ped/bike improvements

[gfish]

I spent the day biking around town as part of my half-assed attempts to get ready for STP. Without a whole lot to occupy my attention, I thought a lot about pedestrian and bicycle improvements. This is my list of what the city (any city!) needs to do. Nothing extravagant, just basic steps if we're actually serious. It wouldn't be practical to do all this at once, but they should be basic design criteria for all roadwork and upgrades.

• Basic principle: As it is quite legal in Seattle, sidewalks need to be designed with bicycles as well as pedestrians and wheelchairs in mind. You get a lot more riders if you can make sidewalk riding easier -- it's a hell of a lot less scary than most arterials, particularly if you're slogging it out uphill.
  
  
  
  • Utility poles should not be placed in the middle of sidewalks. If bikes and wheelchairs can't get by, you might as well not have the sidewalk at all.
  
  
  • All corners with curbs need pedestrian curb cuts. These should be at the corner, not tucked around the corner. It's important, particularly when riding along a busy street, to stay completely visible when crossing the side streets. People just don't pay enough attention when turing right.


• Marked crosswalks need to line up with the curb cuts. I mean, duh.


• We need more marked crosswalks, and more mid-block crosswalks (with pedestrian bulbs).


• Crosswalk signals and buttons are all fucked up. Currently, they only serve (in the vast majority of cases) to have the walk signal turn on the next time the traffic light cycle gets around to it. The walk signal does not turn on otherwise, and they don't influence the timing of the cycle. This is car-centric bullshit. The walk signal should always turn on for each cycle, rather or not a button has been pushed. The signaling buttons should serve the same purpose as car sensors in the intersection and influence timing. In particularly pedestrian-oriented areas of town, the traffic lights should default to all-red/all-walk until a car arrives to trigger a sensor.


• Any non-residential street that has the room should have marked bike lanes. The burden of proof for bike lanes needs to be on the side of the negative. Bike lanes are cheap and really good marketing.


• Car sensors in intersections seriously need to be able to detect bikes. One should not have to lay a bike down directly over the sensor in order to trigger it. This applies even in industrial areas, dammit.


• Shrubs need to be kept trimmed along sidewalks. Soil creep off of steep slopes need to be kept cleared off of sidewalks. This is probably the responsibility of the property owner, but in that case more fines need to be levied or whatever.


• In a slightly different vein, we need to drop the minimum parking spot requirements for new development. Maximum parking limits like SF has would be even better.

Comments

[cow.livejournal.com]

> All corners with curbs need pedestrian curb cuts. These should be at the corner, not tucked around the corner. It's important, particularly when riding along a busy street, to stay completely visible when crossing the side streets. People just don't pay enough attention when turing right.

> Marked crosswalks need to line up with the curb cuts. I mean, duh.

I actually read an article in the P-I about this a few months back. Apparently, the old federal standard/requirement was the one where the curbs are tucked around the corner, and thus don't match up with the marked crosswalks. At some point, the city realized this was moronic, and new/redone intersections should be the better designed ones, where the curb cuts are right at the intersection and matching the crosswalk. (These changes are good for people on bikes, but were especially done for the disabled.)

[ivy]

I totally agree on the sidewalk riding thing -- it's far less scary than riding along the side of a road where you're constantly having the "don't run me over please" fear. Bike lanes are good, but I would *far* rather ride on a sidewalk than on a road that didn't have a bike lane. I also agree with you about the common sense of lining up curb cuts and crosswalks. Several of my nearest and dearest are in wheelchairs, and it's obviously important for them to not get creamed when crossing the street.

I don't necessarily agree about the button-pushing, though. If the area doesn't have much pedestrian traffic and turning on the walk signal increases the length of the cycle in a way that's not optimal for the traffic that is there, then the requirement to push a button makes sense. For areas where there's constant pedestrian traffic (like downtown), having an automatic walk cycle makes sense. I see it as an engineering optimization problem -- no type of traffic is better than any other sort, but a good solution will optimize for the majority of the traffic while still adequately and safely serving the rest. Pedestrians shouldn't have to wait for nonexistent cars, and cars shouldn't have to wait for nonexistent pedestrians. (And yes, those sensors totally ought to pick up bikes too.)

[etb.livejournal.com]

I totally agree on the sidewalk riding thing -- it's far less scary than riding along the side of a road where you're constantly having the "don't run me over please" fear.

It depends on the road and the sidewalk. If there are driveways or objects to block visibility at intersections, you're almost certainly safer in the street, and certainly less obnoxious to pedestrians. No one, both peds and drivers, expect things on sidewalks to move 15+ mph. The peds won't kill you but the drivers definitely can.
Getting rear-ended is rare. Of course, you could ride a bike on a sidewalk at just above walking speed, but then you might as well just walk.

