feel the general argument--at some point--was over semantics.

Well my stats courses are 10 years behind me and in French, so I remember the fundamental workings of, but not all the proper terminology to debate a point, that's the reason I referred to a website, as background material for my words. But I think the point has now been made clear as at first there was a stat, and then it was disputed, and now it appears to be back on course.

So... we can get back to discussing the conclusions of those Danish results again. To use an unfortunate Chertoffism, I have a gut feeling we'll be seeing a lot more of these studies invalidating the perceived safety of BLs

Then there is topology, the shape of the network.

This is a good read on the subject:

A key principle keeps emerging that breaks yet another orthodoxy of the engineering world. For years, engineers designed road networks to flow a bit like blood vessels, with smaller roads feeding into thicker ones. (In fact, the Green Book term for a major thoroughfare is “arterial.”) Toth has come to believe that engineers are putting too much pressure on these major roads. He’s convinced that what would work better is something more like a grid — where drivers have lots of options for getting around, rather than one overburdened highway. “We’ve created a situation where everyone who wants to go out for a bagel or newspaper has to go out to the same road,” Toth says.

http://www.redbridgeroad.org/urbanspeedway.html

This is a good read on the subject:



http://www.redbridgeroad.org/urbanspeedway.html

So something more along the lines of a world wide web... would be a better road design... where there are always alternative routes to get to some point.

I like it.

The construction of cycle lanes has resulted in an increase in accidents of 5% and 15% more injuries. These increases are not statistically significant. The decline in road safety can be seen both at junctions and on sections. The increases occurred especially amongst cyclists and moped riders where increases in injuries is tendential amounting to 49%.

Tallard. You obviously know more about statistics than I do (I took 2.5 full years of calculus but only a single term in statistics during my education; only so much time to learn things in 8 years, gotta pick and choose). The above bolded section implies that the increases in accidents are not statistically significant. I assume this to say that the correlation between accident increase and the construction of bike lanes are not statistically significant.

Tell me what this sentence means, since I am obviously beneath you on this subject. If I understand you, the phrase "not statistically significant" means either that there is no connection between accidents and bike lane construction, or it means that whatever connection there is cannot be discerned from random chance. Which do you interpret this to mean, and how do you make this distinction?

Why not, in such a hierarchical environment, design in (these are usually newer neighborhoods) as a mandate, a parallel wide bike path that provides similar access to the same areas served by the high speed high volume arterial; a low volume set of streets for instance that will not readily serve the capacity for thru traffic speed and volume.

Such a design could be discontinuous in such a way as to discourage heavy thru motor traffic, yet encourage skinny fast** vehicles. Bike boulevards so to speak. ** I mention fast here on purpose... anything that is going to serve commuting cyclists should be designed to allow 25-30MPH traffic. This could readily serve the cycling community while discouraging motorists looking for shortcuts.

One way to do this is just a set of parallel streets that dead end with a thru connecting narrow path. Often this can be done quite easily by using the same "bulb out" neighborhood streets and a connecting bike path between "bulbs." This type of configuration also serves the dual purpose of allowing emergency access for fire departments by allowing alternate access (beyond the main arterial) for fire equipment, where barriers that restrict access to narrow vehicles only, are removable by fire crew. This ultimately becomes a win-win for the community. Quiet fast bike paths for commuting cyclists, fast freeway like arterials to serve the driving community, and alternate emergency access for the community.

Well-connected collector roads, with or without speed humps, can work well as bike boulevards assuming they have lower priority than the arterials they cross; motorists choose the arterials instead for convenience. Using actual streets rather than paths provides some guarantee that the path will be open in darkness (Cary's greenways are closed outside of daylight) and that the junctions will be designed properly, and may become signalized if the delays waiting for a break in arterial traffic become unacceptable. ( Since the arterials are controlled by the state, you have to convince the _state_ engineers that a signal is warranted, but due to low user volume bike paths rarely meet MUTCD warrants for signals, so the signals will never come.) About the best we can manage for getting greenways across arterials mid-block is an offset marked crosswalk with a median refuge area.

The only place in Cary to put a path that would not be closed in darkness is in the sidewalk location beside arterials, but this would suffer increased junction conflicts compared to cycling on the arterial roadway.

There are some off-road greenway corridors, such as the American Tobacco Trail rail trail, where resident opposition has been low enough that bike commuters have been able to get the city of Durham to leave a section of the trail open in darkness. This trail provides a very long route with few at-grade crossings. However, reports are that the downtown paved portion gets scary at night, and the suburban/rural portion is unpaved, due to superior political influence of the running and equestrian interests there.

I do like to use some of the paved streamside paths in Cary for travel to parks with my family, or to singletrack MTB trails a few miles away. During early hours they are empty enough to cycle at my preferred speeds; in the evening they are congested with dog walkers, kids, runners, skaters, and many I-pod wearers. I support such trails as recreational facilities, but there are few places they provide utility that can come near that of a well-connected collector street system.

Well my stats courses are 10 years behind me and in French, so I remember the fundamental workings of, but not all the proper terminology to debate a point, that's the reason I referred to a website, as background material for my words. But I think the point has now been made clear as at first there was a stat, and then it was disputed, and now it appears to be back on course.

So... we can get back to discussing the conclusions of those Danish results again. To use an unfortunate Chertoffism, I have a gut feeling we'll be seeing a lot more of these studies invalidating the perceived safety of BLs

More generally, I hope that quality data is collected to make strong statements about safe cycling and the impact of various programs on cycling.

More generally, I hope that quality data is collected to make strong statements about safe cycling and the impact of various programs on cycling.

We can always hope, but to date, the strongest "statements" about the impact of various programs on safe cycling (80% accident reduction!) have been founded on sophomoric analysis (at best!) of the flimsiest of data (when the data exists at all.)

There's a whackjob 'bicycle driving' instructor out here in Seattle (david smith) who is actually PAYING PEOPLE TO TAKE HIS CLASSES that are 'afraid of traffic'- does anyone think that 'study' :roflmao: he's doing will produce unbiased results, paying participants to take a 'bicycle driving' class? :eek:

There's a whackjob 'bicycle driving' instructor out here in Seattle (david smith) who is actually PAYING PEOPLE TO TAKE HIS CLASSES that are 'afraid of traffic'- does anyone think that 'study' :roflmao: he's doing will produce unbiased results, paying participants to take a 'bicycle driving' class? :eek:

Hmmm, it might be more representative of the population than people who pay to take or volunteer for such classes. Regardless of whether people volunteer, pay, or are paid to take the class, they all incorporate a bias.

-G