There are a lot of situation where bike lanes are worse than nothing. If there's on-street parking, you need a huge amount of space to allow safe bicycling, which is almost never available. (The "solution" of installing a bike lane that overlaps the door zone is much worse than having no bike lane at all.) Bike lanes are also hard to mix with bulb-outs: putting them out into the bike lane is horrible, but not having them go into the bike lane doesn't narrow the road much.

no type of traffic is better than any other sort,

No, environmentally superior, non-lethal traffic is better. The usual formulation looks like this (http://www.transalt.org/press/magazine/012Spring/09hierarchy.html). Also note that signal design that punishes walking discourages walking, so one reason for an area not having pedestrian traffic is because the signals are stupid. There are also public health benefits of walking and bicycling.

But I think the easiest solution to issues with signals in low-traffic environments is to just make Don't Walk mean "yield". Or better yet, replace everything with four-way stops.

[etb.livejournal.com]

For pedestrians, going back to universal No Turn On Red would be nice, too.

I'm kind of jaded about engineering improvements. If they're done right they can help, but you're not going to get rid of conflicts between bicycles and other vehicles; you're still left with asshole drivers. I'd be quite happy just following the "Effective Cycling" strategy on existing infrastructure if drivers were in genuine fear of getting $1000+ tickets for honking, passing improperly, or cutting off a bicyclist.

[catsittingstill.livejournal.com]

I'm with you on the sidewalk riding. Having spent several years as a bicycle commuter, both on upright and recumbent bikes, I think the best solution is to have the lanes wide enough to allow lane-sharing, and to ride bikes *with* the traffic, by the same rules of right-of-way. My experience on both sides of the steering wheel has been that drivers watch for vehicles (fast moving, need space to stop or turn) on the road and pedestrians (slow moving, stop and turn on a dime) on the sidewalks. Apparently the accident statistics, (as best I could interpret them and bearing in mind that this was a few years ago and there may be new studies) suggest riding with the traffic on the road is safest also.

That said, I certainly sympathize in a visceral way with the uneasy situation of being passed again and again by cars you can't see coming, moving at high speed. I tried to use low-traffic roads as much as convenient.

[catsittingstill.livejournal.com]

I so sympathize with the car sensor not registering bicycles thing. There was one intersection near one of my old dwellings, where I had to stop at the traffic light, get off my bike, walk it across to the pedestrian button, press the button, and walk back into the lane to get the light to change for me. There was a car-sensor in the road--I could see where the asphalt had been cut to put it in--but it didn't work for me, even when I laid my bike down flat on it; and my recumbent was a fair chunk of metal. I used to fantasize about a bucket of superglue and a sheet of copper. Never got around to actually doing it, though.

[beaq.livejournal.com]

I'm with you on the shrubbery-in-the-sidewalk thing. Another issue: large shrubs planted at corners, blocking visibility through intersections. You can't see pedestrians, bikes, or oncoming traffic, especially from a small car or recumbent.

[etb.livejournal.com]

If it's just the fact that you can't see them coming, you can use a mirror, but I've found it doesn't really help as much as I'd hoped—it's difficult to gauge the vehicle's angle of approach from a tiny mirror, and to decide whether the correct response is (a) do nothing, (b) get a little closer to the curb if practical, (c) crash over the curb to avoid being hit. It does let you look without (worrying about) veering off course as you're turning your body.

Yes, the statistics suggest everyone should ride in the road, though I'm skeptical whether they're handled right: if riding on the road were equally as safe as riding on the sidewalk, you'd expect to see fewer crashes for road riders, because they presumably have more experience. Also, IIRC, there are more vehicle-ped crashes in crosswalks than outside, but we don't conclude from that that no one should use crosswalks...

[catamorphism.livejournal.com]

Any non-residential street that has the room should have marked bike lanes. The burden of proof for bike lanes needs to be on the side of the negative. Bike lanes are cheap and really good marketing.

I don't really get what you're saying here. Bike lanes seem to be less safe than encouraging cyclists to take the lane. What happens at intersections? And as etb pointed out, when they force cyclists into door zones, bike lanes are worse than nothing.

pedal power to the people!

[callibr8]

I either agree with or see the sense both of a lot of the points you made, and several of the counter-points in the comments. Might want to fix the typo about "turing" right though - enough geek-types read you that they could get cornfused. :)

Yeah, us Engineers don't like all of it either, but this is the way it goes . . .

[ilmarinen.livejournal.com]

One can officially claim a bike corridor by providing 2' additional feet adjacent to a parking lane. I know it sounds crazy, but that is the design standard. Vehicle doors must yield to bicycles, legally (in all cases). Probably doesn't make you feel much better after being doored, but at least the vehicle passenger/driver is at fault. Most cars are 6' wide, if they park against the curb, they have 2' of the 8' parking lane for door swing before the door enters the bicycle path. 2' is considered normal required door swing clearance.

The crosswalk-ADA ramp-markings location issue is interesting. Actually, Seattle *used* to put the ramps dead middle of the curb bulb/return, so that it would be one ramp per corner. This is not the preferred option these days (as it has wheelchairs pointed right into the middle of the intersection), so they now provide two ramps, one to each crosswalk. Location of ramps depends on a lot of factors, including clearance for a landing behind the ramp (4' required IIRC). Current design standards/guidelines does place the ramps into the crosswalk.

Location of crosswalks has several conflicting criteria. You don't want the pedestrians too close to the moving traffic lanes--so 2' clear minimum from edge-of-traveled-way to the edge-of-crosswalk. You also want to keep the pedestrians visible to cars making turning movements, that pushes the walks further towards the intersections. Also, (marked) crosswalk locations determine the stop bar location, and there are requirements for that. As an interesting aside, it is assumed that vehicles can and will pull up from the stop bar in order to see, prior to turning. The car just as to stop at the bar first.

But then, one wants the curb ramps to be as close to parallel to the cross walk as possible, which also pushes the crosswalk way back around the curb. On streets with curb bulbs, these geometry issues are more difficult in some ways, and ramps/crosswalks also want to be in line with the sidewalks down either block if possible. Current crossings require audible push-button posts adjacent to each ramp/crosswalk, and minimum spacing between them.

and part 2 as I exceeded allowed comment length (WTF?)

[ilmarinen.livejournal.com]

On our current job we are placing the ramps so that the outer (away from intersection) wing of the curb ramp ends at the end of major radius of the curb bulb. There are other options, but any solution is a compromise.

Somewhere among all this, one has to fit the signal poles, etc.

They are installing curb ramps in all curbed intersections (that allow crossings). One does not go through and install ramps in every legacy curb in the country in one year. Everytime a street receives a certain level of work or upgrades, the ADA improvements have to be installed, as practical.

All sidewalks must maintain the minimum clear width of 36" for ADA purposes. All the junk on the sidewalks annoys me too, but that's where we stuff all our utilities, etc. We look at all clearances when placing the poles, lights, fire hydrants, etc. Typical minimum design width is about 5', with occasional obstructions down to 36".

We don't even strip centerlines or fog lines on small streets, much less bike lanes. That's silly.

The marking of crosswalks is something I disagree with the standards on, but designers are essentially legally bound by them. Traffic controls typically require "warrants." If the intersection/street doesn't "warrant" the control based on the current and predicted traffic volumes, it is very difficult to justify the installation. Crosswalks have additional restrictions--crosswalks are considered a way to channel pedestrians to safer crossing points, so they are not installed on intersections that are bad crossings. Doesn't make sense, does it? Well, that's the way it is. Given sufficient existing/predicted pedestrian traffic, crosswalk signals will be installed. But one has to meet sight distance, etc. So installations are very involved.

I wish that pedestrian crossing that require button pushing at least put a sign up on the pedestrian signal that told one that. I often end-up standing at an intersection wondering why the crosswalk sign didn't trigger. Although I don't like it, the reason for not automatically having pedestrian signals each cycle is that the pedestrian signal can often be the controlling time for the intersection, and it is unacceptable to degrade intersection performance for foot traffic that is sporadic and only is actually there a small fraction of the time.

-B.

[gfish.livejournal.com]

In an ideal world, there wouldn't be bike lanes. But if we're going to have them at all, we should have a lot of them.

[randomdreams.livejournal.com]

John Forester's book Effective Cycling claims that sidewalk-type riding -- 'mixed use' -- is something like 10-50x as dangerous as on-road riding, because of the cars-not-seeing-bikes thing. But the thing is, people *feel* more comfortable riding that way, so maybe getting them to start that way isn't bad.

I've only ever seen one rear-end car-bike accident. I've seen dozens of car-turning-right-in-front-of-bike (and been in three of them) and dozens of oncoming-car-turning-left-in-front-of-bike (and been involved in two.) But again, people *feel* like that's a really serious danger (hence why so many people ride on the wrong side of the street, against traffic) so it's probably something that needs to be addressed before the real danger situations, just to get people riding. Helmet mirrors or handlebar mirrors are cheap.

[randomdreams.livejournal.com]

The light sensor issue is even worse, oddly, when I'm riding a carbon fiber bike -- I would've expected that the rims, being *much* closer, would be the driving metal detection signal, but apparently the steel frame (and presumably spokes) are better signals.

I hate to admit this but I usually wait for about one light cycle and then just treat it as a stop sign and make sure the traffic is clear and go through it, like I would at a broken traffic light.

[gfish.livejournal.com]

I fundamentally dislike forcing bikes to act like cars. If you want more people to bike, it's a bad sell. Bikes make crappy cars -- they're just motorcycles that can't go very fast at that point. Bikes are cool because they're flexible. They work well on streets and on sidewalks and playgrounds and fields and you can park them anywhere including your bathroom. They're cheap and green and they make you sexy. Effective Cycling is cargo-cult magic.

[gfish.livejournal.com]

Oh, eventually I run the light too -- it is broken. But if the street is busy enough, that can be problematic.

[orawnzva.livejournal.com]

Yeah — I can imagine, and probably know a few, cyclists who can hold their own with cars, but I'm not one of them. Does that mean that, when I can't maintain a speed that the SUV behind me is comfortable with, I shouldn't ride?

What we really need is a separate class of rights-of-way for "light vehicles" like bicycles and segways.

[neuro42.livejournal.com]

BBB was telling me Sunday that there's a strong movement afoot to replace induction loops with video cameras and image analysis software. It's not clear whether this is better or worse for bikes. (It's clearly worse for privacy...)